The 2028 POTUS ballot is D+15 in our new Strength In Numbers/Verasight poll. This is notable given all the talk about turnout fx in 2026 that will (people speculate) fade by 28. A lot of the 2024 Trump voters who didn't show up preferred D to R candidates
www.gelliottmorris.com/p/2026-11-19-strength-in-numbers-v…
G. Elliott Morris
George Elliott Morris (born June 7, 1996) is an American data journalist who is best known for his work on election polling and predictive analytics. From 2018 to 2023, Morris was a data journalist for The Economist. From 2023 until 2025, he was the editorial director of data analytics at ABC News, overseeing FiveThirtyEight. More on Wikipedia
-
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
I have written a short paper with Peter Enns and Benjamin Leff of @verasight_data writing up a true out-of-sample test of "synthetic" "polling" applied to 14 key Senate races ahead of the 2026 U.S. midterms.
We find that synthetic samples did worse than the polls at forecasting candidate vote margins both at the topline and subgroup levels, with particularly large deviations in vote intention for young people and 2024 non-voters.
If you want to understand *public* opinion — that is, attitudes about the human public — you need human data.
www.verasight.io/reports/synthetic-polls-vs-real-surveys-… -
G Elliott Morris
@gelliottmorris
Notably, not a single person volunteered Rahm Emmanuel's name as their first choice in our initial 2028 primary poll. As a second choice, he gets 0.3%.
G Elliott Morris
@gelliottmorris
New Strength In Numbers/ @verasight.io poll of the 2028 POTUS primaries: No clear Dem leader, Rubio leads Vance by 6
Voters were asked who they supported in an open-text & single-choice question, revealing support for Harris is mostly name recognition
www.gelliottmorris.com/p/2026-11-19-strength-in-numbers-v… -
G Elliott Morris
@gelliottmorris
New Strength In Numbers/ @verasight.io poll of the 2028 POTUS primaries: No clear Dem leader, Rubio leads Vance by 6
Voters were asked who they supported in an open-text & single-choice question, revealing support for Harris is mostly name recognition
www.gelliottmorris.com/p/2026-11-19-strength-in-numbers-v… -
G Elliott Morris
@gelliottmorris
New poll: Trump approval at low of 31%; Americans split on impeachment, favor bills to reduce costs and corruption and roll back POTUS's mass deportations program.
Partial Strength In Numbers/ @verasight_data release here. More on 2028 coming next wk
www.gelliottmorris.com/p/2026-11-18-november-strength-in-… -
G Elliott Morris
@gelliottmorris
Time for the “moderate vs progressive” debate to die — or so say the moderates! Those of us arguing the L-R axis is a poor way to understand voters have won this fight. Just let Gallego, Shapiro, & Elissa Slotkin tell you. (Quote from this Atlantic piece www.theatlantic.com/ideas/2026/11/democrats-intraparty-di… )
G Elliott Morris
@gelliottmorris
This is like, straight up vintage 2025 Strength In Numbers posting www.gelliottmorris.com/p/the-strategists-fallacy
-
G Elliott Morris
@gelliottmorris
The 120th Congress will have the highest proportion of first-time members in at least three decades. This is how “wave babies” took over Congress, particularly the House
open.substack.com/pub/fiftyplusone/p/how-wave-babies-took… -
G Elliott Morris
@gelliottmorris
According to @CharlieCookDC , the Democrats didn't win a "wave" election this year because their national party favorability rating is negative. But that didn't stop him from calling the 2010 GOP win a wave.
-
G Elliott Morris
@gelliottmorris
The conventional wisdom is that midterms don't predict the next POTUS elec: specifically, that swing in the pop vote from POTUS year Y to midterm Y+2 does not predict the swing from Y to year Y+4.
Has ne1 looked at this if you "control for" turnout? Does vote-switching persist? -
G Elliott Morris
@gelliottmorris
A lot of pundits were quick to write immediate takes after Nov 3 about how this was A Prices Election, relying heavily on one poll question about the country's "most important problem."
But if you look at voters' revealed behavior/preferences, a very different story emerges
G Elliott Morris
@gelliottmorris
Among 2024 Trump voters, disapproval of the president's handling of immigration/deportations predicted voting for Democrats in the midterms more than views on prices did, and their views on his corruption scandals predicted 2026 vote about as well.
www.gelliottmorris.com/p/2026-11-17-republicans-midterms-… -
G Elliott Morris
@gelliottmorris
Voters told pollsters their most important problem in 2026 was prices. But MIP doesn’t tell us what people voted.
If you ask 2024 Trump voters who didn’t vote GOP why they defected, ICE/deportations + corruption explain more lost votes than affordability www.gelliottmorris.com/p/2026-11-17-republicans-midterms-… -
G Elliott Morris
@gelliottmorris
Among 2024 Trump voters, disapproval of the president's handling of immigration/deportations predicted voting for Democrats in the midterms more than views on prices did, and their views on his corruption scandals predicted 2026 vote about as well.
www.gelliottmorris.com/p/2026-11-17-republicans-midterms-… -
G Elliott Morris
@gelliottmorris
The next piece in my series on why Dems won in 2026 — and why so many people underestimated them — comes out tomorrow. In this one, I show that it wasn't just prices that boosted Dems. Deportations may have even driven as much net D vote margin as prices www.gelliottmorris.com/p/2026-11-17-republicans-midterms-…
-
G Elliott Morris
@gelliottmorris
New: The 2026 election was not all about turnout. On several issues, Dems held their biggest leads since 2007!
This matters because it shows the limits of the "popularist" approach to electoral strategy. It also gives the Ds a policy roadmap for 2027
www.gelliottmorris.com/p/2026-11-15-sunday-roundup -
G Elliott Morris
@gelliottmorris
The discourse on electability has an extremely *mechanically tuning a dial that says “ideology” on it and looking back at the electorate* vibe right now. Stuff like this shows the limits of survey experiment write-ups that lack a coherent behavioral model (20 points?? fr?? cap)
-
G Elliott Morris
@gelliottmorris
Support for impeaching Pete Hegseth is somewhere around +5 to +15 net, and depends heavily on the phrasing of the question blog.fiftyplusone.news/p/hegseths-impeachment-may-be-tabled
-
G Elliott Morris
@gelliottmorris
New post: The bad news — and good news — about how the polls did in 2026
Polls didn't suffer from the non-response bias that rocked the industry in 2016 and 2020.
But they understated Democrats for classic reasons that will never be completely "fixed"
www.gelliottmorris.com/p/2026-11-13-polls-missed-turnout -
G Elliott Morris
@gelliottmorris
It seems like a lot of polls that missed last week were just weighting by recalled vote incorrectly
Raphael Nishimura
@rnishimura
"Anchoring on 2024 presidential recall pulls the sample toward an electorate that didn’t come back."
I wish people, especially pollsters (!), realized that, when applied correctly, that's not what recalled vote weighting is doing! -
G Elliott Morris
@gelliottmorris
Solid chance Trump’s approval is in the 20s among adults before Election Day 2028. He’s at 35% now according to fiftyplusone.news but has flirted with 34% previously
-
G Elliott Morris
@gelliottmorris
Republican campaign and right-leaning polling firms that looked super accurate from 2016-2024 just whiffed for 2026. Their house effects persisted, but because bias from nonresponse and LV filters changed, the firms tilted the wrong direction www.ms.now/opinion/midterm-polls-mississippi-senate-scott…
-
G Elliott Morris
@gelliottmorris
Just posted the first in a series of articles about how Dems won in 2026. This one starts — where else? — with the Trump of it all. He claims Republicans would have done better if he was “on the ballot,” but the data shows the opposite: he dragged Rs down www.gelliottmorris.com/p/2026-11-11-how-democrats-won
-
G Elliott Morris
@gelliottmorris
From January 2026:
"If you plug in Dems’ current 4.2-point lead today, to the formula in section 1 (party out of power gains, on average, 5 points from Feb. to Nov.), you get an estimate of Dems winning the election by roughly 9 points in November."
www.gelliottmorris.com/p/when-midterm-polls-matter
G Elliott Morris
@gelliottmorris
(This, of course, ends up undershooting Democrats eventual margin of 13+ points, but is exactly where the polls were on Election Day)
-
G Elliott Morris
@gelliottmorris
Elite party strategists need to spend less time posting on X and more time talking to real voters in real life.
G Elliott Morris
@gelliottmorris
by this logic, democrats should also run campaigns that promote gambling, getting a divorce, and smoking a fat one
Jeff Stein
@jstein_sun
.@AJentleson: “i think this is yet another teachable moment for lefties on twitter to engage in some introspection and realize that they are not the average voter! they may think spanking is wrong, but three-quarters of americans don’t” including half of Dems. [he cites below]
-
G Elliott Morris
@gelliottmorris
by this logic, democrats should also run campaigns that promote gambling, getting a divorce, and smoking a fat one
Jeff Stein
@jstein_sun
.@AJentleson: “i think this is yet another teachable moment for lefties on twitter to engage in some introspection and realize that they are not the average voter! they may think spanking is wrong, but three-quarters of americans don’t” including half of Dems. [he cites below]
-
G Elliott Morris reposted
Will Jordan
@williamjordann
99% of House districts moved against president's party this year -- biggest in modern history & clearly a step change, but part of a long term trend towards "universal swing" in House districts (2022 was a non-wave, but similar to 2010 at 84%).
Median district +14D, also a high -
G Elliott Morris
@gelliottmorris
now taking nominations for the 'worst polling take of 2026' award. (lots to choose from!)
-
G Elliott Morris
@gelliottmorris
this is one hell of a chart (from Adam Bonica)
-
G Elliott Morris
@gelliottmorris
Incorporating 2026 results, the cycle-to-cycle correlation in polling error since 1990 (I mean this directionally — "signed error" as AAPOR calls it) is now -0.14. Before 2026, it was -0.04. There is no relationship between lagged cycle polling error and error in the next cycle.
-
G Elliott Morris
@gelliottmorris
An interesting counterfactual (and has the benefit of hindsight) is to look at the 2026 midterms with less campaign spending. Given the effect size of extra ads, etc, Democrats arguably could have won back the Senate w a ~25-35% reduction in spending. (Don’t tell the consultants)
-
G Elliott Morris
@gelliottmorris
Notwithstanding the Mississippi miss, I do think it's time to retire the term "tossup" permanently. All the race handicappers agree that they don't actually believe tossups are 50-50, with some saying they expect all of them to go one way. Just call them "pending"
Scott Colom
@scott_colom
At some point, we have to talk about @CookPolitical and other “prognosticators” who had me and other southern candidates labeled as solid R for much of the campaign cycle. They made it much more difficult to fundraise and are part of the reason the South has been neglected for so long.
They’re only a reflection of the past and mysteriously avoid any accountability when they’re wrong. One of their “analysts” was extremely condescending to me (not Jessica) when I tried to make my pitch for why they were wrong about my rating like he knew something about MS politics when he’s never been to the Sip. This attitude hurt Jaime Davis & others so much. -
G Elliott Morris
@gelliottmorris
An evaluation of one’s forecasts should also include what forecasters said about control of each chamber blog.fiftyplusone.news/p/fiftyplusone-first-among-forecas…
Gil
@RFishBlueFish
Log loss / brier score of correlated outcomes is not very useful aghhhhhhhhh. Publish (some approximation of) the joint log likelihood of the total results!
This post is unavailable. See it on X. -
G Elliott Morris
@gelliottmorris
TFW donald trump has a better grasp of public opinion than you do
-
G Elliott Morris
@gelliottmorris
Trump says his immigration policy was “too tough” and “hurt [Republicans]” in the midterms.
This was obvious in the data in April 2025 (see link), despite many pundits’ continued calls for Dems to ignore ICE and deportations as recently as last month
www.gelliottmorris.com/p/kilmar-abrego-garcia-immigration…
60 Minutes
@60Minutes
President Trump says his administration’s deportation policy was “too tough.” www.cbsnews.com/news/donald-trump-2026-post-midterm-elect…
This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
Final political yard sign count from rural Virginia: Trump/Republicans 2, Warner/Dems 6
All the usual Trump 2024/28 barns have been repainted -
G Elliott Morris
@gelliottmorris
A lot of cool data here
Wave Polling and Research
@wave_polling
Wave’s Atlas of Public Opinion has been updated to use the congressional district boundaries for the 120th Congress 🎉🎉
wavepolling.com/atlas/q/immig_legalize/?level=cd -
G Elliott Morris
@gelliottmorris
The Economist found a 1-point & inconsistent penalty for progressives this year, with close to zero effect in the House. According to them Melat Kiros and Claire Valdez beat expectations, while Troy Jackson and El Sayed fell short. Turnout fx much larger
www.economist.com/graphic-detail/2026/11/06/the-blue-wave… -
G Elliott Morris
@gelliottmorris
Models > markets
Polls > price action
www.status.news/p/kalshi-midterm-elections-odds-media-par… -
G Elliott Morris
@gelliottmorris
Here’s the final House level generic ballot predictions from my MRP modeling of surveys gathered for the team at Wave Polling, which came out at D+12 nationally among LVs. Median simulation was 250 D seats. (Guess I should have posted this publicly before the election!).
-
G Elliott Morris
@gelliottmorris
NYT says increasing the salience of deportations contributed to Democrats’ victories
-
G Elliott Morris
@gelliottmorris
An independent analysis by the New York Times finds FiftyPlusOne had the most accurate election forecasting model over the whole 2026 election cycle, beating both prediction markets and other forecasters at almost all time horizons. Congrats team! www.nytimes.com/interactive/2026/11/06/business/polymarke…
G Elliott Morris
@gelliottmorris
You too can get the data behind the most accurate forecasts of American politics:
fiftyplusone.news/downloads -
G Elliott Morris
@gelliottmorris
Black voter turnout held at ~96% of 2024, the highest rate for any midterm in at least a decade, in the Deep South on Tues. If Scott Colom manages to win Mississippi's Senate seat, it wouldn't have been possible without sky-high engagement of Black voters
www.gelliottmorris.com/p/2026-11-10-black-turnout-south -
G Elliott Morris
@gelliottmorris
I just posted this analysis of the @50plus1news model vs the prediction markets. Not only is @NateSilver538 correct and @mansourtarek_ wrong about the performance of the models vs markets over the cycle, the models were actually more accurate on Elec Day 2 blog.fiftyplusone.news/p/fiftyplusone-first-among-forecas…
Chris Hayes
@chrislhayes
I am 100% Team Nate in this beef. It’s *obvious* to anyone paying attention that the betting markets just moved over time towards the models. That’s the info they were working off of! They didn’t “discover” anything
Nate Silver
@NateSilver538
Give me a break, the polls and models provided the direction in contast to vibes and dumb insiders. And yes it's good that traders were smart enough to trust them. But the better election models were more confident about the outcome than the markets and got there more quickly.
Tarek Mansour
@mansourtarek_
Last night, the power to decide what America sees changed hands.
I have spent a ton of time in DC over the last few months. The elites were certain the markets were off. I heard many variants of “no chance markets are right about X” and “I’ve been in politics for a decade. ‘These people’ don’t really know much.”
Last night, ‘these people’ beat the elites.
Prediction market favorites won 479 of the first 487 races. That’s 98%.
The elites live in a bubble. They theorize, write, ideate, and mistake credentials for an understanding of lives they’ve never lived. They don’t see and feel what the everyday man, on the ground, sees and feels.
Knowledge is not exclusive to the elites but scattered across millions of people.
Someone knows their hometown better than any national pundit. Someone has followed a local race passionately for months. Someone has followed a local race dispassionately for months. Someone notices a shift in their community before it makes the headlines. Someone is shaped by an incident at work. Someone is shaped by what’s getting harder or easier for their families.
Each holds a piece of the picture that no expert can see in full.
And for the longest time, that knowledge had no way to reach the rest of the world.
For generations, election night meant turning on the TV and waiting for a handful of people to tell you what happened.
Prediction markets change that.
You don't need a title, a large following, or an editor to publish you. All you need is a perspective and the conviction to back it.
That changes who holds power over information.
Anyone can see the market. Anyone can contribute to it. Anyone can question it. No title gives you the final word. No gatekeepers can filter your view.
Sometimes the signal can be wrong. But anyone can challenge the consensus and improve it.
And when people consolidate their knowledge together, something remarkable happens: millions of individual perspectives become a signal that we all can learn from.
Over the last few months, the people showed up on Kalshi to tell the world what they were going to do.
Last night, the megaphone was passed to them. -
G Elliott Morris
@gelliottmorris
Second day in a row in the top 5 growing publications on Substack. You can’t argue with Value Above Replacement of Strength In Numbers, folks! www.gelliottmorris.com/p/democrats-win-us-house-senate-in…
-
G Elliott Morris
@gelliottmorris
some of us have been saying www.gelliottmorris.com/p/not-just-left-vs-right-most-voters
Osita Nwanevu
@OsitaNwanevu
Very good interview between @EricLevitz and @davidshor here.
-
G Elliott Morris
@gelliottmorris
You don't need to look at prediction markets at all. They are not oracular prediction machines. They are gambling sites optimized to transfer your hard-earned money to their shareholders. Get your election data straight from the source www.nytimes.com/2026/10/30/technology/polymarket-kalshi-m…
-
G Elliott Morris reposted
Raphael Nishimura
@rnishimura
"The markets beat the polls" is an utterly nonsensical statement here.
Not only because polls are not trying to predict the election outcome, but also because the bettors in the markets are using polls to inform their bets! -
G Elliott Morris
@gelliottmorris
Extreme levels of cope happening online in how centrists are responding to DSA candidates being electoral overperformers. This invalidates almost all of their predictions from 2024-2025 “WAR” debates
-
G Elliott Morris
@gelliottmorris
I swear the victory lap will end soon, but just noting the total vindication of this take that reducing voters to single position on the left-right axis yields terrible predictions/inferences of voter behavior www.gelliottmorris.com/p/the-strategists-fallacy
-
G Elliott Morris
@gelliottmorris
Preliminary still, but looking like a slightly higher poll error than the average miss. Nothing like 2020 tho—on par with 2024/16.
Two weeks before the election I wrote at Strength In Numbers how a normal sized polling error could result in a Dem blowout www.gelliottmorris.com/p/2026-10-16-senate-polling-miss-b… -
G Elliott Morris
@gelliottmorris
I was really annoyed at the debate about what empirical measure to use for ideology so I coded a lil script to use a LLM judge candidate ideology with pairwise comparisons and then put everyone on a single L-R scale (after Patrick Wu's work).
Results are here. Judge for yourself
Generally, in the Senate, we are seeing a smaller benefit for moderates than the predictions made by the centrist crowd during the 2025 vintage of this debate.
And in the House, DSA candidates are running better than the average Democrat!!
Derek Thompson
@DKThomp
I think this is right.
It's not just that Talarico (et al) combines elements of left populism, abundance pragmatism, and cultural moderation. It's also that the definitions of "left" and "moderate" are in flux.
Take "left." Talarico's all-of-the-above position on energy, eg, would have been grounds for near excommunication from the climate left during the heights of the Green New Deal era. But many leftists now proudly claim him as one of their own, in part bc of his rhetoric on corporate abuse of power and "not left vs right, but top v bottom."
So what we're seeing might be fusionism. But just as importantly, the elements being fused are in flux.
Rotimi Adeoye
@_rotimia
The Democrat factional fighting is a relic of 2016. This election points toward a new liberal fusionism, candidates who combine leftist economic populism with centrist pragmatism. Talarico and Ossoff offer a model on this. The factional stuff is old news, beltway chatter.
Derek Thompson
@DKThomp
i think "moderates did better on tuesday" and "anti-rich vibes used to be identifiably 'left-wing' but they've become fairly normie in the last few years" are both accurate takes, altho i expect a lot of discourse warfare between these positions in the next few days
-
G Elliott Morris reposted
Adam Carlson
@admcrlsn
In the last 20 hours you have posted three times pitting this as a binary ideological battle between the moderate/establishment “faction” and the progressive/leftist “faction.”
So either you were being disingenuous then, or you are being disingenuous now. There is no third thing.
You said that since progressive candidates perform led slightly worse than moderate ones — which on balance is accurate but there are many exceptions (including Randy Villegas) and race/candidate/cycle-specific factors at play in individual races that you intentionally and irresponsibly choose to ignore — the party shouldn’t listen to their advice on how to win. Even though the objectively successful “fusion” message employed by the Ossoffs and Talaricos of the world adopted a lot of the economic populist messaging honed by the progressive-left over the last decade or so.
And you basically said that anyone that disagrees with you is just stupid (a lot of what ails the left “stems from poor epistemic practices”).
I want a party that takes good ideas for all corners of the party (including moderates!).
All of this is incredibly nuanced, but you are — and have been, even in the weeks leading up to the election — trying to lay the groundwork for a blunt and deeply unsophisticated “us vs. them” dynamic that serves nobody but people like you who want clicks on your Substack (and any corporate clients you have that would benefit from said debate/outright moderation on policy).
You took like a campaign tactic in popularism — which at a base level makes sense — and have contorted it into an entire ideology. It’s just poll- and focus-group tested centrism with no room for outside the box thinking, which — as anyone who has worked in politics or campaigns can attest — is not always the optimal strategy! And I say this as a longtime pollster and focus group moderator!
Matthew Yglesias
@mattyglesias
I 100 percent agree with Litman.
Rather than moderates (boo! hiss!) voters are looking for people who solve problems (there's a whole problem solvers caucus!) and who say things they agree with.
It's all about schmoderation.
www.slowboring.com/p/shmoderation-is-the-future
Amanda Litman
@amandalitman
Agree with all this. Moderate vs progressive is not the right axis to evaluate candidates (and it’s not how most voters understand themselves.) They want people who will solve their problems competently and who can speak to their emotional reality, not gaslight them by saying “no, you’re wrong, the economy is good” or “that’s a distraction.”
Rotimi Adeoye
@_rotimia
The Democrat factional fighting is a relic of 2016. This election points toward a new liberal fusionism, candidates who combine leftist economic populism with centrist pragmatism. Talarico and Ossoff offer a model on this. The factional stuff is old news, beltway chatter.
Derek Thompson
@DKThomp
i think "moderates did better on tuesday" and "anti-rich vibes used to be identifiably 'left-wing' but they've become fairly normie in the last few years" are both accurate takes, altho i expect a lot of discourse warfare between these positions in the next few days
-
G Elliott Morris reposted
Adam Carlson
@admcrlsn
So, to borrow your own language from yesterday, people trying to pit the center-left & progressive-left against each other in a blunt, unnuanced, mutually exclusive manner — and that includes Third Way/Welcome PAC in the center & hordes of lefty groups/poasters — suffer from “poor epistemic practices”
Adam Carlson
@admcrlsn
In the last 20 hours you have posted three times pitting this as a binary ideological battle between the moderate/establishment “faction” and the progressive/leftist “faction.”
So either you were being disingenuous then, or you are being disingenuous now. There is no third thing.
You said that since progressive candidates perform led slightly worse than moderate ones — which on balance is accurate but there are many exceptions (including Randy Villegas) and race/candidate/cycle-specific factors at play in individual races that you intentionally and irresponsibly choose to ignore — the party shouldn’t listen to their advice on how to win. Even though the objectively successful “fusion” message employed by the Ossoffs and Talaricos of the world adopted a lot of the economic populist messaging honed by the progressive-left over the last decade or so.
And you basically said that anyone that disagrees with you is just stupid (a lot of what ails the left “stems from poor epistemic practices”).
I want a party that takes good ideas for all corners of the party (including moderates!).
All of this is incredibly nuanced, but you are — and have been, even in the weeks leading up to the election — trying to lay the groundwork for a blunt and deeply unsophisticated “us vs. them” dynamic that serves nobody but people like you who want clicks on your Substack (and any corporate clients you have that would benefit from said debate/outright moderation on policy).
You took like a campaign tactic in popularism — which at a base level makes sense — and have contorted it into an entire ideology. It’s just poll- and focus-group tested centrism with no room for outside the box thinking, which — as anyone who has worked in politics or campaigns can attest — is not always the optimal strategy! And I say this as a longtime pollster and focus group moderator!
Matthew Yglesias
@mattyglesias
I 100 percent agree with Litman.
Rather than moderates (boo! hiss!) voters are looking for people who solve problems (there's a whole problem solvers caucus!) and who say things they agree with.
It's all about schmoderation.
www.slowboring.com/p/shmoderation-is-the-future
Amanda Litman
@amandalitman
Agree with all this. Moderate vs progressive is not the right axis to evaluate candidates (and it’s not how most voters understand themselves.) They want people who will solve their problems competently and who can speak to their emotional reality, not gaslight them by saying “no, you’re wrong, the economy is good” or “that’s a distraction.”
Rotimi Adeoye
@_rotimia
The Democrat factional fighting is a relic of 2016. This election points toward a new liberal fusionism, candidates who combine leftist economic populism with centrist pragmatism. Talarico and Ossoff offer a model on this. The factional stuff is old news, beltway chatter.
Derek Thompson
@DKThomp
i think "moderates did better on tuesday" and "anti-rich vibes used to be identifiably 'left-wing' but they've become fairly normie in the last few years" are both accurate takes, altho i expect a lot of discourse warfare between these positions in the next few days
-
G Elliott Morris
@gelliottmorris
AtlasIntel has deliberately excluded Strength In Numbers/ @verasight_data from their analysis of pollster accuracy. Depending on whether you use a benchmark natl vote of D+13 or D+14, we are either all tied for first or coming in at 1 and 2. Very dishonest www.gelliottmorris.com/p/democrats-win-us-house-senate-in
AtlasIntel
@atlas_intel
POLLSTER ACCURACY
🇺🇸 2026 MIDTERMS
National Generic Ballot
AtlasIntel was the most accurate pollster of the 2026 Midterms.
Please note: the ranking is unlikely to change but exact errors depend on the final vote count. An updated table will reflect the final data.
G Elliott Morris
@gelliottmorris
They say they get their data from Silver Bulletin, where Verasight is included, so this is an intentional omission
-
G Elliott Morris
@gelliottmorris
I can’t *believe* the centrist think tanks, with their hundreds of millions donated to push centrist policy, and their friend in the press who get paid for the same, are publishing laughable analysis claiming centrists beat avg Dems by 20 points. I am truly shocked
-
G Elliott Morris
@gelliottmorris
Updating my county swing dashboard projects.gelliottmorris.com/election-night-2026/
-
G Elliott Morris
@gelliottmorris
DSA candidates and Blue Dogs both beat the Democratic Party average last night. Edge is small esp. once you add controls for demographics and other candidate characteristics (e.g., anybody would have crushed Paxton; Collins would have done well against any Maine Dem).
Post not available here, view on X
G Elliott Morris
@gelliottmorris
it is notable tho that certain pundits spent years posting misleading charts on "WAR" ignoring the empirical objections (esp that this isn't causal and can't predict out of sample) and lo and behold, DSA House candidates are outperforming now and they are all silent about it
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
New post: FiftyPlusOne first among forecasters in calling 2026 U.S. House election, second in Senate
The FiftyPlusOne model saw a big Democratic victory coming while prediction markets and expert forecasters lagged blog.fiftyplusone.news/p/fiftyplusone-first-among-forecas… -
G Elliott Morris
@gelliottmorris
From me for MS NOW (part of their deal w FiftyPlusOne!) on how the Democrats managed to win their biggest* (*or second biggest, depending on exact digits) victory in the House popular vote since 1974. Big swings with Latinos, youngsters — and rural voters www.ms.now/opinion/how-gop-lost-trump-swing-voters
-
G Elliott Morris
@gelliottmorris
chamber briers tell a dramatically different story FWIW (split ticket still does well)
Lakshya Jain
@lxeagle17
If you impute according to who markets have favored in NC-03/AZ-02/NV-02/FL-09/WA-05 and MS-Sen and AK-Sen, this is how the model grades shake out for this cycle.
Pretty happy with this! But we'll have to wait for the full results for a more detailed postmortem. -
G Elliott Morris
@gelliottmorris
with the benefit of hindsigh, yeah, you could have just taken the small dollar donor map from vote hub to the bank 2 weeks before the election votehub.com/2026/10/24/2026-donations-map/
-
G Elliott Morris
@gelliottmorris
if you had been making trades on prediction markets based on the 50+1 election model, which I absolutely do not recommend or endorse, congratulations you made a lot of money last night
-
G Elliott Morris
@gelliottmorris
Last night was a bad night for any1 who argued, at some point over the last 2 yrs,
- trump gains with latino + young voters marked a durable realignment to the right
- higher turnout helps GOP
- trumpapalooza was going to save GOP
- polls would overshoot dems
god forbid all 4 -
G Elliott Morris
@gelliottmorris
looks like the average BlueSky poster may in fact be closer to the median voter than the average X user
-
G Elliott Morris
@gelliottmorris
I'm off to bed, but the House seats and popular vote forecast at the Strength In Numbers site will continue updating as new ballots are counted. (It's possible something will break while I'm asleep. Use official sites for actual vote returns!) projects.gelliottmorris.com/election-night-2026/
-
G Elliott Morris
@gelliottmorris
New post: Democrats win U.S. House, Senate in biggest incumbent-party defeat since Watergate. The Strength In Numbers/Verasight poll was tied for first in predicting the Democratic House blowout
www.gelliottmorris.com/p/democrats-win-us-house-senate-in -
G Elliott Morris
@gelliottmorris
Sup?
Lakshya Jain
@lxeagle17
What is the point of presenting it like this? What favors are you doing an audience that doesn't know the composition of the seats? When you have partisanship shrinking the battlegrounds, this just doesn't work. I see your disclaimer, but nobody reads that far down.
This is really misleading to an audience that is desperately looking for confirmation that Trump is unpopular and that his partly will lose seats (they will). You know as well as I do that Dems are very likely to regain the House, but that this model way overstates the gains. -
G Elliott Morris
@gelliottmorris
Preliminary Senate polling error chart here. This is why you shouldn't assume that polling error will go in one direction! Fatal blow for the unskewers!! www.gelliottmorris.com/i/223234792/1105-pm-preliminary-po…
-
G Elliott Morris
@gelliottmorris
ann selzer wasn't wrong, she was just early
-
G Elliott Morris
@gelliottmorris
it is officially woke 3
-
G Elliott Morris
@gelliottmorris
Tsunami watch update
www.gelliottmorris.com/i/223234792/1015-pm-tsunami-watch-… -
G Elliott Morris
@gelliottmorris
This is shaping up to be the biggest House popular vote victory for any party since 1974
This is a Watergate-level repudiation of the party in power
(A prospect I raised here: www.ms.now/opinion/trump-polls-unpopular-midterms-undecid… ) -
G Elliott Morris
@gelliottmorris
SUBSCRIBE TO STRENGTH IN NUMBERS
www.gelliottmorris.com/welcome -
G Elliott Morris
@gelliottmorris
FORECASTERS VINDICATED fiftyplusone.news/forecast/house
-
G Elliott Morris
@gelliottmorris
POLLSTERS VINDICATED
-
G Elliott Morris
@gelliottmorris
This map shows the change in Democratic margin in House vs Pres elex from 2024 to 2026
Blue means Dems doing better, red Republicans doing better
www.gelliottmorris.com/i/223234792/825-pm-this-map-shows-… -
G Elliott Morris
@gelliottmorris
has anyone checked in on ruffini yet
-
G Elliott Morris
@gelliottmorris
54-55 seats very possible at this point
G Elliott Morris
@gelliottmorris
wasn't going to share this publicly but it seems to be working for now and... looks pretty tsunami-esque chat, don't think you could conclude anything else from these results so far www.gelliottmorris.com/i/223234792/745-pm-early-but-looki…
-
G Elliott Morris
@gelliottmorris
wasn't going to share this publicly but it seems to be working for now and... looks pretty tsunami-esque chat, don't think you could conclude anything else from these results so far www.gelliottmorris.com/i/223234792/745-pm-early-but-looki…
-
G Elliott Morris
@gelliottmorris
Looking at swings in fully reported counties, and folks, things are looking blue... definitely a wave. tsunami adjacent
www.gelliottmorris.com/i/223234792/745-pm-early-but-looki… -
G Elliott Morris
@gelliottmorris
So if you apply that 14 point swing to the whole district, which is not not wrong, IN-05 is a pure tossup (was Trump +17 in 2024 and +16 in 2020)
Zachary Donnini
@ZacharyDonnini
This drop was only D+13 by past primary history. Ford won it by 27.
Zachary Donnini
@ZacharyDonnini
Indiana U.S. House District 5
Hamilton County, IN (Early Vote):
🔴 2024: Spartz +4
🔵 2026: Ford +27
This 31-point swing toward Democrats is significant. Victoria Spartz, whom VoteHub rates Likely R, is now in serious trouble. IN-05 is a genuinely competitive race. -
G Elliott Morris
@gelliottmorris
House district benchmarks for a "blue wave" vs tsunami ("bluenami," i'm told)
www.gelliottmorris.com/i/223234792/550-pm-wave-or-tsunami… -
G Elliott Morris
@gelliottmorris
THESE ARE THE ONLY NUMBERS I WILL SHARE FROM THE EARLY EXIT POLLS ("voter polls")www.gelliottmorris.com/i/223234792/515-pm-the-only-number…
G Elliott Morris
@gelliottmorris
if this ends up being right... wowza
-
G Elliott Morris
@gelliottmorris
Time for a quick level-set about how much early turnout data matters. This post tells you how to update priors from a pre-election forecasting model with turnout data but without tallied votes.
www.gelliottmorris.com/i/223234792/311-pm-a-quick-level-s… -
G Elliott Morris
@gelliottmorris
Some people are posting misleading analyses of turnout reports (which show a Dem enthusiasm advantage of ~10 points, but which you shouldn't read too much into for predicting votes, because what really matters in close races is how the independents break)
www.gelliottmorris.com/i/223234792/230-pm-some-people-are… -
G Elliott Morris
@gelliottmorris
2:20 PM update: THE REGRESSION COEFFICIENT IS THE SAME EVERYWHERE YOU LOOK www.gelliottmorris.com/i/223234792/220-pm-its-the-same-re…
-
G Elliott Morris
@gelliottmorris
I do not think the numbers in FL are good enough for Dems to pull off a victory, but it depends on how the NPAs break.
What ballot data is primarily useful for is identifying the likely swing since the last election and comparing that to the polls, so we can have a prior on polling error going into vote tallying.
The vote returns all over the country — FL, VA, NV, AZ, KY etc — all show a relative turnout advantage for Dems that is above 2024, 2022, and verging on 2018. Somewhere around 10 point differential turnout/
This is basically what the polls showed, and so gives some validation of at least one of the underlying mechanisms of Dem gains in the polls — a turnout edge for Dems manifesting in a large LV - RV gap.
There is a lot of remaining uncertainty about (a) partisan vote loyalty rates and (b) the precise breakdown of NPAs.
But you know what, it's not true, as some people are suggesting on here, that the FL data show Rs having a good night. In nominal terms, yes -- but actually what they show is the polls picked up on a real enthusiasm advantage for Dems.
Will that hold? We'll have to wait and see!
I am keep track of all these turnout reports here: www.gelliottmorris.com/i/223234792/220-pm-its-the-same-re…
Dave Wasserman
@Redistrict
Update: as of 2pm, Florida has crossed 7M ballots cast (final was 11M in 2024, 7.8M in 2022). Registered Ds are at 71% of their 2024 total turnout, Rs at 63%, NPA/others at 56%.
Overall turnout is R+7.1 pts (2024 final was R+11.1 pts). Modest but clear Dem enthusiasm advantage. -
G Elliott Morris
@gelliottmorris
Nevada Election Day numbers now on the dashboard
www.gelliottmorris.com/i/223234792/200-pm-nevada-election… -
G Elliott Morris
@gelliottmorris
1:35 PM live blog update: Use this tool from Strength In Numbers to update our election forecast as votes are counted
www.gelliottmorris.com/i/223234792/100-pm-use-this-tool-t… -
G Elliott Morris
@gelliottmorris
Fresh E-Day turnout data in VA, courtesy of State Navigate, implies Dem **E Day voters** are voting at a rate about 9-10 percentage points higher than Republicans. This is on top of the Democratic lead in engaging early voters in the state.
www.gelliottmorris.com/i/223234792/1143-am-election-day-t…
G Elliott Morris
@gelliottmorris
Have been writing this for weeks with early vote data. Big turnout problem for Rs everywhere you look
-
G Elliott Morris
@gelliottmorris
The current GOP path to victory in a nutshell:
www.gelliottmorris.com/i/223234792/1143-am-election-day-t… -
G Elliott Morris
@gelliottmorris
it's the same story everywhere
The AZ - abc15 - Data Guru
@Garrett_Archer
Election day ballot types so far in Maricopa as of 9am. Turnout is running under 2022.
-
G Elliott Morris
@gelliottmorris
Florida election simulator based on live returns is now online!
Make assumptions about remaining turnout and party loyalty rates and see how that affects the likely outcome of statewide races.
Defaults will update live w/ actual ballots as they are cast
www.gelliottmorris.com/i/223234792/1126-am-florida-electi… -
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
My current read on the Election Day returns, and then I'm stepping away to go on a run, is that things look roughly in line with both:
(A) extrapolations from early vote
and
(B) the polls
Maybe even a hint of modest (1-2pt) underestimation of Dems
www.gelliottmorris.com/i/223234792/1015-am-suggestions-of… -
G Elliott Morris
@gelliottmorris
GOP turnout is also lagging in Virginia, per State Navigate. Combined with a larger Dem. lead in early voting relative to 2022, that's bad news for GOP
www.gelliottmorris.com/i/223234792/1000-am-virginia-turno… -
G Elliott Morris
@gelliottmorris
Look it's early and just one state and etc etc but I'm not seeing anything indicating that the inference from early voting, that GOP turnout is way down, ISN'T applying to Elec Day voters as well.
There's currently no Republican Election Day surge making up for their decline in early voting.
Here fresh numbers from Florida:
www.gelliottmorris.com/i/223234792/915-am-election-day-tu… -
G Elliott Morris reposted
Peter Schorsch
@PeterSchorschFL
The polling-error chart: half the factors flatter Democrats, half shortchange them, most by a point or less.
Typical final Senate miss is about 3 points. No call on direction.
via @gelliottmorris for Strength In Numbers
www.gelliottmorris.com/p/2026-10-29-your-guide-to-polling… -
G Elliott Morris
@gelliottmorris
I'm watching the @StateNavigate team live call city/county election officials in Virginia and record turnout live. Riveting stuff and very helpful in verifying if extrapolations from early vote are holding
projects.statenavigate.com/interactives/va/election-day-2… -
Post not available here, view on X
G Elliott Morris
@gelliottmorris
Will be updating this throughout the day on my election day live blog:
www.gelliottmorris.com/p/2026-11-03-election-day-live-blog -
G Elliott Morris
@gelliottmorris
No sign yet of the Election Day surge in turnout Republicans need to meet their 2024 or 2022 turnout patterns. But it’s early, and will have much better picture in a few hours.
www.gelliottmorris.com/i/223234792/807-am-early-turnout-l… -
G Elliott Morris
@gelliottmorris
Follow the results of the midterms live as they come in with Strength In Numbers! The Election *day* blog is up here, and for election *night* we’re doing a livestream from 6:45 Eastern until who knows when!
www.gelliottmorris.com/p/2026-11-03-election-day-live-blog -
G Elliott Morris
@gelliottmorris
FINAL forecast: Will it be a blue wave, or a blue tsunami? Democrats favored to win 240 House and 52-53 Senate seats on average. Polling error could be enough for Rs to hold the Senate, but House would need biggest miss in 25 years www.gelliottmorris.com/p/2026-11-03-election-day-preview
-
G Elliott Morris
@gelliottmorris
Our final 2026 election forecast update at FiftyPlusOne: The final polls and race ratings show a late shift toward Dems
The handicappers had been lagging quant forecasts like ours, which saw the shift toward Democrats coming months ago. Now they all align
blog.fiftyplusone.news/p/the-final-polls-and-race-ratings… -
G Elliott Morris
@gelliottmorris
It's 12:01! Happy Election Day!
No more polls will be aggregated at fiftyplusone for 2026. We have started our final average and forecast model runs. Final article to come.
Happy voting! -
G Elliott Morris
@gelliottmorris
The 50+1 polling averages and forecasting model will be updated sporadically this evening as our teams add the flood of final surveys coming out this evening, and we'll lock it shortly after midnight tomorrow morning. fiftyplusone.news
-
G Elliott Morris
@gelliottmorris
as a bonus, here is NC early voting data. partisan turnout differentials among infrequent *early voters* tends to predict partisan turnout differentials *among election-day voters* pretty well. www.gelliottmorris.com/p/2026-early-vote-blog
G Elliott Morris
@gelliottmorris
FINAL post on the Strength In Numbers early vote blog, with the data from the final in-person early voting day in Florida. Democrats posted their highest share of early voters since 2018.
The only hope for Republicans is a massive shift to voting on E-Day
www.gelliottmorris.com/p/2026-early-vote-blog -
G Elliott Morris
@gelliottmorris
FINAL post on the Strength In Numbers early vote blog, with the data from the final in-person early voting day in Florida. Democrats posted their highest share of early voters since 2018.
The only hope for Republicans is a massive shift to voting on E-Day
www.gelliottmorris.com/p/2026-early-vote-blog -
G Elliott Morris reposted
FiftyPlusOne.news
@50plus1news
Hey pollsters! We'll be running our final average and forecast simulations ahead of the election just after midnight on Tuesday morning. No 2026 polls will be aggregated if they are published later than midnight. Send your data our way at data -at- fiftyplusone -dot- news.
-
G Elliott Morris
@gelliottmorris
Democrats are more likely than not to win both the House and the Senate tomorrow, according to our forecasts at @50plus1news. But remember:
-
G Elliott Morris
@gelliottmorris
Texas Latino voters say they will vote against Republicans because of Trump’s mass deportation policies, which they say are “scary” and hurting the economy. Immigration sure seems like a winning issue for Dems this year, at least if you talk to voters
open.spotify.com/episode/5PbCV6cAP5p8CK825HvBff -
G Elliott Morris
@gelliottmorris
This is a very helpful analysis mailchi.mp/goodauthority.org/the-unmaking-of-a-viral-grap…
-
G Elliott Morris
@gelliottmorris
Early vote data shows Latinos, young people, Democrats surging in turnout
New voters are younger, more Latino and much less Republican than new voters in 2022
www.gelliottmorris.com/p/2026-11-02-new-voters-younger-mo… -
G Elliott Morris
@gelliottmorris
I think this is easy to explain. Past NYT polls did not weight by past vote and only controlled party at the sampling stage. Unsurprising that the move to using the “synthetic past vote” variable in the weight would control D bias when non-response among Rs is so high
-
G Elliott Morris
@gelliottmorris
lol “Black voters, who offered tepid support to Dr. El-Sayed in September, now back him, 93-4. Mr. Rogers did not receive the support of any respondent in the city of Detroit (among 43 people).”
www.nytimes.com/2026/11/01/upshot/poll-times-siena-senate… -
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
Something like 40-50% of ballots have been cast in Michigan already. And evidence is that voters take cues from party elites, many of whom on D side are now rallying behind AES. Possible Stevens’ stunt backfires, while she has validated all the progressive critiques of centrists
-
G Elliott Morris
@gelliottmorris
Early voting forecast suggests blue wave, Dems could exceed 2018 turnout advantage. Republicans need highest-ever Election Day vote to meet 2022/24 levels
www.gelliottmorris.com/p/2026-11-01-gop-early-vote-dropoff -
G Elliott Morris
@gelliottmorris
FINAL early vote update in Texas: A GOP bloodbath on the "engagement ladder"
www.gelliottmorris.com/p/2026-early-vote-blog -
G Elliott Morris
@gelliottmorris
Seems relevant given today's news from WaPo about lack of participation from Republicans
Jonathan Robinson
@jon_m_rob
Over on LinkedIn(!!!), Courtney Kennedy (now @ Ipsos) reports that an incentivized mail survey they conducted w/ a 9% response rate (!!!) that shows D+6 vs D+10 in their poll with Reuters on the Generic Ballot.
www.linkedin.com/posts/courtney-kennedy-15b70419_days-bef…This post is unavailable. See it on X. -
G Elliott Morris
@gelliottmorris
Also new at 50+1 today (it's a double-post day!): A record number of women ran for office in the 2026 primaries. Two-thirds of them were Democrats.
From our contributor Meredith Conroy! blog.fiftyplusone.news/p/a-record-number-of-women-ran-for -
G Elliott Morris
@gelliottmorris
New at 50+1: Crosstab-level generic ballot averages! For every age, race, education, gender and party group!
Trends will update live as we aggregate new surveys, and are on the polls page for paying subs: fiftyplusone.news/polls/generic-ballot/generic-ballot
Plus, a write-up here: blog.fiftyplusone.news/p/how-different-demographic-groups -
G Elliott Morris
@gelliottmorris
New poll: Trump won in 2024 by carrying low-engagement voters by double digits. But our fresh polling shows "low-propensity" voters favor Democrats for the midterms.
Just published this chart-loaded article based on 6,000 interviews conducted this month.
www.gelliottmorris.com/p/2026-10-30-low-propensity-voters -
G Elliott Morris
@gelliottmorris
This is why I would be particularly careful with any poll this year that isn't balancing its sampling by party or past vote
www.washingtonpost.com/politics/2026/10/30/we-polled-kans…
I have also found that low-effort survey respondents are much more GOP-leaning, even within party: www.gelliottmorris.com/p/2026-10-29-your-guide-to-polling… -
G Elliott Morris
@gelliottmorris
Picking up absolutely no movement in the generic ballot or Trump’s approval rating in my weekly tracking poll this week. Just completely stable. RV margin here is about D+9. (It’s possible LV numbers have moved because of changing correlates of turnout, though I doubt large fx.)
-
G Elliott Morris
@gelliottmorris
The Strength In Numbers early voting blog has been updated with fresh numbers from Texas:
www.gelliottmorris.com/p/2026-early-vote-blog
In Harris County (Houston), Republicans are failing to get both their infrequent **and habitual** early voters out as often as Dems. 1st time voters are heavily D. -
G Elliott Morris
@gelliottmorris
From the 50+1 team and me: Democrats currently lead in 7 GOP-held Senate seats. Use our new (free!) tool to explore what would happen if the polls miss like they did in 2016/2020/2024 — or 2018/2022
blog.fiftyplusone.news/p/democrats-lead-in-7-gop-held-senate -
G Elliott Morris
@gelliottmorris
Here is a comprehensive overview of all the ways polls could misfire next week, and how much error each factor could cause
www.gelliottmorris.com/p/2026-10-29-your-guide-to-polling… -
G Elliott Morris
@gelliottmorris
Here's a final installment in my series on how polls could go wrong next week. This time, I show evidence that pollsters may not be reaching enough Trump voters from red, rural areas
www.gelliottmorris.com/p/2026-10-28-lab-trump-voters-blue… -
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
Florida early voting update:
Early in-person turnout among Republicans has fallen off a cliff.
At R+0.4, the total early vote in the state is about as blue as 2018, 5 points higher than this point in 2022, and 10 points higher than in 2024.
www.gelliottmorris.com/p/2026-early-vote-blog
G Elliott Morris
@gelliottmorris
Nixon is very likely losing the Florida Senate race. The utility of early vote analysis is mainly in estimating what the aggregate shift in the electorate will be from turnout relative to some past benchmark. Uncertainty from persuasion and NPA exists
Lorenzo Maldonado
@TheRealLoDown
Let’s be honest though. Florida is safe republican and nobody can tell me otherwise. Except for a congressional race in south Florida possibly
-
G Elliott Morris
@gelliottmorris
I think it is wrong to suggest Marist is intentionally rigging their LV screen to hedge their polls toward Republicans. It's much more likely that whatever vote history model they're using is not suited for this turnout environment. Pretty baseless accusation from Lakshya here
Lakshya Jain
@lxeagle17
I try to avoid criticizing other pollsters much nowadays, but this is ridiculous. This is basically just what you do if you don't believe your polls and don't have the guts to stand by your numbers. Democrats aren't doing 7% worse with likely voters in 2026.
-
G Elliott Morris
@gelliottmorris
Not sure what’s up with some of these polls with 10 pt differences between RV and LV samples. Implausible in size generally and also tend to point in the wrong direction. Single samples can be weird when you’re weighting by 12 variables but this is definitely adding uncertainty
-
G Elliott Morris
@gelliottmorris
Dem Senate candidates have lost a little less than 0.5 pts of vote margin over last 2 weeks, while the generic ballot has trended up for them by about 1 point.
Overall state of the race is unchanged: Dems have leads well below size of normal poll error
fiftyplusone.news/forecast/senate -
G Elliott Morris
@gelliottmorris
Brazil elected Lula again tonight.
Poll averages appear to have had Lula and Flávio Bolsonaro effectively tied, maybe Bolsonaro up 1 on Lula — so not sure why the markets had this race at 80-20 Bolsonaro.
Markets might miss big in the U.S. midterms in 9 days too. -
G Elliott Morris
@gelliottmorris
Small-dollar donations to Republican candidates are just apocalyptic. Selected examples here.
Our forecasting research found in-state individual contributions are a strong predictor of results. Way stronger than PAC spending!
votehub.com/2026/10/24/2026-donations-map/ -
G Elliott Morris
@gelliottmorris
In this piece…
1. Some data on what undecided and low-turnout-propensity voters care about, from our latest poll
2. Another EV update: Infrequent Democratic voters are turning out at a much higher rate
www.gelliottmorris.com/p/2026-10-25-sunday-roundup -
G Elliott Morris
@gelliottmorris
We don’t know if the early vote will be predictive bc all Reps could just be waiting to vote until Elec Day. What this means in practice is that to win, the GOP will have to have the most successful EDay mobilization operation in modern history — and that it’s completely stealth
G Elliott Morris
@gelliottmorris
Another possibility for Rs is that undecideds, who are about 5-10% of the electorate, could all break 100-0 for Trump and shift the generic ballot to mid single digits. I’m skeptical of that based on Trump’s closing message but technically there’s a chance!
-
G Elliott Morris
@gelliottmorris
Another example of how these low-quality social media poll “aggregator” accounts operate with no standards or vetting of info is that they tweeted 50+1 was conducting a poll, but are we are very public about how and why we don’t conduct polls!
Just unfollow them!
Adam Carlson
@admcrlsn
Yes, Fifty Plus One explained this in their piece exposing how Wedgewood was faking their data.
They replicated the Wedgewood “poll” to prove it. -
G Elliott Morris
@gelliottmorris
This is not a “poll” that we published. We did gather a pollfish sample in MS to prove the Wedgewood data was fraud, but we have zero faith in the quality of that data and do not recommend weighting it or using the platform weights to create vote estimates.
-
G Elliott Morris
@gelliottmorris
Matthew Yglesias continues to say that talking about ICE and excessive force in immigration enforcement helps Republicans. He is obviously wrong about this and refuses to acknowledge the data. Way past time for everyone to update their priors on this.
G Elliott Morris
@gelliottmorris
It might be true that “Democrats shouldn’t talk about the border.” I’m not sure actually, but let’s grant it’s true for the sake of the argument. It does not follow from that that “Ds shouldn’t talk about ICE shooting people or sending them to foreign gulags” (paraphrasing)
G Elliott Morris
@gelliottmorris
Adding one more thing here: We did a survey experiment in May 2025 that showed that when people heard more about ICE overreach, they moved against the president on deportations. Then the news coverage of the Kilmar Abrego Garcia story verified this www.gelliottmorris.com/p/kilmar-abrego-garcia-immigration…
G Elliott Morris
@gelliottmorris
all the available evidence suggests that attacking trump for immigration overreaches hurts him. that's why the GOP has lost 20 points on trust to handle immigration since 2024 www.ms.now/opinion/trump-polls-unpopular-midterms-undecid…
-
G Elliott Morris
@gelliottmorris
It seems increasingly likely that the fake polls from Wedgewood Polls and SoCal Strategies were simply part of a coordinated shadow op to move betting markets, rather than people trying to boost progressives. Neera is pushing an ideological narrative
x.com/Dan_Zmann/status/2113112455488385131
Neera Tanden🌻
@neeratanden
What I find fascinating here is that a journalist like you doesn’t seem to care at all about an actual fraud and its impact on an election. You’re not curious what happened here? Or is your ire and inquiry just directed at things that hurt your political allies instead of help them?
Yeah I think fraud polls are bad and you can still believe that and lock in to support AES. Doing anything else seems to be sanctioning fraud.
Mehdi Hasan
@mehdirhasan
You spent the past 48 hours on here repeatedly suggesting the Democratic Senate candidate in Michigan won his primary as a result of a left-wing, fake-poll-driven ‘op’ - which is now a Fox News anti-Dem narrative. Maybe you should look in the mirror, Neera.
Neera Tanden🌻
@neeratanden
We are two weeks from the election @mehdirhasan - feel free to ask away afterwards. You are choosing to push out these videos. It really does seem you want to both keep talking about this streamer while chastising people for having talked about him in the past.
Mehdi Hasan
@mehdirhasan
It’s a fair point on the surface, but how I am supposed to point out the double standards (hypocrisy?) without asking the question? It is meta, you’re right.
Neera Tanden🌻
@neeratanden
What I find meta about this whole discourse is @mehdirhasan attacking everyone for talking about a streamer in California when he constantly asks people about him and then pushes out the videos. I pointed this out in my conversation with him at Georgetown when he asked me about Piker. Seems kinda hypocritical
Mehdi Hasan
@mehdirhasan
Me: "Do you know how many times you've tweeted about Piker since you wrote that tweet [about not tweeting about him] on August 30th?"
@mattyglesias: "Probably more than I intended to."
Me: "8 times.... you tweeted about Mike Rogers 6 times. So you're tweeting about the streamer you didn't want to talk about more than the Republican candidate you say you want to defeat."
Yglesias: "Yeah, that's fair."
Me: "Why?"
Yglesias: "What do you want me to say? You got me."
From my new Zeteo interview with the top centrist pundit - on the left, Michigan, Gaza, immigration, polls, electability, and more:
zeteo.com/p/matt-yglesias-mehdi-debateThis media couldn't be displayed. See it on the original post. -
G Elliott Morris reposted
Chaz Nuttycombe
@ChazNuttycombe
Recall 1 point left of 2024
CJ Warnke
@cjwarnke
🚨NEW @HouseMajPAC poll in #va01🚨
🔵Shannon Taylor: 47%
🔴Rob Wittman: 46%
2024 Recall: Trump+4
PPP, 10/21-/10/22, 570 Likely Voters -
G Elliott Morris reposted
Cooper Burton
@cooperrburton
An example of the @50plus1news vetting process at work: The Kingsport Times News reported on a TN-Gov poll yesterday from 'Political Intelligence.' We investigated them back in Feb. and couldn't verify key information, so we aren't aggregating their polls. timesnews.net/news/425247/two-polls-two-different-picture…
-
G Elliott Morris reposted
Verasight
@verasight_data
Thank you to the @McCourtSchool for hosting the second annual Political Analytics Conference.
Pictured below: the State of Polling 2026 panel featuring @RuthIgielnik of The New York Times, @gelliottmorris of Strength in Numbers, @PatrickRuffini of Echelon Insights and Verasight co-founder @pete_enns, moderated by Michael Bailey of Georgetown University.
Photo: Georgetown McCourt School of Public Policy -
G Elliott Morris
@gelliottmorris
pollsters would say no
Matthew Yglesias
@mattyglesias
Polling has Hamilton up by 1 point in a context where the ~10 percent who are still undecided are almost all going to be Trump/Trump/Trump voters.
Shouldn't we expect this to be a GOP win? Conceptually speaking, is it "polling error" if these late deciders break for Hamilton? -
G Elliott Morris
@gelliottmorris
New: Low-info voters, who put Trump in the White House, now look even more anti-Trump than high-info voters, and give Dems bigger margins on the generic ballot.
1 out of 3 Trump voters who can't name the party in charge of Congress now back Dems for 2026
www.gelliottmorris.com/p/poll-low-information-voters-have -
G Elliott Morris
@gelliottmorris
In which the president of the United States references FiftyPlusOne’s sleuthing of fraudulent pollsters
blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-polls…
G Elliott Morris
@gelliottmorris
to be clear he is using the incident for BS partisan political purposes, but it still validating to see our reporting from a very small team having this big of a splash!
-
G Elliott Morris reposted
Mary Radcliffe
@marywitha4
What the hell, I'll pile on. We refused to use a "pollster" earlier this cycle called "Civic Lens." They wouldn't identify themselves, joined twitter just days before their first "poll," and have since deleted their account. Twitter aggregator accounts STILL have their polls up.
G Elliott Morris
@gelliottmorris
Well, the difference between FiftyPlusOne and the social media polling aggregator accounts like this one is that FiftyPlusOne is the organization uncovering the problems here!
And our standards have prevented other fraudsters from being platformed at 50+1 who were highly circulated by the social media aggregators. For example, the fake 'social experiment' polls from "Median Strategies" published earlier this year never appeared on 50+1 or e.g. the New York Times, but moved like wildfire on here because of now-deleted tweets from this account..
Yes, standards will have to get higher because of AI and the rising incentives for fraud from betting markets. But accounts like these that do engagement farming based on polls, with absolutely no standards for sharing, also create an incentive for fraudsters who want attention.
50+1, the NYT, DDHQ etc have actual rules-based systems for deciding what to cover. They will now get updated because of new threats, but these rules are valuable and prevent more fraud than would otherwise exist out there.
Politics & Poll Tracker 📡
@PollTracker2024
.@gelliottmorris, I appreciate your work but your polling aggregator 50 + 1 also published Wedgewood’s polls along with the NYT, VoteHub and 270toWin.
These outlets should be held to a higher standard than some random polling account.
G Elliott Morris
@gelliottmorris
re: polling, these accts played an integral role in spreading the fake wedgewood "polls," and earlier the obviously fake stuff from "median strategies". there is no reason to follow these accounts—we have @50plus1news and other aggregation sources that are better and often faster
James
@jameslineky
“News aggregator” accounts like this that post half-quotes without context for engagement are gonna wreak havoc in the 2028 primaries
-
G Elliott Morris
@gelliottmorris
Well, the difference between FiftyPlusOne and the social media polling aggregator accounts like this one is that FiftyPlusOne is the organization uncovering the problems here!
And our standards have prevented other fraudsters from being platformed at 50+1 who were highly circulated by the social media aggregators. For example, the fake 'social experiment' polls from "Median Strategies" published earlier this year never appeared on 50+1 or e.g. the New York Times, but moved like wildfire on here because of now-deleted tweets from this account..
Yes, standards will have to get higher because of AI and the rising incentives for fraud from betting markets. But accounts like these that do engagement farming based on polls, with absolutely no standards for sharing, also create an incentive for fraudsters who want attention.
50+1, the NYT, DDHQ etc have actual rules-based systems for deciding what to cover. They will now get updated because of new threats, but these rules are valuable and prevent more fraud than would otherwise exist out there.
Politics & Poll Tracker 📡
@PollTracker2024
.@gelliottmorris, I appreciate your work but your polling aggregator 50 + 1 also published Wedgewood’s polls along with the NYT, VoteHub and 270toWin.
These outlets should be held to a higher standard than some random polling account.
G Elliott Morris
@gelliottmorris
re: polling, these accts played an integral role in spreading the fake wedgewood "polls," and earlier the obviously fake stuff from "median strategies". there is no reason to follow these accounts—we have @50plus1news and other aggregation sources that are better and often faster
James
@jameslineky
“News aggregator” accounts like this that post half-quotes without context for engagement are gonna wreak havoc in the 2028 primaries
G Elliott Morris
@gelliottmorris
this account deleted its tweets about median to cover this up, bu ti found this with a quick google.
-
G Elliott Morris
@gelliottmorris
Correct take
Jake Rubinstein
@Jlrube
There are way too many otherwise knowledgeable people on this site asking questions like "how does polling recover".
Pollsters didn't do anything wrong! A bunch of people did fraud and got caught by pollsters! -
G Elliott Morris
@gelliottmorris
talking about fakers open.substack.com/live-stream/421014
-
G Elliott Morris
@gelliottmorris
I'm not totally thrilled about @50plus1news having to be the poll police this week. But somebody has to do it, and our investigations have results. We have uncovered two admitted or likely fraudsters this week, and caught another manipulating its respondent-level data in August.
G Elliott Morris
@gelliottmorris
Our investigation into allegations of data fabrication at the firm SoCal Strategies, and an alleged association with the admittedly fraudulent firm Wedgewood Polls, has resulted in the pollster wiping itself from the face of the earth.
Full story at 50+1: blog.fiftyplusone.news/p/more-pollster-funny-business-unc…
FiftyPlusOne.news
@50plus1news
NEWS: The website and social media accounts for pollster SoCal Strategies have been taken offline after FiftyPlusOne reached out to the firm Tuesday to verify data quality and potential ties to Wedgewood Polls, which admitted to fabricating data this week. blog.fiftyplusone.news/p/more-pollster-funny-business-unc…
G Elliott Morris
@gelliottmorris
Basically, the FiftyPlusOne team now has extensive enough sourcing and the data chops that if you're doing bad shit with polls, we are going to find out about it.
-
G Elliott Morris reposted
Scott Clement
@sfcpoll
Anonymity is a common thread in 2026 polling fraud. All three anonymous "pollsters" have disappeared under scrutiny. www.washingtonpost.com/politics/2026/10/22/pollwatcher-an…
-
G Elliott Morris
@gelliottmorris
yeah, this is an example of how people who talk about politics for a living map their own conceptualization of ideology onto all voters. this is problematic and leads to all sorts of bad takes and strategy www.gelliottmorris.com/p/the-strategists-fallacy
Robert Griffin
@rp_griffin
Good thread but I don't find this one convincing.
Decent argument to be made that ideology isn't measuring the same thing across racial/ethnic groups. If the claim is that a candidate like Buttigieg would shift black conservatives voting behvaior signficantly, I'm skeptical. -
G Elliott Morris
@gelliottmorris
Our investigation into allegations of data fabrication at the firm SoCal Strategies, and an alleged association with the admittedly fraudulent firm Wedgewood Polls, has resulted in the pollster wiping itself from the face of the earth.
Full story at 50+1: blog.fiftyplusone.news/p/more-pollster-funny-business-unc…
FiftyPlusOne.news
@50plus1news
NEWS: The website and social media accounts for pollster SoCal Strategies have been taken offline after FiftyPlusOne reached out to the firm Tuesday to verify data quality and potential ties to Wedgewood Polls, which admitted to fabricating data this week. blog.fiftyplusone.news/p/more-pollster-funny-business-unc…
-
G Elliott Morris
@gelliottmorris
Not sure why you would try to push expectations up for the GOP right now
-
G Elliott Morris
@gelliottmorris
Strength In Numbers is covered here www.thebulwark.com/p/a-guide-to-surviving-election-season
-
G Elliott Morris
@gelliottmorris
Here’s a piece from me out this AM for MS NOW as part of their partnership with FiftyPlusOne. Trump’s net approval rating is the worst ever for a president heading into a midterm, and the GOP has lost its advantage on handling the economy and immigration
www.ms.now/opinion/trump-polls-unpopular-midterms-undecid… -
G Elliott Morris
@gelliottmorris
Last wk, Strength In Numbers polled people who previously said they voted for Trump in 2024 but will no longer cop to it. Most are now committed anti-Trump Democratic voters. This "recall error" could be biasing polls against Democrats (by <1 point)
www.gelliottmorris.com/p/2026-10-22-trump-voters-rewritin… -
G Elliott Morris
@gelliottmorris
great point by max here
Max
@maxtmcc
In 2018, Republicans managed to have a decent night in the Senate largely by just winning Trump approvers. And if the same thing happened this November, they’d get crushed.
-
G Elliott Morris
@gelliottmorris
FiftyPlusOne's investigation into "Wedgewood polls" also covered here
www.theguardian.com/us-news/2026/oct/21/polling-fabricate…
and here
www.pbs.org/newshour/politics/a-company-says-its-polls-we… -
G Elliott Morris
@gelliottmorris
re: polling, these accts played an integral role in spreading the fake wedgewood "polls," and earlier the obviously fake stuff from "median strategies". there is no reason to follow these accounts—we have @50plus1news and other aggregation sources that are better and often faster
James
@jameslineky
“News aggregator” accounts like this that post half-quotes without context for engagement are gonna wreak havoc in the 2028 primaries
-
G Elliott Morris
@gelliottmorris
This basic pattern shows up for POTUS elections too. Gelman, Merlin and I wrote about this in 2020
Owen Winter
@OwenWntr
In *every* election for *both* House and Senate since 1998, polls have overestimated Democrats by more in red states than blue states (and vice versa)
G Elliott Morris
@gelliottmorris
In terms of whether this predict polling error, what really matters is the intercept, which measures the average uniform miss across states. This is pretty random from year to year. Also note the deviation in individual state misses from the trend — that means uncertainty
-
G Elliott Morris reposted
Chris Warshaw
@cwarshaw
If the NYT/Siena poll is correct and the generic ballot ends up D+11, this would be the largest midterm victory (for either party) since the Watergate era. It also implies that Democrats would probably win over 250 seats.
www.nytimes.com/2026/10/21/upshot/times-siena-poll-trump-… -
G Elliott Morris
@gelliottmorris
First Brendan Carr, now Mike Johnson. Our story has completely broken containment (in this case by people trying to spin this for other purposes)
Mike Johnson
@MikeJohnson
I’ve been sounding the alarms that the polls are WRONG—and new evidence of that is now piling up!
Polling organizations are being forced to admit they are dramatically under-sampling Republican support, but the media is delighted to use it all to depress GOP turnout. We can’t let that happen!
Today, the Wedgewood scandal—where they’ve admitted it was “far too easy” to produce and spread completely fake polling data—adds an even darker element with enormous implications.
**IGNORE THE POLLS, and GO VOTE! 🇺🇸🇺🇸 We have 13 days to SAVE AMERICA!** -
G Elliott Morris reposted
Jonathan Robinson
@jon_m_rob
The latest NYT poll shows how much the BIG Dem leads are built upon gains among younger cohorts. And this is shift in two-way support, translating to ~nearly 40-point and 20-point changes in margin (two-way margin, but still) for under 30 and 30-44.
Trip Gabriel
@tripgabriel
In final NYT poll, Ds lead nationally by 11 pts when LV's are asked about their specific congressional candidates
"Eleven is a big number. No party has won the national vote by such a large margin since the Democrats did so in 2008," writes @Nate_Cohn
www.nytimes.com/2026/10/21/upshot/times-siena-poll-trump-… -
G Elliott Morris reposted
Dan Rosenheck
@DanRosenheck
If Rogers beats AES and Dems finish with 50 seats, we will never be able to disprove the theory that one fabricated primary poll cost Dems the Senate.
Mallory McMorrow
@MalloryMcMorrow
Just before this poll hit, the Mitchell poll that also showed me in single digits hit. That poll was bullshit - and our team proved it. It was an open SurveyMonkey link that people were able to send around to friends and take multiple times. We expended a lot of capital (and took a serious risk) - calling it out as bullshit. MIRS ultimately declined to publish it because of shoddy methodology.
But then this one hit immediately after. It was a one-two punch. Then all of the sudden it was a pattern - and people believed it had to be true.
It killed us.
These were some of the absolute worst days on the campaign. Especially for my team. There are no words to explain the morale blow, and nothing I could do or say to my team to fix it. Just an excruciating, nauseating blow.
By that point, we took a risk and we spend our money. We put our ads up. They were working. But the damage was done. And I was drowned out by outside spending and the narrative that we had completely collapsed. -
G Elliott Morris reposted
Meg Schwenzfeier
@_Schwenzfeier
Kudos to FiftyPlusOne for catching this. It's going to be a big problem for Democrats in the 2028 primary -- huge incentives for bad actors to manufacture momentum or move betting markets.
We need to align on a way to prove that a poll talked to real voters.
G Elliott Morris
@gelliottmorris
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollsterWedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
Coverage of FiftyPlusOne investigation: “New Entrant Pollster Was Publishing All Fake Polls”
talkingpointsmemo.com/edblog/new-entrant-pollster-was-pub… -
G Elliott Morris reposted
Scott Clement
@sfcpoll
Big scoop for @50plus1news
G Elliott Morris
@gelliottmorris
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollsterWedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
Updated the Trump approval and party trust trends with our new Strength In Numbers/ @verasight_data poll. All-time lows for Trump on prices, deportations — pretty much everything. Plus, Democrats now lead on immigration, foreign policy, and the economy
www.gelliottmorris.com/p/2026-10-21-october-strength-in-n… -
G Elliott Morris
@gelliottmorris
New NYT and Strength In Numbers polls this morning showing roughly same thing regarding national environment, around D+11-12 lead for Dems in the generic ballot depending on population. Double digit margin now a 50-50 chance in my modeling of all polls
www.gelliottmorris.com/p/2026-10-21-october-strength-in-n… -
G Elliott Morris
@gelliottmorris
New Strength In Number/Verasight poll conducted Oct. 13–19, 2026:
- Generic ballot: D+12 LV, D+10 RV
- Trump approval: 34 / 62
- Trust to handle most important problem: D+13
- Dems overtake Rs on trust to handle immigration, economy
More (free to read): www.gelliottmorris.com/p/2026-10-21-october-strength-in-n… -
G Elliott Morris
@gelliottmorris
The NYT editors' note on the Wedgewood Polls fiasco mentions FiftyPlusOne and links back to our investigation. We are happy our reporting on this bad actor is being recognized, and I applaud the reporting team, including Mary Radcliffe Cooper Burton, on their hard work!!
-
Post not available here, view on XPost not available here, view on X
G Elliott Morris
@gelliottmorris
your argument would go a lot farther if there hadn't been a million other polls published before Wedgewood's that showed the same thing. for what it's worth, here are my messages with a person on these servers where the Wedgewood team coordinated their polls.
At the very least, both of you are wrong about this being a product of the 'left wing advocacy ecosystem' or whatever. Wedgewood was not rigging polls to discredit your personal ideological agendas -
G Elliott Morris reposted
Ryan O'Donnell
@ryanodonnellpa
always very telling when people rush to amplify the explanation they most want to be true
-
G Elliott Morris reposted
Andrew Gelman et al.
@StatModeling
Real political reporters use data journalism techniques to catch fakers at “Wedgewood Polls”
statmodeling.stat.columbia.edu/2026/10/20/legit-political… -
G Elliott Morris
@gelliottmorris
Let's not get carried away with our priors here. Nobody of significance in the "left-wing advocacy community" was paying for "polls" by some high school and college kids with a claude subscription. And the "data" side of Wedgewood was being run by a Republican
Post not available here, view on X
G Elliott Morris
@gelliottmorris
it's because many dem operatives have to punch left to be taken seriously by corporate donors
-
Jeremiah Johnson 🌐
@JeremiahDJohns
So the conspiratorial take that the left wing was inventing fake polls to help their preferred primary candidates turned out to just be true
G Elliott Morris
@gelliottmorris
For what it's worth, one of the guys involved (in fact, the only one we have been able to confirm actually exists) is an active political operative in Florida and well known in online right-wing spaces
-
G Elliott Morris reposted
Spooky Charlotte Swasey
@charlotteeffect
Major props to the folks at Fifty Plus One for doing this digging, this is what you want responsible aggregators to look like
G Elliott Morris
@gelliottmorris
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollsterWedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
FWIW, we have updated all the averages at 50+1, and you can assess the impact for yourself by comparing the estimates from yesterday to today (plus or minus some movement from new polls that came out).
Impact is ~0 bc we down-weighted Wedgewood to hell
fiftyplusone.news/averages
Preston Hill
@realPrestonHill
Just a few examples of Wedgewood Polls holding significant weight in Silver Bulletin's polling averages:
-
G Elliott Morris
@gelliottmorris
'it was a prank bro'
noah
@nowalch
this francesca hong ad cites a poll that we now know to be fake as the “ny times”
i think its actually the only poll that ever found her with a general lead. -
G Elliott Morris
@gelliottmorris
If 10 more people subscribe to 50+1 because of this, we can recoup the cost of the actual polling data we bought to prove that Wedgewood was faking its polls (see chart). (If not, it was a worthwhile expense!)
blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollster -
G Elliott Morris
@gelliottmorris
yeah, definitely a social experiment
-
G Elliott Morris reposted
World's Biggest Stagflation Truther
@corncobanalysis
Yeah, it's a bit hard to stop someone getting into an polling aggregator if they appear to be connected to polling and solicited crowdfunding. While we need to be careful, couldn't have been caught without @50plus1news's experiment.
blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollster
G Elliott Morris
@gelliottmorris
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollsterWedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
and a pretty banger chart too
Joe
@electionsjoe
Conducting your own poll of Mississippi using the same platform to figure out if the data was fabricated or not is honestly such an awesome way to prove that a poll is fake. Absolutely genius idea.
G Elliott Morris
@gelliottmorris
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollsterWedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post. -
G Elliott Morris reposted
Joe
@electionsjoe
Conducting your own poll of Mississippi using the same platform to figure out if the data was fabricated or not is honestly such an awesome way to prove that a poll is fake. Absolutely genius idea.
G Elliott Morris
@gelliottmorris
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollsterWedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post. -
G Elliott Morris reposted
Taniel
@Taniel
.@50plus1news had evidently conducted extensive data analysis and reporting to establish that Wedgewood was fraud. *And* it appears from their article that Wedgewood tried to protect themselves and their lies until they couldn’t (making their statement today unconvincing.)
G Elliott Morris
@gelliottmorris
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollsterWedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
We were investigating Wedgewood for faking their data and emailed them for comment this morning, ahead of publishing the findings tomorrow. They are trying to cover up their fraud with the "we were doing a social experiment" excuse.
Full story here: blog.fiftyplusone.news/p/fiftyplusone-just-caught-a-pollsterWedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post.
G Elliott Morris
@gelliottmorris
They DM'd us this 2 minutes before publishing their official statement
-
G Elliott Morris
@gelliottmorris
they're spinning. story to come
Wedgewood Polls
The polling industry needs major reform. Our statement below:
This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
New Strength In Numbers/Verasight poll out tomorrow, final of the cycle. gelliottmorris.com/welcome
-
G Elliott Morris
@gelliottmorris
Every time Donald Trump says “pretend like I’m on the ballot,” a Republican loses their seat in Congress fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
North Carolina early voting update: Democrats currently make up the highest-ever share of early ballots cast at this point in the EV window. But will that translate to a lead on Election Day? I look at the historical EV data and voter rolls to find out
www.gelliottmorris.com/p/2026-10-20-north-carolina-early-… -
G Elliott Morris
@gelliottmorris
I wouldn't be that surprised at a double-digit Dem generic ballot result at this point.
Do I think it's the median outcome? Not currently, though polls could trend that direction very quickly as low-info voters tune in.
But error happens
fiftyplusone.news/polls/generic-ballot/generic-ballot -
G Elliott Morris
@gelliottmorris
Holy moly, look at the trend here (from the latest NBC poll)
Would rather see (Dems, Reps) control Congress...
Feb. '25: 54% Dems - 41% Reps
Sept. '25: 53 - 42
Oct. '26: 59 - 35 -
G Elliott Morris
@gelliottmorris
indeed www.gelliottmorris.com/p/2026-09-15-could-polls-be-undere…
Politics & Poll Tracker 📡
@PollTracker2024
Nate Cohn: Why Midterm Polls May Not Be Inflating Democrats
Are the midterm polls overestimating the Democrats?
It’s hard not to wonder. We’ve seen the polls inflate Democrats’ chances many times, but we’ve never seen Democratic candidates doing this well in red states during the Trump era. Recent polls have shown them leading in enough states to take the Senate.
Let’s state it plainly: It is absolutely possible that the polls could be overestimating Democrats. Polls are inherently imprecise, subject to many sources of error, and they could be wrong in much the same way they’ve been before.
Even so, there is no reason to assume that the polls will systematically overstate the Democrats’ standing as they did in the 2016 and 2020 presidential elections.
In fact, the polls haven’t really done so in midterm or off-year elections during the Trump era, including the 2018 or 2022 midterms. Just last November, the polls actually underestimated Democrats in New Jersey and Virginia — and by a lot.
www.nytimes.com/2026/10/19/upshot/polling-midterm-electio… -
G Elliott Morris
@gelliottmorris
VA early voting update:
- 2024 early voters at this point broke 55% D to 40% R, according to L2’s party model..
- 2026 EV so far is 60% D to 36%.
- Among 2024 early voters who are modeled D, about 44% have voted early for 2026. Among R, just 38% have.
www.gelliottmorris.com/p/2026-early-vote-blog -
G Elliott Morris
@gelliottmorris
New MRP estimates from the Strength In Numbers/Wave Polling and Research poll find more Republicans say they're financially worse off than better off in red states with close Senate races. These voters are 3x as likely as other Rs to vote Dem in Nov.
www.gelliottmorris.com/p/2026-10-18-sunday-roundup -
G Elliott Morris
@gelliottmorris
Subscribe to Strength In Numbers and get fresh insights about public opinion four months ahead of the herd www.gelliottmorris.com/p/2026-06-16-personal-mip-lead-dem…
-
G Elliott Morris
@gelliottmorris
Important post from my colleagues at FiftyPlusOne: Pollsters are publishing fewer polls in fewer races this year, which could increase uncertainty in polling averages and election forecasts:
blog.fiftyplusone.news/p/where-are-all-the-polls -
G Elliott Morris
@gelliottmorris
you should subscribe to strength in numbers gelliottmorris.com/subscribe/
Josh Kraushaar
@JoshKraushaar
“Internal GOP polling shows that support for Republican candidates cratered after the party’s midterm convention in Dallas, with voters saying the party was not attuned to their economic concerns, according to a Republican strategist briefed on the numbers”
-
G Elliott Morris
@gelliottmorris
Here's what the Senate map will look like if polls are exactly correct, versus what Dems could get with a historically normal shift in the race over the next 2.5 weeks + polling error favoring either Reps or Dems:
www.gelliottmorris.com/p/2026-10-16-senate-polling-miss-b… -
G Elliott Morris
@gelliottmorris
The polls are very close in half a dozen key states. A small uniform miss could have a big impact on Senate control. Democrats are a normal polling error away from 54+ Senate seats — or just 50
www.gelliottmorris.com/p/2026-10-15-podcast-normal-pollin… -
G Elliott Morris
@gelliottmorris
Iowa had its first day of early voting yesterday. Submitted ballots were D+27 by party registration.
In 2022, first-day ABEV ballots were D+11.
And in 2024, it was D+7.
More on the EV blog:
www.gelliottmorris.com/p/2026-early-vote-blog -
G Elliott Morris
@gelliottmorris
Interviewed about the polls and how to read them here. Some rules:
1. Average across surveys
2. Mostly good ones
3. Expect small errors
4. Don’t report on betting market prices as independent source of data. (Better to not report on them at all)
www.cjr.org/analysis/what-to-do-when-polls-are-weird.php -
G Elliott Morris
@gelliottmorris
This really is the rub. Mike Johnson is on Fox this morning saying the polls are all wrong and that the GOP is going to keep the House. But if you follow the money the people around him clearly disagree
Shane Goldmacher
@ShaneGoldmacher
TRIAGE NEWS: The Congressional Leadership Fund and Trump's super PACs are cutting off their first GOP incumbents in NJ-7 and IA-1 as the party races to avoid more widespread losses in final three weeks.
Also giving up OH-1 and MI-8.
New at NYT —> www.nytimes.com/2026/10/13/sports/top-gop-groups-pull-sup… -
G Elliott Morris
@gelliottmorris
I wrote this up a while ago, but Gallup's "who's best on your top issue" question has a very good predictive track record, especially recently www.gelliottmorris.com/p/2026-06-16-personal-mip-lead-dem…
Will Jordan
@williamjordann
Gallup rolls out another batch of indicators suggesting this midterm might have more in common with 2006 than 2018.
Better on top issue
06: D+15
18: D+1
26: D+11
Keeping US prosperous
06: D+17
18: R+5 (!)
26: D+14
Keeping US safe
06: R+2
18: R+6
26: D+4
G Elliott Morris
@gelliottmorris
This D+11 number from Gallup is also *right on* the Strength In Numbers/Verasight result from September, and we re-ask this question each month
www.gelliottmorris.com/i/217002614/democrats-lead-on-10-o… -
G Elliott Morris
@gelliottmorris
The torrent of money being spent on Ken Paxton's behalf in Texas has yet to move the needle, at least according to the polls fiftyplusone.news/polls/senate/general/texas
-
G Elliott Morris
@gelliottmorris
nice work
Owen Winter
@OwenWntr
Voters who live near data centres are swinging hard towards the Democrats, in our polling with YouGov www.economist.com/graphic-detail/2026/10/12/data-centres-…
-
G Elliott Morris
@gelliottmorris
1. I would be comparing ABEV totals (mail + early) across years for this reason.
2. Has anyone asked in a recent large-N poll vote by method intent with a like comparison to 2022 or 2024? Like means same poll. Lookin 4 support for this thesis that Dems are voting early in person
Nathaniel Rakich
@baseballot
I’m always skeptical of AV/EV analysis, but there’s extra reason to be cautious this year: Trump’s proposed mail-voting restrictions might be encouraging more Democrats to vote in person.
petebray
@petebray
Michigan keeps tilting toward Trump country. Those counties now have 47.5% of mail returns, vs. 45.2% at this point in ‘24. Already above their Election Day shares in both ’22 and ’24, with 22 days still to go. Encouraging numbers for the GOP.
G Elliott Morris
@gelliottmorris
Best thing here IMO would be a file-matched survey with VBM on it from 2024 and 2022. Then look at changes in behavior by party, not tabs of VBM raw
-
G Elliott Morris
@gelliottmorris
Posted an update about early voting in Pennsylvania to the Strength In Numbers early voting blog.
www.gelliottmorris.com/p/2026-early-vote-blog -
G Elliott Morris
@gelliottmorris
yeah.... can we get a margin of error on these precinct estimates please?
Owen Winter
@OwenWntr
find this kind of thing a bit absurd, to be honest. Polling is too uncertain to reliably tell you the statewide winner in Michigan and you're telling me you can forecast the result in a precinct of 500 people? come off it
-
G Elliott Morris
@gelliottmorris
Just published: early voting live blog, Strength In Numbers style.
I use voter-file data to approximate changing turnout dynamics by party compared to past elections.
In VA, GOP 2024 early voters are voting early in 2026 at a rate 11 points below Dems.
www.gelliottmorris.com/p/2026-early-vote-blog -
G Elliott Morris
@gelliottmorris
Are Democrats only doing well because Trump is unpopular? That's undoubtedly part of their advantage. But in shift from 2025, the party org and its candidates are now also viewed more favorably than the GOP's. www.gelliottmorris.com/p/2026-10-11-sunday-roundup
-
G Elliott Morris
@gelliottmorris
Charlie Cook says there's no "blue wave" but a "red undertow." I think he's wrong about this. The Dem Party and candidates are currently more popular across the board than the Republican Party and candidates. And Ds are in better shape than 2018
www.gelliottmorris.com/p/2026-10-11-sunday-roundup -
G Elliott Morris
@gelliottmorris
Update on redistricting with some numbers from Strength In Numbers here www.newyorker.com/news/the-political-scene/who-won-the-re…
-
G Elliott Morris
@gelliottmorris
I am personally not sure what Trump's approval rating floor is, but it's definitely lower than pundits were predicting in early 2025. It is a possibility that he never gets above the mid-30s again (among all adults)
fiftyplusone.news/polls/approval/president -
G Elliott Morris
@gelliottmorris
good point (though potential confounder is hard-to-poll people are the ones Rs are trying to reach with ads)
Brent Buchanan, Pollster
@brentbuc
Despite holding a massive campaign cash advantage, only 24% of voters report seeing more Republican ads, while over 70% see more Democratic ads or an equal split.
Ad volume does not equal ad recall when the message fails to stick. A cash advantage is useless if ad buys focus on media weight instead of voter resonance.
Ad efficiency is an emotional connection challenge, not a media-buy problem...
www.politico.com/live-updates/2026/10/08/congress/gop-ads… -
G Elliott Morris
@gelliottmorris
Perhaps relevant to current handicapping: In 2018, voters who decided who to vote for at the very last minute were the most Democratic group in the electorate
-
G Elliott Morris
@gelliottmorris
the nerds have taken over ms now
Mary Radcliffe
@marywitha4
We started this company 1 year ago with no money, just a belief that people care about nonpartisan, high quality data and analysis. This partnership with @MSNOWNews will help bring our data to a whole new audience. Thank you all for following along on the journey!
FiftyPlusOne.news
@50plus1news
Exciting update: FiftyPlusOne is partnering with @MSNOWNews for polling and election analysis ahead of the midterms! See below for more details.
blog.fiftyplusone.news/p/fiftyplusone-to-power-ms-nows-mi… -
G Elliott Morris
@gelliottmorris
New from me for MS NOW (part of deal with 50+1): Republicans have a bad candidate problem. Democratic candidates are viewed more favorably than GOP in all key Senate races, sometimes with a net gap (fav - unfav) up to 20 points higher (Paxton v Talarico)
www.ms.now/opinion/ken-paxton-gop-texas-polling-candidate… -
G Elliott Morris reposted
FiftyPlusOne.news
@50plus1news
Exciting update: FiftyPlusOne is partnering with @MSNOWNews for polling and election analysis ahead of the midterms! See below for more details.
blog.fiftyplusone.news/p/fiftyplusone-to-power-ms-nows-mi… -
G Elliott Morris
@gelliottmorris
FiftyPlusOne is proud to announce a new deal with MS NOW to provide polling averages and interactive data journalism for digital and broadcast platforms!!
blog.fiftyplusone.news/p/fiftyplusone-to-power-ms-nows-mi… -
G Elliott Morris
@gelliottmorris
👀👀fiftyplusone.news/polls/governor/general/ohio
Aaron Blake
@AaronBlake
Including this new CNN poll, the last three quality polls in the Ohio governor's race have Ramaswamy trailing by 4, 6 and 8 points.
-
G Elliott Morris reposted
Alex Roarty
@Alex_Roarty
This is a strange metric to cite because Democrats in many states have turned around their registration losses this year, after years and years of seeing their numbers decline. It’s fair to point out the bigger picture of how many registered voters they’ve lost in the last decade or so — I did so here www.washingtonsun.com/2026-election/nrcc-analysis-voters-… — but at the moment the registration story is a positive one for Dems
L A R R Y
@LarryOConnor
Democrats are LOSING the voter registration war in every key battleground.
As political pros will tell you, these numbers are much more important than public polls. -
G Elliott Morris
@gelliottmorris
i would like to see the error bars on those regression coefficients
Drew Savicki
@DrewSav
There is zero truth to this. To suggest that reputable pollsters are making up their results is absurd. Doing that would destroy their business
-
G Elliott Morris
@gelliottmorris
building a simulator that takes our forecasts and looks at what odds would be if polls are as biased as they were in 2016/20 or 2018/22
-
G Elliott Morris
@gelliottmorris
People are now accepting this as fact, and there is very limited evidence of this. Pundits just want bonus points for calling things early. Pollsters stratifying their samples by party registration and using inverse propensity sampling (which helps a lot here!!) have the same toplines as Chaz
Ryan Grim
@ryangrim
Bad news for Dems in Iowa and Kansas that rural Trump supporters are just refusing to answer pollsters. If they do vote, it'll make these polls off by a decisive amount.
G Elliott Morris
@gelliottmorris
The interpretation of the state navigate data should be that rural Republicans had a low response rate to **their poll**. That does not mean that they have a low response rate to **all polls**, especially those that use sampling techniques to combat this exact problem.
G Elliott Morris
@gelliottmorris
Anyway, the reaction to this tweet from Chaz is a good example of how this site has devolved into spreading pure uncut copium for Republicans. When you’re down 9 points in the generic ballot, I guess denial is the best medicine
-
G Elliott Morris
@gelliottmorris
what did you think we meant by anti-incumbency bias? vibes? papers? essays? Dem state executives are underperforming in many blue states
Nick Reisman
@NickReisman
Marist poll this Wednesday morning:
Hochul 53%
Blakeman 42%
maristpoll.marist.edu/polls/midterm-elections-in-new-york… -
G Elliott Morris
@gelliottmorris
FWIW, I weight the polls I do for clients by the share of the vote Trump won in respondents’ local zip code (so sample is balanced by quintile of local Trump vote), and this moves natl numbers by <1point after weighting by past vote. Could be larger at state level
Matthew Zeitlin
@MattZeitlin
i'm sorry but i'm starting to buy the "polls aren't reaching rural voters" republican cope. you can believe adam hamilton is a fantastic candidate and el-sayed is a meh one, but D+3 in KS and D+1 in MI.......
-
G Elliott Morris reposted
Will Jordan
@williamjordann
It's possible to have good theories about many of these branches, but multiple can be at play! In 2022 and 2024, sampling/measurement errors on these pathways largely balanced out enough to preserve it as a useful tool, but there are probably always exceptions. It's hard.
Will Jordan
@williamjordann
Take classic "missing Trump voters" example. Whether this leads a recall-weighted poll to be biased to Dems, Reps or neither, depends on several other questions, starting w/ is it primarily:
* Non response by Trump-Rs
* High response by Harris-Rs
* False recall? -
G Elliott Morris
@gelliottmorris
misc. poll nerdery here
FiftyPlusOne.news
@50plus1news
A dispatch from the Methods Desk: we're no longer accepting campaign internals conducted without a professional pollster. Find out why:
blog.fiftyplusone.news/p/campaigns-need-to-hire-real-poll… -
G Elliott Morris
@gelliottmorris
dc insiders find any other lens through which to see american politics challenge. rating: impossible
Adam Carlson
@admcrlsn
Can we wait until we actually have results — and WAR scores from Split Ticket and the like to control for other variables — before we start talking about who is underperforming relative to fundamentals?
The 2028 ideological posturing can wait another 30 days, I promise. -
G Elliott Morris
@gelliottmorris
this is great
Matt Grossmann
@mattgrossmann
You can model your own state legislative results here with a variety of scenarios from @ejfagan :
claude.ai/code/artifact/f635abd3-d5bc-496a-85f3-c544ce29311b -
G Elliott Morris
@gelliottmorris
unfortunately you do gotta capitalize on that election season referral traffic
Ben Yelin
@byelin
Sadly the reveal here about which party this helps is behind the paywall.
-
G Elliott Morris
@gelliottmorris
New post at Strength In Numbers today with a fresh take on whether the polls are overstating support for Democrats or Republicans.
I use a measure of respondent engagement to proxy non-response and simulate potential bias. Error could be quite large
www.gelliottmorris.com/p/2026-10-06-survey-effort-nonresp… -
G Elliott Morris
@gelliottmorris
tomorrow at strength in numbers: what a new measure of in-survey respondent effort (that any pollster can use!) might tell us about potential poll bias ahead of November . get subscribed www.gelliottmorris.com/p/2026-10-06-survey-effort-nonresp…
-
G Elliott Morris
@gelliottmorris
The polls could be off, but you shouldn't necessarily expect them to be off in any particular direction. Historically speaking, "unskew"ing has made forecasts worse, not better. + unlike 2016-2024 polls in the Midwest basically agree with the fundamentals www.gelliottmorris.com/p/2026-10-04-sunday-roundup-polls-…
-
G Elliott Morris
@gelliottmorris
So about that 2024 realignment?
www.gelliottmorris.com/p/trumps-winning-2024-coalition-has
Sahil Kapur
@sahilkapur
Latino voters — NBC News/Telemundo poll
Trump job approval rating
• 27% approve
• 69% disapprove
Trump issue approval
20% on cost of living
25% on economy
30% on immigration
38% on border security
Generic ballot: D+24
Fav/unfav
DEM: 41/42
GOP: 27/62
www.nbcnews.com/politics/2026-election/poll-latino-voters… -
G Elliott Morris reposted
Ruth Igielnik
@RuthIgielnik
Our Time/Siena poll shows Talarico with a 6-pt lead in Texas. Notably, in 2018, our poll showed Beto down by a similar margin in the state. www.nytimes.com/2026/10/03/us/senate-polls-texas-kansas-o…
-
G Elliott Morris
@gelliottmorris
All of the pundits saying so confidently that polls are overestimating Democrats fail to mention that the polling error in 2025 in VA and NJ was twice as large as the average state poll bias in 2016 and 2024. Nobody knows direction of error ahead of time!
www.gelliottmorris.com/p/2026-10-04-sunday-roundup-polls-… -
G Elliott Morris
@gelliottmorris
New at Strength In Numbers this morning: Despite claims from some pundits and Republican strategists of biased and "fake" polling data, the polls and fundamentals basically agree on the blue wave: www.gelliottmorris.com/p/2026-10-04-sunday-roundup-polls-…
-
G Elliott Morris
@gelliottmorris
insane that he keeps doubling down on this, and on the conditional in particular, which polls terribly www.gelliottmorris.com/p/2026-09-27-sunday-roundup
Donald J. Trump
@realDonaldTrump
If Republicans win the House of Representatives and the Senate in the 2026 Midterm Elections, I’m going to give all adult citizens in the United States of America, $5,000! Thank you for your attention to this matter, and I look forward to signing those checks!
This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
Short read on Senate polls and "fundamentals" coming out at Strength In Numbers later this AM. Many are surprised by how blue state polls look. And they could be wrong! But they really aren't that surprising in the context of other non-polling predictors. www.gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
After 2024, a lot of people wrote about how low-turnout voters were conservative now and supported Trump's agenda.
This was a dramatic misreading of the election, and was easy to call ahead of time
www.gelliottmorris.com/p/trumps-big-weakness-is-the-economy
www.gelliottmorris.com/p/2026-09-11-low-turnout-maga-no
www.gelliottmorris.com/p/2026-07-15-pew-non-voters
Nate Cohn
@Nate_Cohn
that would be right in 2024, but not anymore, as democrats are very strong with low propensity voters in our polls
24 primary: trump+2 recall, peltola+5
no 24 primary: trump+16 recall, peltola+9 -
G Elliott Morris
@gelliottmorris
ballot measure polling here!
FiftyPlusOne.news
@50plus1news
New Trendline: Where do voters stand on the most consequential ballot measures this year? We dig in to all the latest polling on taxes, voting, transgender issues, and more.
blog.fiftyplusone.news/p/where-do-voters-stand-on-this-years -
G Elliott Morris
@gelliottmorris
state navigate is doing some really cool stuff with live polling
Chaz Nuttycombe
@ChazNuttycombe
UPDATE.
The Vermont Governor's race looks like it's a Toss-Up.
We're using our Virginia method to weight by party ID, and right now we have a raw and LV sample that's Harris +41, Scott +61. Really...odd, Harris won VT by 31.5%, Scott won re-election by 51.6%! Winner's bias? -
G Elliott Morris
@gelliottmorris
NEW: Amid rising diesel prices, Republicans turn against Trump for first time on handling inflation. A majority of Republicans and Republican-leaning independents (52%) now say they disapprove of Donald Trump’s handling of prices:
www.gelliottmorris.com/p/2026-10-02-republicans-sour-on-t… -
G Elliott Morris
@gelliottmorris
New Kansas Senate polling average up at 50+1 (not sure I believe these polls FWIW, but the data's the data) fiftyplusone.news/polls/senate/general/kansas
-
G Elliott Morris
@gelliottmorris
Subscribe to Strength In Numbers today for unparalleled Value Above Replacement
www.gelliottmorris.com/subscribe/ -
G Elliott Morris
@gelliottmorris
Strength In Numbers called this eight months ago
www.gelliottmorris.com/p/meet-americas-new-swing-voter-the
(from The Bulwark's latest Focus Group pod) -
G Elliott Morris
@gelliottmorris
Starting to think it’s part of CNN’s social media strategy to keep doing this extraordinarily dumb take (“90% of people who tell us they like to eat apples also enjoy apple cider”) as a way to get the president’s attention and thus more clicks. Harry knows better
-
G Elliott Morris
@gelliottmorris
We have a new high poll for Dems AFAIK of D+16 on the generic ballot among likely voters. I think pollsters are playing kind of fast and loose with these filters to only use decided voters but still. Holy hell
G Elliott Morris
@gelliottmorris
And seconds after I tweeted this, YouGov dropped at D+15
d3nkl3psvxxpe9.cloudfront.net/documents/econTabReport_CCp…,
G Elliott Morris
@gelliottmorris
These are outliers and you should read a polling average like ours at fiftyplusone.news/polls/generic-ballot/generic-ballot
-
G Elliott Morris
@gelliottmorris
Tomorrow, Strength In Numbers will release the latest findings from our September survey with Verasight on voters' worries about ICE being deployed to the polls, and other "election security" measures proposed by the president. Get it in your inbox:
www.gelliottmorris.com/welcome/ -
G Elliott Morris
@gelliottmorris
Shot, chaser. This is why you subscribe to Strength In Numbers, folks!! www.gelliottmorris.com/welcome
-
G Elliott Morris
@gelliottmorris
I can't wait to get my 'goodies' from the government
-
G Elliott Morris
@gelliottmorris
MAGA male Trump voters are seeking gender-affirming care goodauthority.substack.com/p/republicans-fascination-with…
-
G Elliott Morris
@gelliottmorris
I have built a novel dataset of news headlines that get shared by the press that updates daily and stretches back to 2016.
Looking at the last month, Trump's first-ever midterm convention failed to change the news agenda. Iran, ICE, Epstein lead 2026.
www.gelliottmorris.com/p/2026-09-28-convention-headlines-… -
G Elliott Morris
@gelliottmorris
i am not sure an ad that says "paid for by the U.S. government" is going to have the effect the president wants it to have...
-
G Elliott Morris
@gelliottmorris
Donald Trump's promise to send $5,000 checks to every adult if the Republicans win the midterms has backfired horrifically.
www.gelliottmorris.com/p/2026-09-27-sunday-roundup -
G Elliott Morris
@gelliottmorris
We have found one thing Donald Trump is more popular than: Flock cameras open.substack.com/pub/fiftyplusone/p/flock-cameras-are-ev…
-
G Elliott Morris
@gelliottmorris
Our new Strength In Numbers/ @Verasight_data poll: 58% say the U.S. is worse off under Trump vs Biden, 24% say it’s better
www.gelliottmorris.com/p/2026-09-25-58-percent-country-wo… -
G Elliott Morris
@gelliottmorris
No president has ever gone into a midterm election as unpopular as Trump is today. Not even George W. Bush in 2006. Trump's net approval is twice as low as it was at this point in 2018. blog.fiftyplusone.news/p/no-president-has-gone-into-a-mid…
-
G Elliott Morris
@gelliottmorris
Begging pollsters to just stop for a single hour releasing new surveys with the president hitting new lows so I can finalize these graphs for an article
-
G Elliott Morris
@gelliottmorris
Democratic incumbents underperforming in an otherwise very Democratic year would be consistent with elevated levels of anti-incumbent bias driven by anxiety over excess prices and just generally negative political conditions
Nick Riccardi
@NickRiccardi
NY polling at D+9 in what polls are saying is a D+9 year nationally should be a flashing red light to Dems
-
G Elliott Morris
@gelliottmorris
This chart uses the latest data from fiftyplusone.news
Trump’s approval is significantly below where it was at this point in 2018
G Elliott Morris
@gelliottmorris
This reflects a new All Time Low for Trump set this morning. (This has the new SIN/Verasight poll and will get even worse for him after we add the CNN and Marist numbers)
-
G Elliott Morris
@gelliottmorris
76% wrong priorities numbers is catastrophic. Have we ever seen this number this high?
Aaron Blake
@AaronBlake
The folly of focusing on ballrooms, arches, reflecting pools and the Kennedy Center (oh, and Iran).
(via new @CNN poll) -
G Elliott Morris reposted
Charles Monaco
@charlesmonaco
Great convention team
G Elliott Morris
@gelliottmorris
Trump's approval rating has hit a new low of 33% in our new Strength In Numbers/ @verasight_data poll just out this AM. POTUS's disapproval is 63%, for a net of -30.
Democrats lead by 10 points among likely voters on the generic ballot
Read more:
www.gelliottmorris.com/p/2026-09-23-september-strength-in… -
G Elliott Morris
@gelliottmorris
Trump's approval rating has hit a new low of 33% in our new Strength In Numbers/ @verasight_data poll just out this AM. POTUS's disapproval is 63%, for a net of -30.
Democrats lead by 10 points among likely voters on the generic ballot
Read more:
www.gelliottmorris.com/p/2026-09-23-september-strength-in… -
G Elliott Morris
@gelliottmorris
sorry didn't have time to post earlier when this happened but: fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
The way one person explained it to me, the handicappers aren’t trying to forecast outcomes ahead of time so much as “skate to where the puck is” on Election Day (by delaying judgment on likelihood in competitive races). That is of… debatable utility
www.gelliottmorris.com/p/how-to-update-your-beliefs-with-…
G Elliott Morris
@gelliottmorris
If you use the term “toss-up” to describe an event with an 80-20 Dem likelihood, then… well I think you should find a new term www.gelliottmorris.com/p/2026-09-18-race-raters-lagging-m…
-
G Elliott Morris
@gelliottmorris
It's polling season! How should you update your beliefs about the election in the face of new — often contradictory — data?
Here's a step-by-step example of how we do exactly that at Strength In Numbers:
www.gelliottmorris.com/p/how-to-update-your-beliefs-with-new -
G Elliott Morris
@gelliottmorris
featuring this great insight
FiftyPlusOne.news
@50plus1news
If it feels like every poll you’ve seen this year came from a D campaign and showed a toss-up, you’re not imagining things. D internals make up an all-time high share of internals this year. But is this real support, or just fuel for candidate fundraising?
blog.fiftyplusone.news/p/why-does-every-poll-seem-to-show -
G Elliott Morris reposted
FiftyPlusOne.news
@50plus1news
If it feels like every poll you’ve seen this year came from a D campaign and showed a toss-up, you’re not imagining things. D internals make up an all-time high share of internals this year. But is this real support, or just fuel for candidate fundraising?
blog.fiftyplusone.news/p/why-does-every-poll-seem-to-show -
G Elliott Morris
@gelliottmorris
The Democrat running in Montana could very easily cost Democrats the Senate www.gelliottmorris.com/p/2026-07-28-exclusive-independent…
Adam Carlson
@admcrlsn
-
G Elliott Morris reposted
Dan Rosenheck
@DanRosenheck
Democrats hit a new record high of 59% in our Senate forecast today and reached a lofty 93% chance of flipping the House—the highest figure since the brief gap b/w Virginia's later-annulled redistricting and Florida's. Our story explains what's changed 1/6 www.economist.com/graphic-detail/2026/09/17/high-petrol-p…
-
G Elliott Morris
@gelliottmorris
New analysis: Expert handicappers like the Cook Political Report are probably underestimating Democratic odds of winning the House by about 30 points. That's based on an analysis of their historical ratings, not polls or other models.
www.gelliottmorris.com/p/2026-09-18-race-raters-lagging-m…
G Elliott Morris
@gelliottmorris
Historically the out-party wins about 70% of races Cook calls toss-up at 45 days before the election, and 20% of races they say lean toward the party in the White House.
-
G Elliott Morris
@gelliottmorris
This is great viz from Adam and co. and an important, though not new, finding. Most Americans are not ideological and there are neglected dimensions of voter psychology that can be leveraged to win elections
- www.gelliottmorris.com/p/not-just-left-vs-right-most-voters
- www.gelliottmorris.com/p/the-strategists-fallacy
-
Adam Jentleson
@AJentleson
At @SearchlightInst we talk a lot about heterodoxy.
Now, partnering with @nytimes, we can show you exactly what we mean.
“In reality, people often have heterodox views that defy strict partisanship.”
Take a minute to read this—it might change how you think about politics. -
G Elliott Morris
@gelliottmorris
anti-system axis www.gelliottmorris.com/p/meet-americas-new-swing-voter-the
Amy Walter
@amyewalter
Voters competitive CD’s polled for @CookPolitical see election not just as a referendum on Trump (who’s underwater by 16pts), but as a repudiation of the entire political system. Indies prefer candidate who will “reform entire system” vs serve as a “check on Trump”by 39pts.
-
G Elliott Morris
@gelliottmorris
Good get for CBS!
Zachary Donnini
@ZacharyDonnini
We at VoteHub are very excited to partner with the CBS News Data Desk for coverage of the 2026 midterms.
It’s going to be a long election season, Election Night, and likely election week, and we’ll be there for every minute of it. -
G Elliott Morris
@gelliottmorris
Two things. 1) When will the people who argued in April 2025 that Democrats should stay silent on immigration own up to being frankly just disastrously wrong about the politics of mass deportations? Their punditry ran against the polling data and basically bordered on enablement.
Shane Goldmacher
@ShaneGoldmacher
NEW: Rep. Maria Elvira Salazar (R-FL) just rolled out a straight to camera ad breaking with Trump.
"Mr. President, some of your immigration enforcement efforts have gone too far....The same Hispanics who helped you get to the White House in 2024 feel betrayed today"
G Elliott Morris
@gelliottmorris
2) Maria Elvira Salazar is talking tough on immigration now, after voting for ICE funding bills, because internals show a tied race in FL-27
-
G Elliott Morris
@gelliottmorris
race raters are starting to move
Kyle Kondik
@kkondik
Crystal Ball: 8 House Rating Changes in Favor of Dems
Most notable: Kean (R, NJ-7) Toss-up to Leans D, IA-2 Open and Bill Huizenga (R, MI-4) Leans R to Toss-up
New math: 210 at least Leans D, 203 at least Leans R, 22 Toss-ups
centerforpolitics.org/crystalball/seven-weeks-to-go-polli…
Full list: -
G Elliott Morris
@gelliottmorris
There's one survey methodology decision that could be causing polls to *UNDER*estimate Democrats this year, in a reversal from recent patterns www.gelliottmorris.com/p/2026-09-15-could-polls-be-undere…
-
G Elliott Morris
@gelliottmorris
it's time for people to come around to the idea that this question isn't asking what they think it is www.tandfonline.com/doi/full/10.1080/17457289.2013.811245
also you have to wrangle with information; lots of voters don't know the GOP's "position" on data centers, which is quickly evolving against the public's
Derek Thompson
@DKThomp
Americans are concerned about AI, hate data centers, and hate Trump, who has loudly embraced the industry over and over again.
So, naturally, the electorate is ...
<squints> <squints harder>
... R+2 on artificial intelligence -
G Elliott Morris reposted
Tom Bonier
@tbonier
An important point here - this isn't, for the most part, pointing to the GOP redraws being dummymanders. The takeaway should be that the Dems are poised to do so well in November that they will overcome GOP gerrymanders in many districts. But the gerrymandering will still limit Dem gains relative to pre-gerrymandered districts. And, absent something to even the playing field, this will make it harder for Dems in future elections when a wave isn't at their backs.
Decision Desk HQ
@DecisionDeskHQ
7 of the 9 states that redrew their U.S. House maps this cycle are using Republican-drawn lines.
But our 2026 forecast has Democrats making an average net gain of about 2 seats across those 9 states.
Redistricting isn't shaping up to be the majority-saver the GOP hoped for. -
G Elliott Morris reposted
David Wallace-Wells
@dwallacewells
The two largest and most conspicuous drops in Trump issue approval are both on inflation and the cost of living. The first, right after inauguration, reflects tariffs and the trade war. The second, beginning this spring, reflects the war in Iran.
-
G Elliott Morris
@gelliottmorris
As we said when we launched our election forecast: Yes, it's looking like a blue wave. The data are even stronger for Democrats than in 2018. Glad to see everyone finally catching up to us blog.fiftyplusone.news/i/209533131/1-democrats-are-even-b…
FiftyPlusOne.news
@50plus1news
According to our election model at fiftyplusone.news, control of the Senate now leans toward Democrats. The party has been buoyed by new polls showing it ahead in TX, IA, AK, OH & ME, and has a 64% chance to control the chamber after the midterms
fiftyplusone.news/forecast/senate -
G Elliott Morris
@gelliottmorris
Have had a few requests to update the Trump approval graphics at www.gelliottmorris.com/p/data . Ask and you shall receive! (Our long-term plan is to adopt these with custom charts that live update at FiftyPlusOne.)
-
G Elliott Morris
@gelliottmorris
Trump is taking the 10% side of a 90-10 issue on AI/data centers. Talked about this on the podcast two weeks ago www.gelliottmorris.com/p/donald-trump-is-acting-like-he-w…
-
G Elliott Morris reposted
Roose
@RuNoseP
if the 2026 elections end up being a blue tsunami, the explanation will be pretty simple
-
G Elliott Morris
@gelliottmorris
Democrats have done a great job at recruiting candidates this cycle. In both House and Senate races, named candidates are out-polling the generic ballot www.gelliottmorris.com/p/2026-09-13-sunday-roundup
-
G Elliott Morris
@gelliottmorris
Democrats hit 50% on average in House generic ballot polls for the first time today, per our average at fiftyplusone.news
-
G Elliott Morris
@gelliottmorris
Trafalgar is at D +2 in Iowa folks
Mary Radcliffe
@marywitha4
We’ve definitely seen a shift this week. We’ve gone from D-aligned pollsters posting great D results in reach districts to R-aligned pollsters showing MOE races in reach Senate seats. This is looking like an absolute wipe out. (Opinion subject to change based on new information.)
-
G Elliott Morris
@gelliottmorris
iPhone fold is cool but you can buy an iPhone 17e and iPad mini for less than half the price (my setup). but I guess if you want to scroll tik tok and instagram reels at the same time then fold??
-
G Elliott Morris
@gelliottmorris
Was starting to think our MRP was too negative on Trump. But this result from YouGov validates our registered voter numbers almost exactly
wavepolling.com/pro/maps
Ronald Brownstein
@RonBrownstein
A 57% disapproval for Trump in Iowa continues his pattern of substantial majority disapproval in key Senate states that he carried by double digits in 2024, including TX, AK, and OH-not to mention ME (60% in @YouGovAmerica) and NC (around 55% in two polls yesterday).
-
G Elliott Morris reposted
Owen Winter
@OwenWntr
We have a data-heavy package in this week's @TheEconomist on the state-of-play ahead of the midterms. Donald Trump has tied himself to inflation-boosting policies that swing voters hate. His historic unpopularity could sink the Republican Party: www.economist.com/interactive/united-states/2026/09/10/do…
-
G Elliott Morris
@gelliottmorris
Over the last week we have seen several polls w Ds leading by 10 nationally and ahead in seats Trump won by 15+ points. Starting to get hard to deny that Nov could plausibly be a bloodbath for the GOP.
Our forecast has Dems above 60% to win the Senate
fiftyplusone.news/forecast/senate -
G Elliott Morris
@gelliottmorris
We have updated our House election forecast with the new old congressional district boundaries in Missouri. fpo.news
-
G Elliott Morris
@gelliottmorris
Based on 12,000 survey interviews, low-turnout-likelihood Republicans are comparatively anti-Trump, contrary to most analysis and punditry about the group after the 2024 election. Less engaged voters aren’t MAGA, but anti-status-quo
www.gelliottmorris.com/p/2026-09-11-low-turnout-maga-no -
G Elliott Morris
@gelliottmorris
Apparently every newspaper in the world has decided to just run the president’s midterm campaign for him
-
G Elliott Morris
@gelliottmorris
Here’s the data on why the Republicans’ “Trumpapalooza” convention will backfire on the GOP this fall:
1) POTUS is at 35% approval
2) Low-turnout-likelihood 2024 Trump voters are NOT die-hard MAGA.
www.gelliottmorris.com/p/2026-09-09-trumpapalooza-gop-hurt -
G Elliott Morris
@gelliottmorris
Here’s my report on the utility of “Trumpapalooza,” using polling data from over 11,000 interviews this year.
Low-turnout Republicans are not die-hard Trump supporters sitting on the sidelines. He doesn’t understand this, and has scores a big own-goal
www.gelliottmorris.com/p/2026-09-09-trumpapalooza-gop-hurt -
G Elliott Morris
@gelliottmorris
Bit of a wonky post, but here I share a lot about how I’m thinking about polls and polling error these day
www.gelliottmorris.com/p/2026-09-15-all-polls-are-models-now -
G Elliott Morris reposted
Data for Progress
@DataProgress
We tested @SenSanders and @RepCasar's bill to pause AI development and permanently ban the creation of AI superintelligence.
It's supported by 68% of voters — including 72% of Democrats, 70% of Independents, and 63% of Republicans. -
G Elliott Morris
@gelliottmorris
Republicans are fear-mongering about “Chinese AI” and “going backwards” because they want the government to stay out of AI.
This is bad because AI could kill all of us. And that’s not me saying that, that’s the people working AT the AI companies!
David Shor
@davidshor
I think government mandated independent oversight of the labs is a good idea with broad bipartisan public support and that failure to act quickly could have grave irreversible consequences
-
G Elliott Morris
@gelliottmorris
Feels like people are getting out over the skis with this poll.
Cycle on cycle, Jewish voters are 8 points more Dem now, according to this survey.
The shift among RVs in the generic ballot overall is, drum roll… D+8!
Drew Savicki
@DrewSav
Another cycle, the rightward swing among Jews continues. This is why it's important to have an ideologically diverse coalitions that includes people like Jared Moskowitz, Josh Gottheimer, and Tom Suozzi all of whom represent districts with large rightward trending Jewish populations.
-
G Elliott Morris
@gelliottmorris
The convention ain’t gonna help
Jonathan Martin
@jmart
Man alive, new @CNN poll of MAINE has 46% of all voters saying their Senate vote is aimed at "Sending a message that you oppose Donald Trump"
www.cnn.com/2026/09/09/politics/cnn-polls-maine-senate-mi… -
G Elliott Morris
@gelliottmorris
Here is a screenshot from a new CNN/SSRS poll in Michigan.
It shows that Democratic voters are 14 percentage points more likely than Republicans to say they are extremely/very motivated to vote in November, 87% vs 73%.
s3.documentcloud.org/documents/28608784/cnn-poll-conducte… -
G Elliott Morris
@gelliottmorris
GOP strategists live in an alternative reality (on Twitter/X) where Trump's approval rating among his 2024 voters is 100%. In our polling (and Pew's), it's closer to 75%. Last wk Navigator came out showing 1/5 of 2024 Trump voters regret voting for him! www.gelliottmorris.com/p/poll
-
G Elliott Morris
@gelliottmorris
We are adding some new charts to the FiftyPlusOne forecast page. These charts are behind a partial paywall to incentivize people to pay to support our aggregation and forecasting work. First up is a pop vote - seats scatter. Line charts soon fiftyplusone.news/forecast/house
FiftyPlusOne.news
@50plus1news
We’re adding a new scatter chart to our forecast for subscribers only that shows how the national popular vote in the 2026 midterm election translates to seats in the House and Senate. fiftyplusone.news/forecast/house
-
G Elliott Morris
@gelliottmorris
hey @VoteHub honest question, how is your House forecast model at 76% with Kalshi odds integrated but 77% without, when the Kalshi market is 85% (and Polymarket 89)
-
G Elliott Morris
@gelliottmorris
New YouGov/Economist poll out this AM has Democrats up 11 points on the House vote among likely voters. This is with named candidates, not the usual "generic ballot," which in my testing with CES data is typically more accurate in forecasting outcomes
d3nkl3psvxxpe9.cloudfront.net/documents/econTabReport_Slc… -
G Elliott Morris
@gelliottmorris
Missed this: Trump approval now solidly in the 35s, as the GOP launches its "Trumpapalooza" midterm convention in Dallas. What a stupid idea fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
My best guess after a lot of on-the-ground research is that so many political types, esp reporters, spend so much time on Twitter/X because they aren’t terribly interesting people and crave likes from other people to convince themselves they are interesting
G Elliott Morris
@gelliottmorris
Also it’s a lot easier for pundits to stare at the feeds and come up with reactionary “takes” than it is to do the hard work of interviewing people or gathering original data, partly because editors are overworked and goals from the biz side are warped
-
G Elliott Morris
@gelliottmorris
Good article, just out, from the Times. Mentions the public as a thermostat + cyclical nature of party ID. (Stops short of saying the current environment was easy to predict, though that’s true & some of us did!).
Covered this on the podcast on Thurs: www.gelliottmorris.com/p/donald-trump-is-acting-like-he-w…
G Elliott Morris
@gelliottmorris
If you want to support independent, data-driven coverage of politics and elections that has a track record of being ahead of the curve, become a paying subscriber of Strength In Numbers today: www.gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
voters literally only want one thing and it's disgusting www.gelliottmorris.com/p/2026-09-04-trump-wrong-priorities
-
G Elliott Morris
@gelliottmorris
Here's another incidence of a pollster (who we already knew was shady) allegedly getting caught passing off surveys done for a partisan campaign client as nonpartisan public work
www.theguardian.com/politics/2026/sep/04/investigation-po… -
G Elliott Morris
@gelliottmorris
New study: High prices (nominal level in excess of 10-yr trend) explains “vibecession” in the Netherlands, while media coverage does not. Replicates our findings in the U.S. (Line Go Up theory undefeated) esb.nu/hoog-prijspeil-verklaart-pessimisme-onder-consumen…
-
G Elliott Morris
@gelliottmorris
This guy is a total joke. Was spreading conspiracies about dead people voting for Biden in Michigan in 2020, IIRC. Nobody should take him seriously. Even if he’s right that I’m right youtu.be/NinVQap86iA?t=1077
-
G Elliott Morris
@gelliottmorris
the real Trump Derangement Syndrom was thinking his agenda was ever popular
-
G Elliott Morris
@gelliottmorris
Yeah the joke is bad but running against tariffs in Michigan, where approval of Trump on trade is -20 net, seems like a smart strategy
wavepolling.com/tracker/
Jonathan Martin
@jmart
A non-zero percent chance Schumer and/or his Super Pac ask AES to spend rest of cycle deep in Mackinac woods (at the lodge Grand has) while they handle messaging
CSPAN
@cspan
Michigan Sen. candidate @AbdulElSayed on the Trump-Canada trade war: "So if you're starting a trade war with Canada is cause you're the asshole."
G Elliott Morris
@gelliottmorris
Like if you think Chuck Schumer, the most unpopular Democratic politician in the country, has some magical powers in messaging to swing voters in Michigan… well I disagree www.gelliottmorris.com/p/2026-06-29-most-popular-pols-par…
-
G Elliott Morris
@gelliottmorris
Given the current moment in history I think it is weird for Third Way to be lovingly quoting Stephen Miller and ceding all rhetorical ground on immigration and deportations to the far right. Seems like they actually want Republicans to win?
Sarah Pierce
@SarahPierceEsq
"Immigration reform should mean something else entirely: it should mean improvements to our laws and policies to make life better for American citizens."
I'd love if Dems said this. As much as I hate to admit it, Stephen Miller wrote those words.
More: www.thirdway.org/memo/a-decade-ago-trump-set-an-immigrati… -
G Elliott Morris
@gelliottmorris
Trendline of Collins overperformance is also significantly negative, from +40 to +12 in a decade. She has never faced a real Dem challenge in a Dem midterm year. People are just using bad heuristics for her because the polls were off with Gideon
Adam Carlson
@admcrlsn
I don’t think enough people realize that the Susan Collins of 2026 isn’t the Susan Collins of the 1990s - 2010s.
She voted to confirm all but one current Trump cabinet nominee (Hegseth), including RFK Jr. — despite knowing he was deeply unqualified for the position.
And by doing so, she put the health of Mainers — including children, right as they are about to go back to school — at risk.
Either she knew better and did it anyway to appease Trump, or her judgment is not what it once was.
And as a result we have:
— Preventable disease outbreaks (like measles) after he removed all 17 members of the CDC's Advisory Committee on Immunization Practices and replaced them with his own people, and made changes to childhood vaccine recommendations (not to mention undercutting faith in vaccines writ large, which has an impact on herd immunity)
— Seemingly daily terrifying food recalls after laying off about 3,500 FDA employees — including scientists, inspectors & lab personnel looking for potential contaminations — even after concerns were raised at the time that it could impair inspections, lab work, outbreak investigations & safety communications
Susan Collins owns every decision that he makes. -
G Elliott Morris
@gelliottmorris
we need a complete and total shutdown of the harrisx poll until we can figure out what’s going on
Annie Andrews Democrat
@GorillaHpn
Never forget, HarrisX had the VA governor’s race at D+3. Actual result: D+15.
-
G Elliott Morris
@gelliottmorris
“Would you vote for candidate A, who loves America, or candidate B, who kicks puppies for fun?”
InteractivePolls
@IAPolls2022
Should Democrats embrace or reject people like Hasan Piker, who said America “deserved 9/11”?
❌ Reject: 77%
✅ Embrace: 23%
Reject/Embrace
GOP: 79-21
Dem: 72-28
——
Should Republicans embrace reject people like Candace Owens, who said “the Jews are behind everything”?
❌ Reject: 75%
✅ Embrace: 22%
Reject/Embrace
GOP: 73-27
Dem: 76-24
Harvard/Harris | 8/28-30
harvardharrispoll.com -
G Elliott Morris
@gelliottmorris
looks about right— for a poll from Arkansas
-
G Elliott Morris
@gelliottmorris
New study from me and the Verasight team on “AI-moderated” online survey interviews. A lot here but the upshot is these tools offer depth the way a focus group does, but a lot of people drop off and in our testing this created some significant biases
www.linkedin.com/posts/benjamin-leffb071_ai-moderated-int… -
G Elliott Morris
@gelliottmorris
This (from Tom Wood) suggests more than just the typical thermostatic backlash is happening right now. In sixteen months, Trump and Republicans have turned the GOP’s largest party ID advantage this century into a tie for its biggest deficit
-
G Elliott Morris
@gelliottmorris
From our August poll: Here's what voters really mean when they say the Democrats should "fight" more, in their own words
www.gelliottmorris.com/p/2026-09-01-what-democrats-really… -
G Elliott Morris
@gelliottmorris
donald trump is actively trying to help as many democratic senators as possible get elected
-
G Elliott Morris
@gelliottmorris
Amid Iran fallout, inflation anxiety, immigration policy moving dramatically farther to the right than the public sanctioned, & all the rest of what has happened in the last 19 months, a new poll shows 1/6Trump voters says they wouldn't vote for him again.
www.gelliottmorris.com/p/2026-08-30-sunday-roundup -
G Elliott Morris reposted
Wave Polling and Research
@wave_polling
Our new poll finds Democrats leading the generic ballot by 7 points among registered voters. Just 32% of voters approve of how Donald Trump is handling the economy. Dems at +10 to handle
wavepolling.com/results/2026-08-17-economic-costs-deporta… -
G Elliott Morris reposted
Rajiv Sethi
@rajivatbarnard
Great conversation with @gelliottmorris on prediction markets and elections, covered a lot of interesting ground:
www.gelliottmorris.com/p/everything-you-wanted-to-know-about -
G Elliott Morris
@gelliottmorris
In this piece, I load up the last 15 months of Strength In Numbers/Verasight polling to answer two big questions:
1. Are respondents too progressive?
2. Is weighting by past vote increasing support for Republicans?
**In our data,** the answer to both is no -
G Elliott Morris
@gelliottmorris
IMO people are underestimating the chance of a double-digit Dem margin in Nov. Modal outcome? No, but certainly 20-30%+ chance. Our LV generic ballot is D+8-9, and out-party tends to gain 1-2 points from Sept to Nov
www.gelliottmorris.com/p/2026-08-25-august-strength-in-nu… -
G Elliott Morris reposted
Cooper Burton
@cooperrburton
There's been a lot of coverage of fake polls and big polling misses lately, which has led to some premature obituaries for the polling industry. The reality is that the polls are actually alright! We ran the numbers if you don't believe me ⬇️
G Elliott Morris
@gelliottmorris
The biannual freakout about the polls has arrived exactly on schedule.
No, polls don't look any worse than usual
No, there is not an atypical number of new firms
Our advice is as always: keep calm and average across firms
blog.fiftyplusone.news/p/no-the-polling-industry-is-not-on -
G Elliott Morris
@gelliottmorris
The biannual freakout about the polls has arrived exactly on schedule.
No, polls don't look any worse than usual
No, there is not an atypical number of new firms
Our advice is as always: keep calm and average across firms
blog.fiftyplusone.news/p/no-the-polling-industry-is-not-on -
G Elliott Morris
@gelliottmorris
New Strength In Numbers/ @verasight_data poll: New low for Trump on prices and foreign policy, and he is significantly net negative on handling the border.
The gap between POTUS's strong disapproval and strong approval is now nearly 3-to-1 (50% vs 17%)
www.gelliottmorris.com/p/2026-08-25-august-strength-in-nu…
G Elliott Morris
@gelliottmorris
topline and crosstabs are now up at the data portal
www.gelliottmorris.com/p/poll -
G Elliott Morris
@gelliottmorris
is that bad?
Aaron Blake
@AaronBlake
How bad things are for Trump:
He's now managed to go 5 points underwater on "border security" -- this despite border-crossings dropping to record lows on his watch.
(via @gelliottmorris) -
G Elliott Morris
@gelliottmorris
this guy went off the deep end a long time ago but this is next level
-
G Elliott Morris
@gelliottmorris
Aug. Strength In Numbers/@verasight_data poll coming out tomorrow 7:00 ET.
In it:
— Trump loses his last positive issue
— Dems lead the generic ballot for a 15th straight poll, with bigger lead among likely voters
— And: subscriber questions on savings, gov program cuts -
G Elliott Morris
@gelliottmorris
hot take but if you're gonna do polling in the us you should try to spell the state names correctly
Raphael Nishimura
@rnishimura
Acabei de receber esse anúncio no Instagram depois de ficar literalmente 1 minuto por lá.
Aliás, eu moro em Michigan, não faço ideia de onde fica "MichiNgan" 😂 -
G Elliott Morris
@gelliottmorris
now i'm no public relations expert but it seems possible to me that an organization could oppose the blockade of cuba without literally celebrating fidel castro -- and therefore telling that the official "democratic" socialists of america account has chosen this path
-
G Elliott Morris
@gelliottmorris
New at FiftyPlusOne: Rural and suburban voters haven’t looked this Democratic since 2018
blog.fiftyplusone.news/p/rural-and-suburban-voters-havent… -
G Elliott Morris reposted
Adam Carlson
@admcrlsn
Here are the average ratings for U.S. Senate races as of today — based on ratings from The Cook Political Report, Inside Elections, Sabato’s Crystal Ball, Split Ticket, VoteHub & FiftyPlusOne.
At the moment, Alaska is the tipping point race for a Democratic majority. -
G Elliott Morris
@gelliottmorris
Not buying this.
www.kark.com/news/your-local-election-hq/latest-poll-show…
My best guess right now is that polls w/o partisanship controls are at least 5-10 points too friendly to Ds -
G Elliott Morris
@gelliottmorris
Democrats are doing very well in early polls of this year’s midterm Senate elections. But are pollsters just overestimating Democrats again, like they have the last 4 cycles?
Maybe. But predictions of poll bias from lagged performance usually fail.
www.gelliottmorris.com/p/2026-08-21-predicting-poll-bias-… -
G Elliott Morris reposted
Cooper Burton
@cooperrburton
NEW: We're suspending two pollsters from @50plus1news after discovering evidence that suggests both firms conducted polls for political campaigns without disclosing their clients publicly. More below 👇
-
G Elliott Morris
@gelliottmorris
You may have heard of a "pollster" that recently admitted to fabricating data for a "social experiment".
We have discovered that another firm, TPSI, falsified data to produce results more favorable to a candidate it did undisclosed partisan work for
blog.fiftyplusone.news/suspending-two-pollsters-data-fals…
G Elliott Morris
@gelliottmorris
Another firm, Patriot Polling, also appears to have conducted surveys for a Republican candidate for governor and then passed them off as public, non-partisan polls.
FiftyPlusOne will no longer use any data from TPSI or Patriot Polling, or the individuals involved with these firms -
G Elliott Morris
@gelliottmorris
Publishing another big story on potential data fabrication by a prolific election pollster tomorrow at FiftyPlusOne.
Subscribe: blog.fiftyplusone.news/welcome -
G Elliott Morris
@gelliottmorris
Matthew Yglesias is the perfect example of the epistemic humility problem in DC. He knows a little bit about everything, which drives a lot of overconfidence about politics and voter psychology. E.g., he totally wiffed on backlash to Trump on immigration www.gelliottmorris.com/p/kilmar-abrego-garcia-immigration…
Matthew Yglesias
@mattyglesias
Learned helplessness.
G Elliott Morris
@gelliottmorris
i have never met someone who is so unwilling to realize they were wrong about something they very very publicly fucked up on
-
G Elliott Morris
@gelliottmorris
We are taking MRP mainstream in the U.S. Check us out
Wave Polling and Research
@wave_polling
Introducing Wave Polling, a new political data firm specializing in model-based methods for tracking opinion across time and geographies.
Here’s our MRP of the generic House vote. Dems are tied or above 50% in 26 states including 6/8 Senate battlegrounds
wavepolling.com/dashboard/ -
G Elliott Morris
@gelliottmorris
Yes. This is best understood as the paper of record giving free public relations services to an ideological ally, not analysis rooted in political science or data. It is frankly akin to wishcasting.
Michael A. Cohen (NOT TRUMP’S FORMER FIXER)
@speechboy71
The argument in this piece -- that Democrats can potentially win a 60-seat supermajority in the Senate -- is so divorced from reality it's embarrassing. The idea that "heterodoxy" or winning on policy can overcome partisanship is, for lack of a better word, a scam. It's absolute nonsense, and I'm genuinely amazed that anyone who has worked in politics actually believes it.
There is one way for Democrats to pass their policy agenda -- abolish the filibuster, add blue-leaning states, expand the Supreme Court, end partisan gerrymandering, increase the size of the House. That's it. Aggressive, ruthless, unsentimental politics that challenges the structural lean of national politics, which has favored Republicans for decades.
Until Democrats stop believing the long-disproven notion that policy wins elections and accept that partisanship drives everything, they will continue to fail.
www.nytimes.com/2026/08/19/opinion/democrats-midterms-sup…Post not available here, view on X
G Elliott Morris
@gelliottmorris
A helpful calculation is to arrange US states by their 2024 vote margins and find the state that would provide the Democrats their 60th seat in the Senate, restricting to seats up in 2026 or 28.
It’s Kansas, implying Dems would need to win the nat pop vote by 16 points to win 60 seats to overcome the filibuster.
Or you need Democrats to become a bunch of Jerry Morans.
Neither of these scenarios are plausible.
The median effect size in the academic literature on running moderate Dems vs the median Dem is about a point on vote margin. The max coef is close to 3 points.
The solution proposed by Jentleson is a fiction. The real problem for Dems is political geography and social sorting, not policy positions by nominees. John Tester does not win MT in 2024 by becoming more conservative. The party does not morph into the Kansas GOP in any scenario
G Elliott Morris
@gelliottmorris
Dems should instead stand down all their candidates in places they aren’t going to win and let independents run instead. More on the brand problem here
www.gelliottmorris.com/p/2026-06-30-democratic-brand-issu…
And here
www.gelliottmorris.com/p/2026-07-28-exclusive-independent… -
G Elliott Morris
@gelliottmorris
Unfortunately people are taking the reaction markets too seriously
Tom Bonier
@tbonier
For the third consecutive week, a candidate trading at greater than 90% odds of winning has lost their race.
I have no issue with these markets existing, but I do have an issue with anyone who pretends they have predictive value. -
G Elliott Morris
@gelliottmorris
Tomorrow’s Strength In Numbers podcast will be a crossover ep w/ FiftyPlusOne’s Mary Radcliffe — talking through 3 recent examples of fake and lying pollsters and what they mean for aggregators/forecasters in 2026. Tune in! open.substack.com/live-stream/327136
-
G Elliott Morris
@gelliottmorris
at @50plus1news we vet every pollster before we put "data" from them on our website. we did not include any polls from "median strategies" because they did not answer our screening Qs.
this is why you should just spam refresh the FiftyPlusOne website, instead of following meme aggregators on X
Nathaniel Rakich
@baseballot
Disseminating fake polling data is Harmful and Bad, but these people certainly got their answer about how quickly it would spread.
-
G Elliott Morris
@gelliottmorris
Is Abdul El-Sayed going to lose Michigan?
Because he's a progressive?
Here's what the data actually says.
www.gelliottmorris.com/p/2026-08-18-el-sayed-michigan-wha… -
G Elliott Morris
@gelliottmorris
This is the opposite of the conclusion from my piece, which is that there were possibly too many young people in WI polls, but hard to know, and able to be weighted mostly — the bigger source of “error” seems to be late movement toward Crowley. No evidence of huge progressive non-response
Max Fisher
@Max_Fisher
2026 polls continue to massively oversample left-wing demos which should be a huge source of concern for Dems going into November but for some reason isn't www.gelliottmorris.com/p/2026-08-13-wisconsin-poll-reanal…
-
G Elliott Morris
@gelliottmorris
Very dumb discourse today. For one thing, this is a take about an unrealized future election outcome using cherry-picked data. For another, Cohn Jain et al all ignore that Haley Stevens was polling just as bad! www.gelliottmorris.com/p/2026-08-06-el-sayed-early-polls
G Elliott Morris
@gelliottmorris
All the stuff about Baldwin is also inaccurate. CES data shows voters viewed her as very progressive (and also that perceptions of ideology are unrelated to overperformance: data4democracy.substack.com/p/the-new-york-times-argues-m… )
G Elliott Morris
@gelliottmorris
Good reminder that people who write about politics for a living (and spend all day on Twitter) see the world through very colored ideological lenses, but this is not how most voters work.
www.gelliottmorris.com/p/the-strategists-fallacy -
G Elliott Morris reposted
Larry Sabato
@LarrySabato
Here’s the polling autopsy we’ve all been asking for:
G Elliott Morris
@gelliottmorris
I have reanalyzed polling microdata from two surveys in Wisconsin, generously shared with me by State Navigate, and put the results up Strength In Numbers for all to learn from. As far as I know this is the only public autopsy from Tuesday's miss www.gelliottmorris.com/p/2026-08-13-wisconsin-poll-reanal…
-
G Elliott Morris reposted
Will Jordan
@williamjordann
Smart & interesting read. This does flatter one ituition, which is that while "non-response" explanations feel sexy and have mattered a lot in GE polling error, in a primary, it is just much easier to get other very consequential stuff wrong (demographic/ideological composition)
G Elliott Morris
@gelliottmorris
I have reanalyzed polling microdata from two surveys in Wisconsin, generously shared with me by State Navigate, and put the results up Strength In Numbers for all to learn from. As far as I know this is the only public autopsy from Tuesday's miss www.gelliottmorris.com/p/2026-08-13-wisconsin-poll-reanal…
-
G Elliott Morris reposted
Jonathan Robinson
@jon_m_rob
Excellent analysis from @gelliottmorris on the dynamics of low favorability ratings for the Dem Party among Dems vs. more popular national left-wing pols). Somewhat self-evident, but younger, more progressive (largely White) D's much less fav, stark contrast to older Black voters
G Elliott Morris
@gelliottmorris
New analysis of Strength In Numbers/Verasight data today: It’s not just young Democrats who like lefty candidates.
Both young and old Democrats rate politicians like Zohran Mamdani highly. The difference is young Dems really don't like the establishment
www.gelliottmorris.com/p/2026-08-14-dem-party-divide-youn… -
G Elliott Morris reposted
Jerry Skurnik
@PrimeNewYork
Why polling primary elections is so hard, by @gelliottmorris open.substack.com/pub/gelliottmorris/p/2026-08-12-primary…
-
G Elliott Morris
@gelliottmorris
The people crave technical write-ups of survey methodology. I don't make the rules. Sorry Hunter
-
G Elliott Morris
@gelliottmorris
KILL THE VAMPIRES
www.gelliottmorris.com/p/kill-the-vampires -
G Elliott Morris
@gelliottmorris
New analysis of Strength In Numbers/Verasight data today: It’s not just young Democrats who like lefty candidates.
Both young and old Democrats rate politicians like Zohran Mamdani highly. The difference is young Dems really don't like the establishment
www.gelliottmorris.com/p/2026-08-14-dem-party-divide-youn… -
G Elliott Morris
@gelliottmorris
I have reanalyzed polling microdata from two surveys in Wisconsin, generously shared with me by State Navigate, and put the results up Strength In Numbers for all to learn from. As far as I know this is the only public autopsy from Tuesday's miss www.gelliottmorris.com/p/2026-08-13-wisconsin-poll-reanal…
G Elliott Morris
@gelliottmorris
correction: Marquette published yesterday I was just too busy coding to see it. law.marquette.edu/facultyblog/2026/08/a-look-at-2026-prim…
-
G Elliott Morris reposted
Matthew Zeitlin
@MattZeitlin
very interesting @gelliottmorris on what happened in wisconsin
-
G Elliott Morris reposted
Raphael Nishimura
@rnishimura
Quer ver como realmente se faz uma avaliação de pesquisa pré-eleitoral de forma metodologicamente séria e honesta que realmente busca entender seus potenciais problemas e limitações? Dá uma olhada nessa análise do G Elliott Morris de duas pesquisas das primárias em Wisconsin! +
G Elliott Morris
@gelliottmorris
I have reanalyzed polling microdata from two surveys in Wisconsin, generously shared with me by State Navigate, and put the results up Strength In Numbers for all to learn from. As far as I know this is the only public autopsy from Tuesday's miss www.gelliottmorris.com/p/2026-08-13-wisconsin-poll-reanal…
-
G Elliott Morris reposted
Jon Green
@_Jon_Green
really great stuff here from @gelliottmorris (and +1 to @ChazNuttycombe for sharing data) on diagnosing the WI polling miss. some evidence for all of a) older electorate than modeled, b) late movement toward Crowley, and c) non-response bias www.gelliottmorris.com/p/2026-08-13-wisconsin-poll-reanal…
-
G Elliott Morris reposted
Chaz Nuttycombe
@ChazNuttycombe
This is the BEST-written breakdown of what happened in WI. The L2 data still has the same problem as other vendors w/ the vampires, so I don't think 18-29% was <3% (it was 5% IIRC for early voting, a vote method not preferred by voters <30), but age wasn't the main problem
G Elliott Morris
@gelliottmorris
I have reanalyzed polling microdata from two surveys in Wisconsin, generously shared with me by State Navigate, and put the results up Strength In Numbers for all to learn from. As far as I know this is the only public autopsy from Tuesday's miss www.gelliottmorris.com/p/2026-08-13-wisconsin-poll-reanal…
-
G Elliott Morris
@gelliottmorris
So what the hell is going on with primary polling?? www.gelliottmorris.com/p/2026-08-12-primary-polls
-
G Elliott Morris
@gelliottmorris
The Democratic primary for governor of Wisconsin is a really great example of why America needs to switch to ranked-choice voting in primaries. As of now it looks like the winner of the race will prevail with <40% of the vote! No good reason not to switch ahead of 2028.
-
G Elliott Morris
@gelliottmorris
Jeremy is a great interviewer. Enjoyed this conversation americaninequality.substack.com/p/decoding-public-opinion…
-
G Elliott Morris
@gelliottmorris
counterpoint (to bankhead): www.gelliottmorris.com/p/2026-07-28-exclusive-independent…
Jonathan Robinson
@jon_m_rob
Anti electability disinformation is going to cost the pro democracy center left a shot at a critical Senate seat in November
-
G Elliott Morris
@gelliottmorris
I've written a lot more about "America's new swing voter" today, showing how "anti-system orientation" is distinct from left-right ideology & driving much if not most of the gains for (1) the DSA in primaries and (2) the Democrats for the general election. www.gelliottmorris.com/p/2026-08-11-anti-system-not-just-…
G Elliott Morris
@gelliottmorris
mystery plot below the paywall
-
G Elliott Morris
@gelliottmorris
it's gonna be trump vs mamdani for nyc mayor in 2029
-
G Elliott Morris
@gelliottmorris
Looks like the conventional wisdom that El-Sayed only does well with rich white people is heavily confounded by the correlation between race and income. Per this, wealthier white ppl actually went more for Stevens! (same pattern in NYC 2025 primary, which people misinterpreted)
Zachary Donnini
@ZacharyDonnini
This type of thing is very tricky...being young and being high ed/lower income is just very closely linked so it's really hard to figure out causality. And this is just precinct data (not voter data). But definitely overall being more black and more wealthy is predictive of voting for Stevens (controlling for everything else) and being more higher ed and younger is predictive of voting for El-Sayed (controlling for everything else).
So El-Sayed's core base is White, lower income, higher ed, and young.This post is unavailable. See it on X.
G Elliott Morris
@gelliottmorris
What's happening e.g. in this Politico piece is that a lot of people in the media want to denigrate progressives as "champagne socialists," but that doesn't appear to be an accurate description, at least if you're looking at the data www.politico.com/live-updates/2026/08/04/michigan-missour….
More on NYC
G Elliott Morris
@gelliottmorris
one of the additional tells is that the MAGA crowd is going after El-Sayed for supposedly not winning among the middle class thehill.com/homenews/administration/6012719-vance-critici…
-
G Elliott Morris reposted
Make Sure You Vote In 2027
@je_mc2
I actually bought his book, Strength In Numbers, when it came out. Great book. I'd recommend.
-
G Elliott Morris
@gelliottmorris
Democrats currently lead the polls in 5 of 7 key Senate races for which we are publishing a poll average today. (Alaska hasn't had its primary yet, so we aren't publishing there, but if you count the hypotheticals with Peltola and Sullivan it's 6/8)
FiftyPlusOne.news
@50plus1news
FiftyPlusOne is very excited to release our polling averages for Senate and governor elections today! There's a new page on the website just for general election averages, and a blog post with some takeaways on the current state of the race here:
blog.fiftyplusone.news/p/general-election-polling-averages -
G Elliott Morris
@gelliottmorris
Learned a lot talking to Peter Enns for my podcast’s latest deep dive ep about his new paper "How a Seemingly Innocuous and Intuitive Methodological Choice Confused a Generation of Research on Policy Responsiveness."
Some rare good news for democracy?
www.gelliottmorris.com/p/deep-dive-episode-what-if-democr… -
G Elliott Morris
@gelliottmorris
Interview with Paul Krugman about the midterms here.
I argue people are updating their priors about November too slowly; It’s a very likely “blue wave,” with more Dem seats than 2018 possible and full Democratic control of Congress just above even odds
paulkrugman.substack.com/p/talking-with-g-elliott-morris-… -
G Elliott Morris
@gelliottmorris
People do no fully appreciate that inflation is a bigger issue to voters now than it was in 2024
open.substack.com/pub/fiftyplusone/p/affordability-concer… -
G Elliott Morris
@gelliottmorris
GOP Rep. Max MIller is so toxic now that sub/exurban Cleaveland might actually elect a guy named Poindexter fiftyplusone.news/forecast/house
-
G Elliott Morris
@gelliottmorris
this is why left-wing democrats are winning so many elections recently
www.gelliottmorris.com/p/2026-08-07-what-dividing-the-left -
G Elliott Morris reposted
Jeremy Ney
@jeremybney
@gelliottmorris and I will be discussing prices, polls, perception, and what Americans get both right and wrong about the economy. Join us on Tuesday at 4pm ET open.substack.com/live-stream/306626
-
G Elliott Morris
@gelliottmorris
I enjoyed reading this behind-the-scenes dispatch from Katie, who designed the graphics for our new election forecast that launched this week! open.substack.com/pub/fiftyplusone/p/how-501-designed-our…
-
G Elliott Morris
@gelliottmorris
Republicans will probably lose Michigan's Senate race despite El-Sayed's win.
Contrary to the Takes about the primary, the polls and election "fundamentals" do not suggest Stevens would have performed significantly better
www.gelliottmorris.com/p/2026-08-06-el-sayed-early-polls -
G Elliott Morris
@gelliottmorris
Polling error in primaries is really high. Last night's in Michigan comes in right at the historical average
"The expected error on the margin between two primary candidates in a race for governor or U.S. Senate is a whopping 13.0 points."
blog.fiftyplusone.news/p/2026-texas-senate-primary-polls -
G Elliott Morris reposted
Mary Radcliffe
@marywitha4
Maybe! But she’s been posting a 25+ point lead, which is quite different than a 13 point lead like we saw from El-Sayed. If she’s overestimated to the same degree as Abdul, she still sails to nomination.
G Elliott Morris
@gelliottmorris
francesca hong overestimated too perhaps?
-
G Elliott Morris
@gelliottmorris
francesca hong overestimated too perhaps?
-
G Elliott Morris
@gelliottmorris
One of the problems with people who spend too much time on sites like this one is they forget how to read (he deleted this tweet right)
-
G Elliott Morris
@gelliottmorris
Guessing but probably something to do with political engagement bias that correlated with vote choice that you can’t weight out of without a past vote/party benchmark (since not a general)
Matthew Zeitlin
@MattZeitlin
can't really knock AES for "underperforming" if he wins by 5 or something because he was probably never doing as well as the polls said, but why were the polls so good for him...
G Elliott Morris
@gelliottmorris
Aka why dem primary polls in 2020 showed Warren doing so well
-
G Elliott Morris reposted
Mary Radcliffe
@marywitha4
At this point, I would not be surprised by either! Polling open primaries is very hard and, as we wrote at @50plus1news earlier this year, the average error on the margin in partisan primaries is 13 points.
blog.fiftyplusone.news/p/2026-texas-senate-primary-polls
Chaz Nuttycombe
@ChazNuttycombe
I told @marywitha4 earlier today I'm holding to my pre-poll prediction of El-Sayed by double digits, and she said she thinks MI polling isn't great (e.g. 2016) and wouldn't be surprised by a Stevens win. Trust Mary folks
-
G Elliott Morris
@gelliottmorris
Looks like a 10-15 point polling miss in Michigan. And people have been writing a lot of electability takes using this data. Be careful out there folks
-
G Elliott Morris
@gelliottmorris
At current county breakdown, Stevens needs at least +20 margin out of Wayne to close the gap. Caveat emptor, that’s the range to watch for
Jordan Schwartz
@JordanSchwartz
It's likely that this race will come down to Wayne County Election Day vote.
Stevens is doing extremely well in rural Michigan relative to expectations, and a good Wayne result could allow her to win this primary.
Keep watching @VoteHub coverage throughout the night. -
G Elliott Morris
@gelliottmorris
How did these not get scrubbed before the campaign even launched? Lmao
sam
@sam_d_1995
bruh she literally just hates every single holiday lmao
-
G Elliott Morris
@gelliottmorris
natalie is right on this one
Natalie Jackson
@nataliemj10
You really can't gauge El-Sayed's or Stevens' "electability" based on cherry-picked polls. The differences are almost all within MOE.
There are other aspects of electability to argue about, but it's just not there in the polls.
My latest in @nationaljournal [link below] -
G Elliott Morris
@gelliottmorris
Feeling pretty good about this article I wrote in May: www.gelliottmorris.com/p/the-texas-senate-election-is-a-t…
fiftyplusone.news/forecast/senate
John McCormack
@McCormackJohn
Three polls in a row showing Paxton losing.
Cruz never trailed in a 2018 poll. -
G Elliott Morris
@gelliottmorris
Our average of polls at FiftyPlusOne gives Trump a roughly 36% approval rating and 61% disapproval rating — a net -25 — among all adults, and he’s rapidly losing ground. The president set a new record low six days in a row just in the last week:
www.gelliottmorris.com/p/2026-08-04-trump-squeezed-from-b… -
G Elliott Morris
@gelliottmorris
The reason Trump’s approval keeps hitting a new all-time low every day, and why Republicans might lose the Senate, is because GOP voters are losing their enthusiasm for his agenda, while independents move in droves toward an energized Democratic base.
www.gelliottmorris.com/p/2026-08-04-trump-squeezed-from-b… -
G Elliott Morris
@gelliottmorris
Cooper is the best and we're happy to have him as part of the team
FiftyPlusOne.news
@50plus1news
fiftyplusone.news is delighted to announce that for the remainder of the 2026 election cycle, the remarkable and talented Cooper Burton will be working with us! Cooper is an incredible researcher and writer, and we're so happy to have him.
-
G Elliott Morris reposted
Politics & Poll Tracker 📡
@PollTracker2024
Fifty Plus One 2026 congressional forecast
US Senate
🟦Democrats 55% (51 seats)
🟥Republicans 45% (49 seats)
——
US House
🟦Democrats 85% (230 seats)
🟥Republicans 15% (205 seats)
Link to forecast: fiftyplusone.news/forecast/senate -
G Elliott Morris reposted
InteractivePolls
@IAPolls2022
New 2026 midterm forecast from FiftyPlusOne:
HOUSE
🟦 Dem: 85% → 230 seats
🟥 GOP 15% → 205 seats
SENATE
🟦 Dem 55% → 51 seats
🟥 GOP 45% → 49 seats
Democrats are even better positioned than at this point in the 2018 blue wave.
——
Senate battlegrounds (win probability)
🟦 GA: Dem 94–6% (Safe D)
🟦 NC: Dem 84–16% (Likely D)
🟦 MI: Dem 69–31% (Lean D)
🟦 ME: Dem 68–32% (Lean D)
🟦 OH: Dem 53–47% (Toss-up)
🟦 AK: Dem 51–49% (Toss-up)
🟥 TX: GOP 52-48% (Toss-up)
🟥 IA: GOP 59-41% (Toss-up)
blog.fiftyplusone.news/p/election-forecast-democrats-stro… -
G Elliott Morris
@gelliottmorris
Follow along at @50plus1news
Larry Sabato
@LarrySabato
BIG Forecast today by Strength in Numbers and FiftyPlusOne:
Democrats strongly favored to take the House (85% chance). Senate tilts Democratic (55% chance). www.gelliottmorris.com/p/democrats-strongly-favored-to-take -
G Elliott Morris
@gelliottmorris
Get it while it’s hot
FiftyPlusOne.news
@50plus1news
Good morning poll lovers! Big drop at fiftyplusone.news today: our 2026 midterm election forecast is now live!
fiftyplusone.news/forecast/house -
G Elliott Morris
@gelliottmorris
🥁
🥁
🥁
🥁
Dems strongly favored to take the House in our new election forecast. Senate just better than a tossup.
Seat probabilities:
Iowa: 41% Dem
Texas: 48
Alaska: 51
Ohio: 53
Maine: 68
Michigan: 69
North Carolina: 84
Georgia: 94
blog.fiftyplusone.news/p/election-forecast-democrats-stro… -
G Elliott Morris
@gelliottmorris
New from Strength In Numbers: Trump’s immigration agenda and esp deportation operation is unpopular in every swing state — including Maine, where Susan Collins is trying to moderate on ICE www.gelliottmorris.com/p/even-republicans-in-congress-can
-
G Elliott Morris
@gelliottmorris
main character mode www.gelliottmorris.com/p/2026-07-28-exclusive-independent…
-
G Elliott Morris
@gelliottmorris
We are publishing more and more journalism from outside contributors at 50+1 by the day. Pitch us!
Tia Yang
@tsyang27
Latest at @50plus1news: We built a dataset tracking every primary endorsement from Trump, progressive groups, party committees, women’s groups, and pro-Israel groups. Halfway through the season, here's who's winning. 🧵blog.fiftyplusone.news/p/primary-project-2026
-
G Elliott Morris
@gelliottmorris
Yes, to be making the argument in 2026 that immigration is a Republican leaning issue so Democrats shouldn’t campaign on it, and especially Trump’s overreach on deportations, is to be completely ignorant of the trends in data and events of the last 18 months
.
@GregTSargent
This is the problem w/the idea that Ds can't go on offense on immigration. It's not okay for Collins to have a clear field on TV to sell herself as a moderate on ICE.
CNN has Trump 22 points underwater on this. Two other polls have Ds tied w/GOP on it!
newrepublic.com/article/213839/susan-collins-ice-funding-… -
G Elliott Morris reposted
Tia Yang
@tsyang27
Latest at @50plus1news: We built a dataset tracking every primary endorsement from Trump, progressive groups, party committees, women’s groups, and pro-Israel groups. Halfway through the season, here's who's winning. 🧵blog.fiftyplusone.news/p/primary-project-2026
-
G Elliott Morris
@gelliottmorris
I took a look at the link between gas prices (high) and approval of Trump’s handling of prices (very bad) over the last 6 months. The data suggests evaluations of POTUS are sticky and negative, wouldn’t return to normal even if gas prices go back below $3
www.gelliottmorris.com/p/cheap-gas-wont-save-trump-or-rep… -
G Elliott Morris
@gelliottmorris
New Strength In Numbers Data Lab post about survey weighting and MRP models
www.gelliottmorris.com/p/adventures-in-survey-weighting-mrp -
G Elliott Morris
@gelliottmorris
FiftyPlusOne.news
@50plus1news
Donald Trump's net approval has reached a new low today in our tracker. His net approval sits at -24.1. The previous low was -23.7, achieved on May 27.
fiftyplusone.news/polls/approval/president -
G Elliott Morris
@gelliottmorris
FWIW, poll out this morning from The Argument (not a progressive outlet) shows 65% of Americans support government-funded health care for all, and this holds for working-class folks.
www.theargumentmag.com/p/americans-know-who-the-real-elite
Mamdani also did best in lower-income precincts in NYC. True in the primary true if you control for race. dockish.com/analysis.html
So IMO not as simple as Neera makes it seem
Neera Tanden🌻
@neeratanden
It has been a pretty big lie people I’ve told themselves on here that working class people are closet leftists. They are in fact much more moderate than college educated folks.
-
G Elliott Morris
@gelliottmorris
Can progressives win in swing states? Maybe not Bernie Sanders, but it largely depends on what you mean by progressive. E.g.,
-
G Elliott Morris reposted
Aaron Blake
@AaronBlake
Democrats still haven't gotten Alani Bankhead to drop out to help the independent candidate (like they got in Nebraska).
Now we're seeing a poll suggesting it could majorly impact the independent's chances.
Politics & Poll Tracker 📡
@PollTracker2024
GQR/Strength in Numbers poll | 7/23-7/26 LV
US Senate Montana 2026
🟥Kurt Alme 41%
⬜️Seth Bodnar 27%
🟦Alani Bankhead 22%
🟨Kyle Austin 4%
——
⬜️Seth Bodnar 50%
🟥Kurt Alme 44%
——
⬜️Seth Bodnar 44%
🟥Kurt Alme 40%
🟨Kyle Austin 11%
Link to poll: www.gelliottmorris.com/p/2026-07-28-exclusive-independent… -
G Elliott Morris
@gelliottmorris
Here’s CNN covering our recent Strength In Numbers/ @verasight_data polling on support for voter ID laws and what groups of voters say they could/couldn’t prove their citizenship: www.gelliottmorris.com/p/2026-07-24-save-act-proof-of-cit…
-
G Elliott Morris
@gelliottmorris
Trump didn’t win because he was more moderate than Clinton, he won because he had a monopoly on anti-system voters. Politics is not one-dimensional, but everyone in DC and the media fails to account for this. (That’s why they were also blindsided by the DSA wins this year )
Source www.gelliottmorris.com/p/meet-americas-new-swing-voter-the
David Weigel
@daveweigel
Clinton ‘16 ran to the left of Obama ‘12, and voters saw her as more liberal than him.
Not a hot take at the time — Dems celebrated that they had the most progressive platform in history — but that election is forever remembered On Here as Dems losing bc they went w a centrist instead of a populist. -
G Elliott Morris reposted
Matthew S. Shugart
@laderafrutal
Coordination-problem watch!
G Elliott Morris
@gelliottmorris
If Democrats want to hold a U.S. Senate seat in Montana, they should support Bankhead.
But if what they want is to actually control the Senate, getting her to drop out is a very high-leverage use of resources/influence.
G Elliott Morris
@gelliottmorris
A new poll shared exclusively with Strength In Numbers finds independent candidate Seth Bodnar polling 6 points ahead of Republican Kurt Alme in a hypothetical one-on-one for Montana's Senate race, but the Democratic candidate Alani Bankhead trailing by 8.
www.gelliottmorris.com/p/2026-07-28-exclusive-independent… -
G Elliott Morris
@gelliottmorris
If Democrats want to hold a U.S. Senate seat in Montana, they should support Bankhead.
But if what they want is to actually control the Senate, getting her to drop out is a very high-leverage use of resources/influence.
G Elliott Morris
@gelliottmorris
A new poll shared exclusively with Strength In Numbers finds independent candidate Seth Bodnar polling 6 points ahead of Republican Kurt Alme in a hypothetical one-on-one for Montana's Senate race, but the Democratic candidate Alani Bankhead trailing by 8.
www.gelliottmorris.com/p/2026-07-28-exclusive-independent… -
G Elliott Morris
@gelliottmorris
A new poll shared exclusively with Strength In Numbers finds independent candidate Seth Bodnar polling 6 points ahead of Republican Kurt Alme in a hypothetical one-on-one for Montana's Senate race, but the Democratic candidate Alani Bankhead trailing by 8.
www.gelliottmorris.com/p/2026-07-28-exclusive-independent… -
G Elliott Morris reposted
Verasight
@verasight_data
See how G. Elliott Morris uses Verasight data for polling at the state, district, and sub-county levels:
G Elliott Morris
@gelliottmorris
I've updated the MRP model Strength In Numbers uses for polling at the state, district, and sub-county levels w a SOTA pipeline that predicts 50/50 states correctly in 2024 + every Senate seat in 2022. Now available to SIN Pro users and enterprise clients
www.gelliottmorris.com/p/what-do-voters-in-each-congressi… -
G Elliott Morris
@gelliottmorris
New polling exclusive in the Montana senate race dropping at Strength In Numbers tomorrow morning at 6 AM Eastern
Get it in your inbox: www.gelliottmorris.com/subscribe -
G Elliott Morris
@gelliottmorris
Small sample size caveat but it is worth pointing out that there is no public polling data that shows Haley Stevens doing better than Abdul El-Sayed against Mike Rogers. If anything, she does worse fiftyplusone.news/polls/senate/general/michigan
Haley Stevens
@HaleyforMI
What The Post says 💁♀️
-
G Elliott Morris
@gelliottmorris
I've updated the MRP model Strength In Numbers uses for polling at the state, district, and sub-county levels w a SOTA pipeline that predicts 50/50 states correctly in 2024 + every Senate seat in 2022. Now available to SIN Pro users and enterprise clients
www.gelliottmorris.com/p/what-do-voters-in-each-congressi…
G Elliott Morris
@gelliottmorris
tried to make this clear in the post but like if you have questions that would be editorially interesting to project to the district level, i can just... run it
-
G Elliott Morris
@gelliottmorris
returning for the midterms election season: the Sunday strength in numbers weekly roundup. got data to share? email us
How Democrats can win over more midterm voters
www.gelliottmorris.com/p/how-democrats-can-win-over-more-… -
G Elliott Morris
@gelliottmorris
Good chart
-
G Elliott Morris reposted
Nathaniel Rakich
@baseballot
To be clear, there is no evidence that this clip affected the primary whatsoever. fiftyplusone.news/polls/senate/democratic-primary/michigan
TMZ
@TMZ
Rep. Haley Stevens’ viral “stick it to ’em” clip could’ve hurt her campaign, but instead gave it a boost.
TMZ DC’s @jacob_wass and @hicharliecotton break down the power of going viral. 💥This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
Someone should do a poll of Maine voters asking about their faith/trust in the Democratic nominating process. Would help with understanding how Americans feel about alternative systems for nominee selection (& that are more common in multiparty democracies)
-
G Elliott Morris
@gelliottmorris
Are the DSA’s policy positions popular? 50+1 takes a look open.substack.com/pub/fiftyplusone/p/how-popular-are-marq…
-
G Elliott Morris
@gelliottmorris
Our new poll finds the Americans who most support Donald Trump’s “SAVE America” Act know little to nothing about it
www.gelliottmorris.com/p/2026-07-24-save-act-proof-of-cit… -
G Elliott Morris
@gelliottmorris
New poll from Strength In Numbers and Verasight this morning: Black voters and Democrats are systematically more likely to say they couldn’t process citizenship. These are **currently registered voters** — and so presumably citizens being disenfranchised.
www.gelliottmorris.com/p/2026-07-24-save-act-proof-of-cit… -
G Elliott Morris
@gelliottmorris
I’ll be presenting research I did with @verasight_data in 2025-26 on using LLMs for survey data generation at the Analyst Institute’s 2026 conference in Washington, DC tomorrow morning. Email me if you want to grab a coffee or chat!
members.analystinstitute.org/wp-login.php?redirect_to=htt… -
G Elliott Morris
@gelliottmorris
New at the Strength In Numbers Data Lab: How pollsters weight by party ID is influencing Democrats' midterm standing. Different methodologies give a margin anywhere between D+7 and D+10 (among registered voters — likely even higher among likely voters).
www.gelliottmorris.com/p/2026-07-23-lab-npors-pid5-weighting -
G Elliott Morris
@gelliottmorris
Here's my latest at the 50+1 blog: Despite some tightening in the generic ballot, Democrats are still heavy House favorites and lead or are tied in 6 GOP-held Senate seats, making them nearly even odds to take it, too. Jackson at +2 in Maine. blog.fiftyplusone.news/p/fiftyplusone-nowcast-update-troy
-
G Elliott Morris
@gelliottmorris
Publishing an update to our House and Senate "nowcast" model tomorrow morning at 50+1
There are some surprises. Get it in your inbox:
blog.fiftyplusone.news/subscribe -
G Elliott Morris
@gelliottmorris
Likely voters for this year's midterms lean more Democratic than registered voters or all adults, a change from last cycle. Democrats are +9 among self-described likely voters in this month's Strength In Numbers/Verasight poll, vs +7 among RVs.
www.gelliottmorris.com/p/2026-07-22-midterm-enthusiasm-ga… -
G Elliott Morris
@gelliottmorris
First article about our monthly Strength In Numbers/ @verasight_data poll is up today
- 59% say gas price hikes have forced them to cut back
- 51% blame Trump
- 68% say prices will influence their vote in Nov. Cake may already be baked on Iran fallout
open.substack.com/pub/gelliottmorris/p/2026-07-21-july-st… -
G Elliott Morris
@gelliottmorris
New Strength In Numbers/Verasight polling out this week on the midterms and the enthusiasm gap, Trump’s mental fitness for office, SAVE Act, AI data centers, gas price worries and more. Get it in your inbox at gelliottmorris.com/subscribe
Data is online for _SIN_ Pro subscribers -
G Elliott Morris
@gelliottmorris
fwiw here is a link to download the north carolina voter file. congratulations you have now apparently done election espionage! www.ncsbe.gov/results-data/voter-registration-data
-
G Elliott Morris
@gelliottmorris
Big story from the new Pew data out yesterday: Non-voters lean Democratic in party ID again, by a 12-point margin. That’s up from R+4 in 2024. Huge shift. The party is approaching its lead with the group from 2020-2021.
www.gelliottmorris.com/p/2026-07-15-pew-non-voters -
G Elliott Morris
@gelliottmorris
Democrats lead by 26 points among “double haters” in the 2026 House midterms vote, according to new Strength In Numbers/ @verasight_data poll www.gelliottmorris.com/p/2026-07-14-double-haters-2026
-
G Elliott Morris
@gelliottmorris
Hey he’s only off by a little over 20 points — give the guy a break right? fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
Does not inspire confidence
-
G Elliott Morris
@gelliottmorris
New post up on the new Strength In Numbers Data Lab, a vertical for writing about our poll, answering the odd reader question, or just thinking through other super nerdy things.
First topic: Let’s revisit those Trump disapprovers who aren’t voting Dem:
www.gelliottmorris.com/p/2026-07-13-lab-gop-anchors-gener… -
G Elliott Morris
@gelliottmorris
Re: the McConnell news, our roundup in May found strong support for both age and term limits for President, Congress and the Supreme Court
blog.fiftyplusone.news/p/janet-millss-defeat-in-maine-isnt -
G Elliott Morris
@gelliottmorris
Here's the definitive numbers take on Graham Platner: He's bad. Specifically, he's running about 10 points behind expectations — and in a race that decides control of the Senate about 10% of the time (the third most likely tipping point).
www.gelliottmorris.com/p/2026-07-07-graham-platner -
G Elliott Morris
@gelliottmorris
Democrats are polling ahead of or even with Republicans in six key GOP-held U.S. Senate races, giving them even odds to take the chamber if the midterms were held today
FiftyPlusOne.news
@50plus1news
New: An early look at House and Senate general election polling averages, + a polling "nowcast" model that predicts results of the midterms if they were held today.* **
Polls suggest a D+7-8 national environment + easy House majority, toss-up Senate
blog.fiftyplusone.news/p/new-fiftyplusone-polling-nowcast -
G Elliott Morris
@gelliottmorris
Here’s a huge data point in the current debate about why DSA candidates are suddenly winning everywhere:
Bernie Sanders is the most favorable Democratic political figure today — even among self-described ideological “moderates.”
Huh?
www.gelliottmorris.com/p/2026-07-07-democratic-fracture-n… -
G Elliott Morris
@gelliottmorris
Tomorrow at Strength In Numbers: The hard data that shows why DSA candidates are suddenly winning everywhere. It’s not because all Americans are suddenly leftists www.gelliottmorris.com/p/2026-07-07-democratic-fracture-n…
-
G Elliott Morris
@gelliottmorris
TIL it’s trivial to replace your own AC compressor
-
G Elliott Morris
@gelliottmorris
lots of anti-system vs establishment discourse going on re: Dem primary outcomes over the last couple weeks
see www.gelliottmorris.com/p/meet-americas-new-swing-voter-the
worth noting that at the general election level, establishment picks are also polling somewhat poorly comparatively
fiftyplusone.news/polls/senate/general/michigan -
G Elliott Morris
@gelliottmorris
to be fair, who among us hasn’t hit “publish” instead of “save” on a prewrite
-
G Elliott Morris
@gelliottmorris
See, I told you. gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
How can Democrats outrun their party’s brand problem in red states and win the Senate again? I ran a survey experiment that tested three proposed fixes.
One insight: policy moderation doesn’t beat generically signaling anti-establishment/party attitudes
www.gelliottmorris.com/p/2026-06-30-democratic-brand-issu… -
G Elliott Morris
@gelliottmorris
It’s not just New York; Democrats everywhere rate lefty anti-establishment candidates better than party leadership. New polling:
www.gelliottmorris.com/p/2026-06-29-most-popular-pols-par… -
G Elliott Morris
@gelliottmorris
Per our new polling out today, the most popular politicians in America are…
Barack Obama
Bernie Sanders
Zohran Mamdani
Pete Buttigieg
Jon Ossoff
& Alexandria Ocasio-Cortez
Trump is tied with Elon Musk and Hakeem Jeffries, Schumer and Carlson in last
www.gelliottmorris.com/p/2026-06-29-most-popular-pols-par… -
G Elliott Morris
@gelliottmorris
i'd read a mcmorrow campaign autopsy of this race fiftyplusone.news/polls/senate/democratic-primary/michigan
-
G Elliott Morris
@gelliottmorris
Here's the current state of play in the Senate if you use polls and reasonable guesses about the state of play in each seat projects.gelliottmorris.com/senate-2026/?n=7&s=AK%3A9%2CI…
-
G Elliott Morris
@gelliottmorris
Our new poll: A clear majority of voters think there are grounds to impeach President Trump, 53% to 39%. Read what they said in their own words: www.gelliottmorris.com/p/2026-06-25-grounds-to-impeach-trump
-
G Elliott Morris
@gelliottmorris
New Strength In Numbers/ @Verasight_data poll:
- Only 18% think U.S. achieved its goals in Iran
- Trump's approval flirts with historic low at 37%
- Plurality says Trump admin. has benefited the wealthy over everyday Americans
- and more! Free to read: www.gelliottmorris.com/p/2026-06-24-june-strength-in-numb… -
G Elliott Morris
@gelliottmorris
In this post I introduce the Strength In Numbers “cost of ruling index” — a metric that combines presidential approval, the generic ballot, and other data into a single measure of the incumbent party’s strength. Currently exclusive for paying subscribers
www.gelliottmorris.com/p/2026-06-23-generic-ballot-noise -
G Elliott Morris
@gelliottmorris
Democrats' lead on the House generic ballot "dropped" 1-2 points over the last month. It will likely go back up, but this is a good reminder that even polling averages can fake you out
www.gelliottmorris.com/p/2026-06-23-generic-ballot-noise -
G Elliott Morris
@gelliottmorris
Tomorrow's Deep Dive post at Strength In Numbers: ZOMG! Trump's approval has gone up by a point! The generic ballot sky is falling! Democrats are doomed!
Or is this just polling noise? We've got the tools to figure out:
www.gelliottmorris.com/p/2026-06-23-generic-ballot-noise -
G Elliott Morris
@gelliottmorris
More U.S. House members than ever are running for other offices this year — and more than ever are losing their races open.substack.com/pub/fiftyplusone/p/house-members-keep-l…
-
G Elliott Morris
@gelliottmorris
This poll question has predicted the winner of every election since 1952. Democrats now lead Republicans on it by 13 points — their largest lead of the 2026 election cycle so far www.gelliottmorris.com/p/2026-06-16-personal-mip-lead-dem…
-
G Elliott Morris
@gelliottmorris
A new poll from the New York Times shows potential Democratic voters want the party to move to the right, are happy with where it is ideologically, and rate socialism highly. Huh?
www.gelliottmorris.com/p/2026-06-12-dem-voters-moderate-a… -
G Elliott Morris
@gelliottmorris
Yes, Americans are polarized. But there can still be big swings in voter behavior between elections. How do we solve this puzzle? www.gelliottmorris.com/p/2026-06-09-party-loyalty-and-202…
-
G Elliott Morris
@gelliottmorris
Americans say keeping up with key news topics is crucial. But fewer say they’re highly informed themselves www.pewresearch.org/short-reads/2026/06/05/americans-say-…
-
G Elliott Morris
@gelliottmorris
This chart from the latest YouGov/Economist poll result is really quite stunning. Trump’s net approval is not only deeply negative, but about 15(!!) points below what it was at this point in 2018, putting the GOP deeper in blue wave territory now vs then substack.com/@gelliottmorris/note/c-269482494?r=a9pj
-
G Elliott Morris
@gelliottmorris
would love to see gallup update this chart for 2026
news.gallup.com/poll/651092/2024-election-environment-fav… -
G Elliott Morris
@gelliottmorris
New report out today with my friends at Verasight writing up a couple new studies I designed on AI “digital twins” in polling. The upshot is basically these tools are very inaccurate unless you have raw polling data to build on (which defeats the purpose!) www.verasight.io/reports/can-ai-digital-twins-replace-hum…
-
G Elliott Morris
@gelliottmorris
I did the math: Here's exactly how James Talarico can beat Ken Paxton for the Texas Senate race. (He doesn't need as many Trump voters as conventional wisdom dictates.)
www.gelliottmorris.com/p/2026-01-01-how-many-republicans-… -
G Elliott Morris
@gelliottmorris
made some fun new data viz for this article. out tomorrow morning www.gelliottmorris.com/p/2026-01-01-how-many-republicans-…
-
G Elliott Morris
@gelliottmorris
Here's why so much election analysis is useless: A race decided by a point and a half can be blamed on almost anything. Most party autopsies don't stand the test of time, and the narratives that get pushed are mainly exercises in confirmation bias www.gelliottmorris.com/p/2026-05-29-dnc-autopsy-few-more-…
-
G Elliott Morris
@gelliottmorris
If the actual result of the Texas Senate race among Latinos ends up anywhere close to this, Paxton is cooked unidosus.org/press-releases/new-unidosus-bipartisan-poll-…
-
G Elliott Morris
@gelliottmorris
a problem for republicans is that the entire party, from the white house to legislators to campaign managers to people who make ads and clip speeches etc, are chronically online, and chronically online on X in particular. it's worse than the bubble Dems used to get criticized for
-
G Elliott Morris
@gelliottmorris
Interesting analysis of Trump's approval by 'county type' and how those counties voted in 2024. From the American Communities Project, using data from Strength In Numbers www.americancommunities.org/new-analysis-shows-trumps-app…
-
G Elliott Morris
@gelliottmorris
this is from a recent poll (not trying to shame them so not including name of pollster). just one data point, but have response rates stabilized around 1%? is this... good news?
-
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
New poll: Americans blame Trump more than Biden, Congress for the cost-of-living crisis. Here's what they want done about it.
www.gelliottmorris.com/p/2026-05-27-trump-owns-the-cost-o… -
G Elliott Morris
@gelliottmorris
Trump approval among registered voters is now positive in just 7 states and below 45% in all swing states, according to my estimates using Strength In Numbers/Verasight polling. The president's approval rating is underwater in 136 GOP-held House districts projects.gelliottmorris.com/trump-approval/
-
G Elliott Morris
@gelliottmorris
New deep dive on the House generic ballot from my colleague Mary Radcliffe and me (mostly Mary):
The Dem lead on the generic ballot is now as large as it was at this point in 2018.
That’s due to sky-high loyalty from Dems and notable defection among Rs
open.substack.com/pub/fiftyplusone/p/the-democratic-lead-… -
G Elliott Morris
@gelliottmorris
NEW: Americans blame Trump for the cost-of-living crisis.
64% of all adults say he is doing worse than they expected — including 30% of Republicans.
Voters support redistributive policies that fund social spending via taxes on the wealthy and corporations www.gelliottmorris.com/p/2026-05-27-trump-owns-the-cost-o… -
G Elliott Morris
@gelliottmorris
IMO, the Texas Senate race is a pure tossup. And if you take a deep and fair look at the data, I think you'll find you have a hard time concluding otherwise www.gelliottmorris.com/p/the-texas-senate-election-is-a-t…
-
G Elliott Morris
@gelliottmorris
Fine, I'll say it: The Texas Senate race is a tossup
www.gelliottmorris.com/p/the-texas-senate-election-is-a-t… -
G Elliott Morris
@gelliottmorris
Here's estimated generic ballot vote intention for key states, per the latest modeling from Strength In Numbers. The Pro data portal is updated with our May survey results: pro.gelliottmorris.com
G Elliott Morris
@gelliottmorris
(two-party Dem share BTW)
-
G Elliott Morris
@gelliottmorris
NEW Strength In Numbers/ @verasight_data poll: Trump approval on prices falls to -47; new high says inflation is the “most important problem” (39%); voters say they trust Ds over Rs to handle their most important problem by 13 points; D+8 for House vote
www.gelliottmorris.com/p/2026-05-26-may-strength-in-numbe… -
G Elliott Morris
@gelliottmorris
one of the best tweets of all time
-
G Elliott Morris
@gelliottmorris
not to dwell on the events of last week, but every problem nate lists here (and there are a lot of them) was at least as true, if not worse, during my tenure heading FiveThirtyEight. i cannot imagine living through that for a decade (i mean this literally) www.natesilver.net/p/disney-erased-fivethirtyeight
-
G Elliott Morris
@gelliottmorris
Have got a new poll out today on the reasons many Trump voters give for defecting from him over the last year, how Americans feel about Trump’s prioritization of the economy (bad, and punishing the GOP for it), and tests some immigration messaging morrispredictive.com/assets/surveys/tyg730HYHsJ/memo.pdf
-
G Elliott Morris
@gelliottmorris
The DNC "autopsy" omits the biggest reason Democrats lost in 2024: inflation and anti-incumbent sentiment.
In doing so, it fails to incentivize party actors to think about how they will overcome the same negative forces in 2030 and 2032
www.gelliottmorris.com/p/the-dnc-autopsy-omits-the-biggest -
G Elliott Morris
@gelliottmorris
It would be smart for parties to commission "autopsies" w/ original data collection and analysis from multiple firms across the ideological spectrum. The DNC had to release theirs because of Martin's promises but judging by the reaction among party elites, it ain't helping anyone
-
G Elliott Morris
@gelliottmorris
Americans oppose spending $1 billion on White House ballroom 68% to 21%
www.gelliottmorris.com/p/americans-oppose-spending-1-billion -
G Elliott Morris
@gelliottmorris
Republicans are now deep underwater (-20) with the voters who make decisions based on the price of gas and general economic conditions — the group that put Trump in the White House in the first place
www.gelliottmorris.com/p/2026-05-19-indies-nonvoters-turn… -
G Elliott Morris
@gelliottmorris
Strength In Numbers mentioned here nymag.com/intelligencer/article/polls-show-democrats-have…
-
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
Trump is really wreaking havoc on the GOP's traditional advantage on the economy. Here's the current congressional district-level map on D/R trust to handle the issue. (National break here is D+8 among registered voters.)
More CD-lvl (and county!) maps at pro.gelliottmorris.com -
G Elliott Morris
@gelliottmorris
The blue wave is building. Yesterday's NYT/Siena poll showed independents at D+18 on the generic ballot, the same as 2006.
In our Strength In Numbers/Verasight data, indies and non-voters have been trending blue all cycle — now up to D+20, vs R+4 in 2024.
www.gelliottmorris.com/p/2026-05-19-indies-nonvoters-turn… -
G Elliott Morris reposted
G Elliott Morris
@gelliottmorris
Republicans were as relatively unpopular as Democrats are now in 2014 and in early 2016. Despite the media’s heavy focus on favorability of the parties, these numbers are not very predictive of outcomes www.gelliottmorris.com/p/democratic-party-favorability-ra…
-
Post not available here, view on X
G Elliott Morris
@gelliottmorris
Republicans were as relatively unpopular as Democrats are now in 2014 and in early 2016. Despite the media’s heavy focus on favorability of the parties, these numbers are not very predictive of outcomes www.gelliottmorris.com/p/democratic-party-favorability-ra…
-
G Elliott Morris
@gelliottmorris
If you use the two-party turnout stats here to forecast the national environment, we are currently in something like D+8. And that’s before the typical summer-fall pattern of the polls breaking against the incumbent party
VoteHub
@VoteHub
Turnout in Louisiana’s U.S. Senate primaries was broadly similar across both parties, even with a competitive GOP race, though Republicans outperformed their registration share.
Ballots cast (two-party share)
🔴 Republicans – 401,118 (53.8%)
🔵 Democrats – 344,751 (46.2%) -
G Elliott Morris
@gelliottmorris
Trump’s strategy for dealing with inflation worries and just high economic anxiety in general this cycle is apparently to do a Brian Cox in Succession and yell “oh fuck off” at and walk away from anyone who asks. And it appears to be having exactly the effect you’d predict
-
G Elliott Morris
@gelliottmorris
When I was at ABC I had to rally the 538 team to threaten to quit as leverage to have the old website archived instead of taken down. Not surprising they did this now, after all the real journalists on the digital side of the org have been laid off or left
www.nytimes.com/2026/05/16/business/media/fivethirtyeight… -
G Elliott Morris
@gelliottmorris
Trump’s approval rating hits second-term low among his Latino voters, though many still approve
www.pewresearch.org/short-reads/2026/05/15/trumps-approva… -
G Elliott Morris
@gelliottmorris
Pundits/"analysts" generally overrate candidate effects (especially triangulating whatever "ideology" means) and underrate conditions as an explanation for recent election results
goodpoliticsbadpolitics.substack.com/p/the-right-candidat… -
G Elliott Morris
@gelliottmorris
Despite all the vote rigging news over the last 2-3 weeks, there is one thing I keep coming back to about Donald Trump: Nobody likes him. And you can't gerrymander yourself out of a bad approval rating. www.gelliottmorris.com/p/2026-05-12-you-cant-gerrymander-…
-
G Elliott Morris
@gelliottmorris
Jain raises a good point: If you update an elec forecast w data from a prediction market, which ingests your forecast, are you breaking the logic of ur model?
Yes. I wrote about this last year. You want to "separate" the market signal & the forecast signal www.theargumentmag.com/p/kalshi-crashes-the-polling-party -
G Elliott Morris
@gelliottmorris
In this special deep dive episode of the Strength In Numbers podcast, I talk with voting and democracy expert Lee Drutman, who makes a good case for why you should support proportional representation and fusion voting! In pod feeds now www.gelliottmorris.com/p/deep-dive-episode-democracy-20-with
-
G Elliott Morris
@gelliottmorris
fwiw www.cygn.al/wp-content/uploads/2026/05/23838-Cygnal-Natio…
-
G Elliott Morris
@gelliottmorris
This map shows, in purple, congressional districts where Trump's approval on prices is below 50% — and, in green, districts where he is above 50%.
Local estimates of approval, party trust, and the generic ballot are now available at Strength In Numbers Pro. Link in next tweet
G Elliott Morris
@gelliottmorris
Check out this data and more at pro.gelliottmorris.com
-
G Elliott Morris
@gelliottmorris
Moderate Democrat Ruben Gallego comes out in support of packing the Court. Also popular with a plurality of Americans, including independents, according to our Strength In Numbers/ @verasight_data polling:
www.gelliottmorris.com/p/two-thirds-of-americans-want-term -
G Elliott Morris
@gelliottmorris
Thanks to freshly gerrymandered House maps across the country, the number of competitive districts will fall to a new low of 33 (or less) this year, and only 15 are true toss-ups. www.gelliottmorris.com/p/2026-05-12-uncompetitive-seats
-
G Elliott Morris
@gelliottmorris
As they say, some personal news www.gelliottmorris.com/p/one-year-of-strength-in-numbers?…
-
G Elliott Morris
@gelliottmorris
February: Gallup says that due to cost reasons it will no longer do presidential approval polling
May: Gallup says it is exploring "digital twin" technology that replaces human respondents with AI/LLM-generated answers
news.gallup.com/opinion/methodology/709373/gallup-begins-… -
G Elliott Morris
@gelliottmorris
After losing their map in VA, Strength In Numbers estimates Dems are now down 6 seats in the House from mid-decade + post-VRA redistricting. Could lose 2-3 more across the South.
Dems need to win the House pop. vote by 4 to win the majority of seats. www.gelliottmorris.com/p/2026-05-10-dem-house-pop-vote-th… -
G Elliott Morris
@gelliottmorris
Really is remarkable what a 36% approval rating looks like in map form. The Economist's MRP model has Trump's approval rating among all adults net negative in all but 6 states www.economist.com/interactive/trump-approval-tracker
-
G Elliott Morris
@gelliottmorris
Democrats have now lost a net of six House seats in enacted or likely mid-decade redistricting across the U.S. They will need to win the House popular vote by 4 points to win the majority of seats. New House elec. simulations from Strength In Numbers:
www.gelliottmorris.com/p/2026-05-10-dem-house-pop-vote-th… -
G Elliott Morris
@gelliottmorris
I miss the FiveThirtyEight "Atlas of Redistricting" tracker :(
-
G Elliott Morris
@gelliottmorris
The Supreme Court of VA invalidates the new pro-D gerrymander (www.vacourts.gov/static/opinions/opnscvwp/1260127.pdf), which (with new maps in FL and TN this week) updates the redistricting playing field to something like
D+5
R+14 (net +9)
Will do some math on the impact to the pop vote threshold & blog it -
G Elliott Morris
@gelliottmorris
A very Strength In Numbers-coded article today: It looks like Republican Supreme Court justices made a very rudimentary, but willful, statistical error in their effective invalidating of the VRA in the Callais decision.
www.gelliottmorris.com/p/2026-05-08-simple-math-error-sco… -
G Elliott Morris
@gelliottmorris
You’ve probably heard a lot about Democrats getting “locked out” of the California general election for governor this year. But my colleague at 50+1 Mary Radcliffe shows why this is equally or more likely to happen to Republicans: open.substack.com/pub/fiftyplusone/p/republicans-could-ge…
-
G Elliott Morris
@gelliottmorris
Interview about journalism and prediction markets in the Columbia Journalism Review this AM. Many in the media are using this new “forecast” data incredibly irresponsibly, and are failing to wrestle with the ethics of the companies they are dealing with www.cjr.org/tow_center/a-bet-is-not-a-poll-polymarket-kal…
-
G Elliott Morris
@gelliottmorris
Pleasure having Seth Masket on the pod to talk about everything political scientists have got right about Trump over the last ~6 years. Plus, we talk abt Dem strategy for 2028 and beyond. Can Ds avoid anti-incumbent backlash once they're back in power?
www.gelliottmorris.com/p/deep-dive-episode-political-scie… -
G Elliott Morris
@gelliottmorris
The problem with Matt's tweet is the claim "Republicans lost the Senate in 2022 because they nominated a bunch of election deniers Trump endorsed" is not the same as "Dems would have won the House and Senate in 2024 if they nominated moderates" — because that's what they did! 1/4
Post not available here, view on X
G Elliott Morris
@gelliottmorris
Pundits see the world in ideological terms. This is not how voters experience politics. Voters say the R party is extreme and D party is weak, they don’t know what it stands for. There’s enough $$ for the moderation strategists. Who is doing long-term planning for econ anxiety?
G Elliott Morris
@gelliottmorris
People like Matt have turned this debate into a Go Left vs Go Right debate, when that's not what those of us looking at the data are even arguing. Matt wants all Dem energy to be spent on moderation. Data suggests other options may have higher leverage... so let's try those too
-
G Elliott Morris
@gelliottmorris
Trump's intrusion into GOP primary elections forces elected Republicans to take his side, often endorsing dramatically unpopular stances — which generates severe blowback for the midterms. E.g., this probably cost the GOP control of the Senate in 2022 www.gelliottmorris.com/p/2026-05-06-indiana-primaries-tru…
-
G Elliott Morris
@gelliottmorris
trump is breaking y axes left and right
-
G Elliott Morris
@gelliottmorris
"The new poll finds 34 percent of voters saying they trust Republicans to handle the economy, 33 percent saying they trust Democrats and the rest saying they trust neither or both equally. In 2022, Republicans had a double-digit advantage on the economy."
www.washingtonpost.com/politics/2026/05/03/trump-approval…
G Elliott Morris
@gelliottmorris
As predicted 14 months ago in Strength In Numbers: www.gelliottmorris.com/p/the-trump-vibe-shift
-
G Elliott Morris
@gelliottmorris
New data from me & David Nir for Strength In Numbers: It'll be years before Americans get used to higher prices — and politicians can't just wait it out. Consumers will eventually adjust, but in the meantime, they'll keep punishing leaders who don't act
www.gelliottmorris.com/p/2026-05-05-how-long-until-econom… -
G Elliott Morris
@gelliottmorris
Maul – Shadow Lord: pretty good television
-
G Elliott Morris
@gelliottmorris
Trump's net job approval rating this morning is rounding to -23 for the first time fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
guys i’m so tired of all the winning
-
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
Pundits say Dems are underperforming in the midterms because Trump approval is -20 yet Dems are "only" +5-8 on the GB.
But that misunderstands the disapprove-but-not-Dem group. It’s mostly closeted GOP partisans, not moderates waiting for Ds to pivot right www.gelliottmorris.com/p/2026-05-01-cross-pressured-voter… -
G Elliott Morris
@gelliottmorris
trump literally broke the scale of this graph on my data portal
www.gelliottmorris.com/p/data -
G Elliott Morris
@gelliottmorris
We will be talking about the impacts of SCOTUS's VRA decision on the midterms for our podcast this week — and also looking at the data on what Trump disapprovers who aren't voting Dem in the midterms want.
Join us live at 2PM ET. Questions encouraged!
open.substack.com/live-stream/183304 -
G Elliott Morris
@gelliottmorris
here's an IMO delightfully wonky article about how our polling averages work in primary campaigns:
blog.fiftyplusone.news/p/making-polling-averages-react-to -
G Elliott Morris
@gelliottmorris
AAPOR’s Amicus brief in the Trump vs Selzer/Des Moines Register case is worth reading aapor.org/amicus-brief-overview/?_zs=KdvFo1&_zl=5FPiA
-
G Elliott Morris
@gelliottmorris
This poll writeup feels like two pieces stapled together. The Q here is important (I wrote about it last month), but it's not empirically explained by crime, an issue that's the Most Important Problem of <5% of voters & where Dems have a small -3 trust gap www.theargumentmag.com/p/why-democrats-cant-win-more-trump
-
G Elliott Morris
@gelliottmorris
who could have possibly predicted that firing all your best journalists and marginalizing your core audience would have had a deleterious effect on one’s business model? impossible to see coming, really. truly a mystery
www.status.news/p/tony-dokoupil-cbs-evening-news-ratings-… -
G Elliott Morris
@gelliottmorris
A new theory that Republicans can win in November by getting turnout up is not rooted in the actual turnout data from recent elections, but rather an ideologically convenient refusal to acknowledge the real problem the party faces: Donald Trump www.gelliottmorris.com/p/2026-04-28-no-turnout-wont-save-…
-
G Elliott Morris
@gelliottmorris
Good modeling, though I wouldn't treat a vote for "no" necessarily as a vote against Dems. An aggressive gerrymandering referendum probably offers us the lower bound on Dem vote share for the midterms, not the median www.gelliottmorris.com/p/2026-04-28-no-turnout-wont-save-…
Zachary Donnini
@ZacharyDonnini
Nonwhite voters overwhelmingly shifted back toward Democrats in Virginia’s redistricting referendum last week.
YES ran better in pretty much all Hispanic and Asian plurality precincts compared to Kamala Harris’s 2024 baseline, while White precincts moved right. -
G Elliott Morris
@gelliottmorris
The GOP doesn’t have a midterm turnout problem
It has a Trump problem
“Republicans just need to get their voters off the sidelines” is a much more comforting story than “voters saw the Trump presidency and decided they don't want it"
But the data is clear www.gelliottmorris.com/p/2026-04-28-no-turnout-wont-save-… -
G Elliott Morris
@gelliottmorris
IMO banger post coming tomorrow morning, with fresh analysis of voter file data and polling micro data on key 2025 and 2026 races www.gelliottmorris.com/p/2026-04-28-no-turnout-wont-save-…
-
G Elliott Morris
@gelliottmorris
As to the question of whether the events at the WHCD on Saturday will meaningfully impact the president's popularity (some are saying he'll get a bump), someone very close to Trump was actually assassinated not six months ago and this had no detectable effect on public opinion
-
G Elliott Morris
@gelliottmorris
New from us over at 50+1: The Democrats' favorability rating gap vs the Republicans — which drove so much hand-wringing and many news cycles in 2025 — has basically reverted back to normal
blog.fiftyplusone.news/p/democrats-have-nearly-closed-their
G Elliott Morris
@gelliottmorris
(as we predicted last July) www.gelliottmorris.com/p/democratic-party-favorability-ra…
-
G Elliott Morris
@gelliottmorris
Some ppl are saying the VA referendum shows Republicans can hold the House in November. IMO this is 100% pure uncut hopium
(a) gerrymandering is very unpopular, more unpopular than Trump!
(b) status quo bias dominates referendums
(c) turnout patterns won't be predictive of Nov -
G Elliott Morris reposted
Chris Warshaw
@cwarshaw
PlanScore's current national seat share estimate (using a generic ballot of D+5; +7D point shift from 2024) is similar to the Economist's model. PlanScore thinks Democrats would win ~233 seats if the election were held today vs. Economist's 232 seats. planscore.org
G Elliott Morris
@gelliottmorris
The Economist's new House and Senate midterms forecasting models get their polling data from our API at FiftyPlusOne. Building something that needs raw polling data as an input? Check out our data services at fiftyplusone.news/downloads
-
G Elliott Morris
@gelliottmorris
democrats should have a plan for how they're going to counter anti-incumbent backlash on price vibes once they're back in power, or the political yo-yo will just continue yo-ing
Aaron Blake
@AaronBlake
Fox News poll shows Americans prefer Democrats to Republicans on the economy (D+4) for the first time since 2010.
-
G Elliott Morris
@gelliottmorris
One out of every five 2024 Trump voters supports impeaching the president, per our new poll out at Strength In Numbers. I was pretty surprised by this level of defection among the president's base. www.gelliottmorris.com/p/2026-04-22-strength-in-numbers-v…
-
G Elliott Morris
@gelliottmorris
Left: All the groups that currently disapprove of President Trump’s job performance in our polling
Right: The two groups where he’s above water
😬😬
www.gelliottmorris.com/p/2026-04-24-trump-approval-groups… -
G Elliott Morris
@gelliottmorris
Been working on something cool. Excited to share soon
-
G Elliott Morris
@gelliottmorris
This is what a 35% approval rating looks like at the geographic and demographic level
www.gelliottmorris.com/p/2026-04-24-trump-approval-groups… -
G Elliott Morris
@gelliottmorris
Doing this webinar for NYAAPOR tomorrow on what my polling shows about the state of the midterms. The GOP's trouble is mostly, but not all, about prices. Tune in! www.eventbrite.com/e/reading-the-electorate-what-the-data…
-
G Elliott Morris
@gelliottmorris
From our Strength In Numbers/ @Verasight_data polling this month: Many voters say the Democrats are weak, but what types of things do they want the party to stand up for? We asked respondents to answers in their own words, and the results are revealing. www.gelliottmorris.com/p/2026-04-23-poll-results-party-favs
-
G Elliott Morris
@gelliottmorris
The Economist's new House and Senate midterms forecasting models get their polling data from our API at FiftyPlusOne. Building something that needs raw polling data as an input? Check out our data services at fiftyplusone.news/downloads
-
G Elliott Morris
@gelliottmorris
Good post from Krugman on the economic vibes again, esp the last chart:
paulkrugman.substack.com/p/bad-vibes-and-broken-promises
In my piece last week I shared a new paper from Chris Wlezien and Derek Epp at UT Austin showing he's exactly right, that people respond to inflation worse during good times vs bad: -
G Elliott Morris
@gelliottmorris
Support for impeaching Trump is at 55%, according to a new Strength In Numbers/ @Verasight_data poll, with 37% of adults opposed. Crucially 1 in 5 of Trump’s own voters support impeachment.
Read and share:
www.gelliottmorris.com/p/2026-04-22-strength-in-numbers-v… -
G Elliott Morris
@gelliottmorris
Looks good and looks right! + I like the vertical dot stack for representing simulations. 🦊
www.economist.com/interactive/2026/us-midterms/prediction… -
G Elliott Morris
@gelliottmorris
New polling on impeaching Trump is coming out at Strength In Numbers tomorrow at 7:00 AM Eastern. It's free to read and get in your inbox! www.gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
Just 26% of Americans say they approve of how the president is handling prices/inflation, 35% on jobs and the economy. His net rating is way underwater on both
www.gelliottmorris.com/p/2026-04-21-april-strength-in-num… -
G Elliott Morris
@gelliottmorris
NEW Strength In Numbers/Verasight.io poll: Trump approval falls to 35% as rating for handling prices hits a record -46
www.gelliottmorris.com/p/2026-04-21-april-strength-in-num… -
G Elliott Morris
@gelliottmorris
hint at tomorrow's poll result...
-
G Elliott Morris
@gelliottmorris
new poll on trump, prices, Iran, impeachment, and the democratic party's brand problem coming out this week starting tomorrow at gelliottmorris.com. sign up to get results in your inbox!
-
G Elliott Morris
@gelliottmorris
👀 👀
www.the-downballot.com/p/the-downballots-1q-2026-senate-and -
G Elliott Morris
@gelliottmorris
Cool that Strength In Numbers is part of The Discourse on things talkingpointsmemo.com/edblog/thoughts-on-a-new-civic-cont…
-
G Elliott Morris
@gelliottmorris
While approval is a polarized metric that changes slowly among Trump's supporters, “soft” measures of Trump’s popularity are eroding among Republicans.
New at 50+1:
blog.fiftyplusone.news/p/soft-measures-of-trumps-popularity -
G Elliott Morris
@gelliottmorris
I built a model of consumer sentiment that’s backed by new research showing consumers react more negatively to economic shocks in good times vs bad.
It tells us the same thing polls do: People are really upset about the high price of food, housing, etc.
www.gelliottmorris.com/p/2026-04-17-cotw-more-on-explaini… -
G Elliott Morris
@gelliottmorris
obviously what democrats really need to be doing to win elections in suburban seats is to nominate moderates who vote with republicans on immigration and identity issues
-
G Elliott Morris
@gelliottmorris
steyer-mentum?? fiftyplusone.news/polls/governor/nonpartisan-primary/cali…
-
G Elliott Morris
@gelliottmorris
Why is consumer sentiment lower than you'd predict based on unemployment, inflation, and the stock market alone? Because none of those variables tap into price levels, which voters say they care most about. Multiple measures of price levels beat more naive models of sentiment
G Elliott Morris
@gelliottmorris
The index of consumer sentiment isn't broken; models that try to predict it just don't have the right input variables. I fix that and find that, yes, high nominal price levels explain why the vibes are so off. It's the prices, stupid
www.gelliottmorris.com/p/2026-04-14-economy-sentiment-its… -
G Elliott Morris
@gelliottmorris
The index of consumer sentiment isn't broken; models that try to predict it just don't have the right input variables. I fix that and find that, yes, high nominal price levels explain why the vibes are so off. It's the prices, stupid
www.gelliottmorris.com/p/2026-04-14-economy-sentiment-its… -
G Elliott Morris
@gelliottmorris
back on my bullshit (vaugeposting predicted consumer sentiment charts)
-
G Elliott Morris
@gelliottmorris
Consumer sentiment is at the lowest level ever. This is not what you’d predict given inflation, GDP, spending data. But one variable from UMich sheds some light on the subject: the percent of people who are saying prices are nominally high is at an ATH.
www.gelliottmorris.com/p/2026-04-12-sunday-roundup -
G Elliott Morris
@gelliottmorris
Here's a bunch of charts on presidential approval and gas prices.
That's it, that's the piece!
(Plus some commentary on how a ceasefire with Iran could impact Trump's numbers.)
www.gelliottmorris.com/p/2026-04-10-gas-price-approval-cotw -
G Elliott Morris
@gelliottmorris
Joined The Downballot podcast this week to talk about Trump's approval in GOP districts and Senate seats open.substack.com/pub/thedownballot/p/the-progressive-lan…
-
G Elliott Morris
@gelliottmorris
Nice post from Arin. Raising the minimum wage in Florida, Arizona, Nebraska and Missouri led to higher wages for workers and had no measurable effect on employment
arindube.substack.com/p/these-republican-leaning-states-went -
G Elliott Morris
@gelliottmorris
Lotta new stuff rolling out at 50+1 these days
FiftyPlusOne.news
@50plus1news
New polling averages up at fiftyplusone.news
-
G Elliott Morris
@gelliottmorris
Kyle Kondik on the GOP’s Trump problem
centerforpolitics.org/crystalball/republicans-need-their-… -
G Elliott Morris
@gelliottmorris
Live podcast recording Thursday at 2PM ET covering all the new polling on Trump, Iran, and the midterms — including my new analysis of Trump at the state and congressional district levels substack.com/@gelliottmorris/note/c-240341917?r=a9pj
-
G Elliott Morris
@gelliottmorris
If I'm a GOP member of Congress and I'm in this table, I would be worried about re-election — especially if Trump keeps doing what he's doing and I keep voting to let him www.gelliottmorris.com/p/2026-04-07-trump-approval-gop-ho…
-
G Elliott Morris
@gelliottmorris
The Iran ceasefire deal is pretty bad for Trump and America, as others have pointed out. I have seen some people arguing that voters wanted a ceasefire, so Trump will get a ratings bounce from this. But I don't think you get a bounce for ending an unpopular war that you started
-
G Elliott Morris
@gelliottmorris
NEW: Donald Trump is underwater with registered voters in **135** GOP-held House and Senate seats. His approval is <45% in 44, including 10 senators. Should be a warning for Republicans in Congress weighing electoral cost of Trump's war and crazy tweets.
www.gelliottmorris.com/p/2026-04-07-trump-approval-gop-ho… -
G Elliott Morris
@gelliottmorris
I guess now would be a good time for me to publish my data on Trump approval by congressional district 
-
G Elliott Morris
@gelliottmorris
line go down
FiftyPlusOne.news
@50plus1news
We have just published an average of Trump's approval on handling foreign policy, in addition to specific trends for Iran and Venezuela. The Iran trend is one of, if not the, steepest approval drops of Trump's term in office
Data and charts: blog.fiftyplusone.news/p/trump-is-losing-the-public-over-… -
G Elliott Morris
@gelliottmorris
Trump's posts on Iran have gotten increasingly bad over the last wk (and that's really saying something), from "see you in hell" to "a whole civilization will die tonight." Where are Republicans in Congress? The simplest approx is that they are staying silent bc they support this
-
G Elliott Morris
@gelliottmorris
New deep dive piece on 2 new big studies on ideology and moderation, including a data exclusive from Blue Rose Research. Short version: Only 8% of "moderates" say they want a moderate party. And when candidates do move to the middle, fx are often negative www.gelliottmorris.com/p/2026-04-07-only-8-percent-modera…
-
G Elliott Morris
@gelliottmorris
New podcast is also on YouTube for people who prefer to watch there www.youtube.com/watch?v=XqClYRdzAHc
-
G Elliott Morris
@gelliottmorris
New podcast with Joy Wilke of the data analytics company BlueLabs covering what new focus group interviews in Michigan tell us about economic anxiety, anti-system attitudes, and faith (or lack thereof) in the parties today. Great convo with good qual data www.gelliottmorris.com/p/podcast-joy-wilke-focus-groups-v…
-
G Elliott Morris
@gelliottmorris
Join Mike Podhorzer and me live at 1pm as we discuss the latest polls on Iran, the midterms, regretful 2024 Trump voters and more
open.substack.com/live-stream/154186?r=103bk -
G Elliott Morris
@gelliottmorris
When we polled 2024 Trump voters who now disapprove of the president and asked them why, the top three issues were prices/the economy, immigration, and his personality/conduct in office. We have come to expect insane tweets like this, but they are still taxing with normies
-
G Elliott Morris
@gelliottmorris
New piece: Why Democrats are suddenly winning back the left — and the "double-haters"
www.gelliottmorris.com/p/2026-04-05-sunday-roundup -
G Elliott Morris
@gelliottmorris
FiftyPlusOne.news
@50plus1news
Polls shows how Americans soured on Pam Bondi over her short 14-month stint as attorney general blog.fiftyplusone.news/p/polls-show-how-america-soured-on
-
G Elliott Morris
@gelliottmorris
This might be the largest party lead I've ever seen among the double haters
www.cnn.com/2026/04/03/politics/cnn-poll-double-haters-de… -
G Elliott Morris
@gelliottmorris
The six big events that have dragged down Donald Trump's approval rating www.gelliottmorris.com/p/2026-04-03-cotw-trump-approval-c…
-
G Elliott Morris
@gelliottmorris
This chart, from YouGov, shows non-MAGA Republicans (about half of the party) moving 50 points, from +25 to -23, against Trump’s war in Iran over the last month.
I wrote two weeks ago why this was the group to watch:
www.gelliottmorris.com/p/forget-the-maga-split-on-iran-su… -
G Elliott Morris
@gelliottmorris
Trump is 20+ points underwater on approval. So why are Dems lagging their 2018 generic ballot polls?
The answer is the Dem party is currently losing on the GB among Trump’s “soft disapprovers.” I profile this group of key “cross-pressured” voters here:
www.gelliottmorris.com/p/trump-is-20-points-underwater-so -
G Elliott Morris
@gelliottmorris
trump rapidly heading for an approval rating among all adults that starts with `36` fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
Trump's job approval rating is hitting all-time lows (-21 on avg) across polling firms this week. Yet Democrats are "only" +6 on the House generic ballot. What explains the difference? I take a deep dive into exclusive raw polling data to find out: www.gelliottmorris.com/p/trump-is-20-points-underwater-so
-
G Elliott Morris
@gelliottmorris
Our latest Strength In Numbers / verasight.io poll was cited on CNN yesterday alongside three major pollsters. That’s more evidence that independent media can be a real business model for covering public opinion well.
More poll results: www.gelliottmorris.com/p/poll -
G Elliott Morris
@gelliottmorris
Trump hits new approval low among all adults, rounding to -21 for first time
 fiftyplusone.news/polls/approval/president -
G Elliott Morris
@gelliottmorris
Yeah, that’s… pretty bad!
-
G Elliott Morris
@gelliottmorris
tis the season
-
G Elliott Morris
@gelliottmorris
One reason Democrats are lagging Trump’s disapproval (-20) on the generic ballot (D+5) is that there are a lot of voters who say they somewhat disapprove of Trump but are undecided for the midterms. That alone costs Dems about 4 points on the GB
www.gelliottmorris.com/p/bonus-why-the-gop-midterms-conve… -
G Elliott Morris
@gelliottmorris
Our new poll finds 1 out of 8 Trump voters regrets how they voted in 2024, and Harris would win a hypothetical redo by almost 3 points www.gelliottmorris.com/p/2026-03-27-buyers-remorse-trump-…
-
G Elliott Morris
@gelliottmorris
new data out tomorrow morning at gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
Live Strength In Numbers podcast recording today at 2:00 Eastern. David and I will be talking about AI-generated polls and going deep on the predictive value — and values — of prediction markets. Join us live, paying subs can ask Qs!
open.substack.com/live-stream/145709 -
G Elliott Morris
@gelliottmorris
From our new poll: ICE agents are the least trusted law enforcement personnel in the country www.gelliottmorris.com/p/2026-03-26-flash-poll-ice-airpor…
-
G Elliott Morris
@gelliottmorris
James Carville: “6 months ago I didn’t know who [Elliott Morris] was. Now everybody’s asking me to get this guy on the podcast!”
podcasts.apple.com/us/podcast/teleport-us-to-the-waffle-h… -
G Elliott Morris
@gelliottmorris
NEW POLL:
- 72% oppose ground troops in Iran — including a majority of Republicans.
- 53% of adults oppose ICE at airports
- 59% disapprove of Trump rejecting the DHS funding deal
Full results from my new Strength In Numbers / @Verasight_data survey:
www.gelliottmorris.com/p/2026-03-26-flash-poll-ice-airpor… -
G Elliott Morris
@gelliottmorris
over the past 6 months strength in numbers has been gaining an average of 150 new readers per day. covering politics/public opinion isn't always fun but i feel truly blessed to be able make this a functioning career. and big things to come. thank you all who made this possible!
-
G Elliott Morris
@gelliottmorris
new poll 6am tomorrow www.gelliottmorris.com/p/2026-03-26-flash-poll-ice-airpor…
-
G Elliott Morris
@gelliottmorris
Just got data back for a special SIN/Verasight.io survey on the ICE airport deployment, Trump's rejection of the Senate's DHS budget deal, opposition to boots on the ground in Iran, and support for a ceasefire. Results tomorrow morning at www.gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
Tomorrow's live recording for the Strength In Numbers Podcast will cover some exclusive crosstabs from our new monthly poll, plus why you should trust real polls over prediction markets and AI-based predictions. You can submit questions early here!
open.substack.com/live-stream/145709 -
G Elliott Morris
@gelliottmorris
trump hits new all time low on average approval for handling prices
www.gelliottmorris.com/data/ -
G Elliott Morris reposted
Laura Rozen
@lrozen
From new @gelliottmorris Strength in Numbers poll:
Trump approval 37%, disapprove 60%.
On Iran war: 51% think will make Americans less safe, double the 26% think will make safer
Trump approval on prices/inflation fell to net -39, the worst single issue rating
Only 10% think things going well in America, 52% poorly.
open.substack.com/pub/gelliottmorris/p/2026-03-24-march-s… -
G Elliott Morris
@gelliottmorris
A lot of what Trump is doing is incredibly unpopular. The GOP is heading for a wipeout in Nov. Yet the gov keeps doubling down on the most toxic parts of his presidency to appeal to extremist audiences on social media. I call it the made-for-X presidency.
newrepublic.com/article/208152/trump-tirades-iran-take-ug… -
G Elliott Morris
@gelliottmorris
Trump is now flirting with a job approval rating among all adults that starts with the number 37, per our average
fiftyplusone.news/polls/approval/president -
G Elliott Morris
@gelliottmorris
New poll: Trump’s approval on handling prices, the border hits new low. 51% of adults say the war in Iran makes Americans less safe www.gelliottmorris.com/p/2026-03-24-march-sin-verasight-p…
-
G Elliott Morris
@gelliottmorris
Fresh polling data on Trump, the war in Iran, economic sentiment, the midterms and more coming out tomorrow from our latest monthly Strength In Numbers/Verasight.io survey. Get it in your inbox: gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
Trump dramatically increasing the visibility of ICE agents, a group with a double digit negative favorability rating, is a real galaxy brain move
-
G Elliott Morris
@gelliottmorris
I thought this segment from Enten was bad use of polling www.gelliottmorris.com/p/2026-03-22-sunday-roundup
-
G Elliott Morris
@gelliottmorris
FiftyPlusOne.news
@50plus1news
New Friday post: How Americans feel about prediction markets blog.fiftyplusone.news/p/how-americans-feel-about-prediction
-
G Elliott Morris
@gelliottmorris
New poll: 58% of adults say the war in Iran is a bad use of taxpayer dollars, and 61% — including 1 out of 3 Republicans — say they would oppose the war if it raises gas prices (it has) www.gelliottmorris.com/p/2026-03-20-cotw-sin-v-poll-iran-…
-
G Elliott Morris
@gelliottmorris
New polling on how Americans view the cost of the war in Iran coming out tomorrow morning @ Strength In Numbers. Get it in your inbox: gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
to put this in the language certain people On Here will understand: doesn't kat abu end up with a pretty high WAR score?
G Elliott Morris
@gelliottmorris
.@mattyglesias détente?
-
G Elliott Morris
@gelliottmorris
.@mattyglesias détente?
-
G Elliott Morris
@gelliottmorris
Pundits who appear to spend all day on social media are declaring that being ‘Very Online’ doesn’t work in politics. The irony. Meanwhile this 26-yo candidate w no experience nearly won a 15-candidate race… New post on what we can actually learn from IL09
www.gelliottmorris.com/p/what-can-we-learn-from-a-close-loss -
G Elliott Morris
@gelliottmorris
the temptation to tweet a dismissive hot take out to perform to your followers how much you hate lefty progressives seriously hampers one's ability to conduct serious analysis of politics
www.gelliottmorris.com/p/what-can-we-learn-from-a-close-loss -
G Elliott Morris
@gelliottmorris
On the other hand, Abughazaleh came in 2nd, and the incumbent mayor only won by 4 points (with <30% of the vote). I agree it's not surprising, but given the close race, maybe there's more to the story than just "very online leftist lost?" Would this have worked in 2022, or 2016?
-
G Elliott Morris
@gelliottmorris
FiftyPlusOne.news
@50plus1news
NEW crosstab-level polling averages from 50+1! Trump is barely above water among white voters without a college degree. But the group still plans to vote for Republicans in November. blog.fiftyplusone.news/p/trump-is-barely-above-water-among
-
G Elliott Morris
@gelliottmorris
Donald Trump's job approval is 34 points lower than his 2024 vote margin among working-class Hispanics — and working-class whites now on net disapprove of how he's handling his job as president, per new Strength In Numbers/Verasight polling
www.gelliottmorris.com/p/workers-trump-economy-2026-03-17 -
G Elliott Morris
@gelliottmorris
this chart will publish tomorrow at 7:00 am eastern www.gelliottmorris.com/p/workers-trump-economy-2026-03-17
-
G Elliott Morris
@gelliottmorris
trump is actually a democratic sleeper agent trying to get james talarico, mary peltola, and sherrod brown elected to the senate
-
G Elliott Morris
@gelliottmorris
“We want to change the rules of the next election so Democrats don’t win anymore”
-
G Elliott Morris
@gelliottmorris
Forget the "MAGA split" (or lack thereof) on Iran. It's the independents and soft partisans that matter.
www.gelliottmorris.com/p/forget-the-maga-split-on-iran-su… -
G Elliott Morris
@gelliottmorris
Strength In Numbers is looking for a smart part-time survey research assistant (or a few)
www.gelliottmorris.com/p/strength-in-numbers-is-looking-for -
G Elliott Morris
@gelliottmorris
73% of Americans say they are worried about the financial cost of Trump's war in Iran, including 79% of political independents. Nearly 9 in 10 Americans are worried about the lives of military personnel
substack.com/@gelliottmorris/note/c-227324965?r=a9pj -
G Elliott Morris
@gelliottmorris
Why hasn't Trump received a "rally around the flag" bump in his approval rating?
I made every presidential approval chart you could ever want to explain www.gelliottmorris.com/p/i-graphed-every-presidents-appro… -
G Elliott Morris
@gelliottmorris
This article claims the gap in Trump's approval between polls of voters and adults is a reason to think the GOP will outperform in the midterms.
The problem is that approval is only weakly predictive of outcomes, and a diff of 1 pt is meaningless. See: projects.gelliottmorris.com/midterms-approval/?approval=-… -
G Elliott Morris
@gelliottmorris
We have generated crosstab-level averages for the House generic ballot:
FiftyPlusOne.news
@50plus1news
New: How are key demographics moving ahead of the 2026 midterms? We model the generic ballot by demographic group, using crosstabs from every 2026 generic ballot poll, to see how the race is shaping up.
blog.fiftyplusone.news/p/how-different-demographic-groups… -
G Elliott Morris
@gelliottmorris
New: The groups Trump made the biggest inroads with in 2024 are the ones most likely to disapprove of how he's doing as POTUS. That includes Black, Latino, working class, and young voters.
So much for the new multi-racial, working-class GOP?
www.gelliottmorris.com/p/trump-borrowed-affordability-vot… -
G Elliott Morris
@gelliottmorris
Subscribe to Strength In Numbers today and you too can understand public opinion 11 months before the White House catches on
⬇️⬇️
gelliottmorris.com/subscribe -
G Elliott Morris
@gelliottmorris
NEW polling data on Trump voters: Overall 15% now disapprove of his job performance, including 19% of Hispanics, 22% of young voters, and 33% of Black 2024 Trump voters.
Affordability voters have shifted left on Trump and the House generic ballot:
www.gelliottmorris.com/p/trump-borrowed-affordability-vot… -
G Elliott Morris
@gelliottmorris
this article (and the data behind it) will publish tomorrow at 7:00 am www.gelliottmorris.com/p/trump-borrowed-affordability-vot…
-
G Elliott Morris
@gelliottmorris
Notable favorability rating numbers for Senate candidates in Texas, per new PPP poll:
- James Talarico: +6 net (41% favorable)
- John Cornyn: –28 net (53% unfavorable)
- Ken Paxton: –24 net (54% unfavorable)
senatemajority.com/wp-content/uploads/signal-2026-03-06-1… -
G Elliott Morris
@gelliottmorris
america has entered "woke" 2.0 www.gelliottmorris.com/p/64-of-americans-want-to-stop-cha…
-
G Elliott Morris
@gelliottmorris
Strength In Numbers coverage in Axios this morning: www.axios.com/2026/03/07/trump-iran-risks-economy
-
G Elliott Morris
@gelliottmorris
By moving right on immigration & other key issues, and generally coming off as incompetent (esp on cost of living), UK Labour is losing vote share to left-wing parties AND to right-leaning ones. This is what happens when a party appeases its opponents & loses its core identity
Steve Akehurst
@SteveAkehurst
Ok, so why are they disaffected with Labour?
Some of it is similar to other groups - the slow pace of change.
But left defection is more distinctly values based than right. Rather than just one issue, there's a sense that Labour has been more right wing than they expected.
G Elliott Morris
@gelliottmorris
This plot shows the highest-leverage policies that win back defectors on both the left and right:
- Increase pay for public sector
- Decrease cost of bus and train tickets
- Wealth tax on billionaires
and the winnable left bloc is bigger than the right
Steve Akehurst
@SteveAkehurst
In our innovative RCT experiment, these policies worked best.
Notice there is *some* ‘lib’ stuff in there - EU, climate. Works for left while not alienating right.
But the key: affordability populism unites left & right! -
G Elliott Morris
@gelliottmorris
FiftyPlusOne.news
@50plus1news
What divides MAGA and non-MAGA Republicans when it comes to foreign policy and Iran? For 50+1's weekly polling roundup, The Trendline, we took a look at the numbers:
blog.fiftyplusone.news/p/what-divides-maga-and-non-maga-r… -
G Elliott Morris
@gelliottmorris
The U.S. lost 92,000 jobs in February.
According to our polling, 24% of Americans say Trump has made the economy better since becoming president, vs 53% who say he's made things worse.
Voters say 2:1 Trump isn't focused enough on things they care about.
www.gelliottmorris.com/p/53-of-americans-say-trump-has-made -
G Elliott Morris
@gelliottmorris
All the polls on the U.S. war in Iran so far
www.gelliottmorris.com/p/polls-us-iran-attack-2026-03-06
If you ignore people who say they “don’t know” how they feel about the war, disapproval of Trump’s actions is already at 56%. That is higher than the resistance to Vietnam in 1971. -
G Elliott Morris
@gelliottmorris
Since everyone is talking about new voters, I pulled up the Virginia voter file to check how many 2025 general election voters were non-2024 voters. Per L2, that number is 200k, or about 6% of all ballots cast in the race. These 200k were modeled at 60% Dem and 20% Rep
-
G Elliott Morris
@gelliottmorris
59% of Americans Oppose the Military Action in Iran www.nytimes.com/2026/03/04/us/politics/iran-us-war-milita…
-
G Elliott Morris
@gelliottmorris
Six data-driven reasons Texas could actually go blue in 2026
www.gelliottmorris.com/p/six-data-driven-reasons-texas-co… -
G Elliott Morris
@gelliottmorris
Put a little spreadsheet together. This is Talarico margin vs Hegar in 2020. Tracking pretty close on average but lots of variance (small counties). If you weight counties by number of votes expected, Talarico is running about 3 points ahead of Hagar.
docs.google.com/spreadsheets/d/1noAYOOOwP-U0zR1nsN1g-Wrix… -
G Elliott Morris
@gelliottmorris
looks cool. methods explainer? conformal prediction? votehub.com/2026/03/03/introducing-votehub-wincast/
-
G Elliott Morris
@gelliottmorris
Every time I read a “x LLM is better at doing my job than I am” article, it turns out the person writing the piece is at best 50th percentile for their job. My theory is these people are trying to position themselves as AI pros because they know they’re going to get fired soon
G Elliott Morris
@gelliottmorris
"and guess what? surprise! this whole post was written by AI! bet you couldn't even tell."
we could. and the fact that _you_ couldn't speaks volumes... -
G Elliott Morris reposted
FiftyPlusOne.news
@50plus1news
Our final polling averages in Texas show a very close race on the Democratic side, and a likely runoff on the Republican side.
ICYMI, our analysis of these races, and primary polling in general, from last week: substack.com/home/post/p-189066848 -
G Elliott Morris
@gelliottmorris
The polling in this Texas Senate race has been wack. Maybe the highest ever standard deviation I've seen for a contest, even higher than NY Mayor last year. Someone is gonna be really wrong fiftyplusone.news/polls/senate/democratic-primary/texas
-
G Elliott Morris
@gelliottmorris
New post!
Many Americans hold contradictory opinions on the same policies. Question wording effects can swing opinions by 20+ points on immigration, the budget, and transgender rights.
That's a problem for people who interpret polls for a living.
www.gelliottmorris.com/p/many-americans-hold-contradictor… -
G Elliott Morris
@gelliottmorris
YouGov's state polling has been very good recently (nailed the NJ and VA gov races last year). Primaries are a different beast, but worth indexing on this new poll of TX. Has Talarico +13 and Paxton +4
yougovamerica.substack.com/p/polls-on-the-texas-primaries… -
G Elliott Morris
@gelliottmorris
Opinion on the U.S. strikes on Iran has been pretty stable as more people are learning about Saturday's events. There has been no "rally around the flag" bump for POTUS, and the strikes are much less popular than previous wars when they started www.gelliottmorris.com/p/polls-trump-iran-2026-03-01
-
G Elliott Morris
@gelliottmorris
tough to come up with a better story about the ignorance and hubris of mankind than “b-list tv news host becomes Secretary of Defense,” but then we get the story “b-list tv news host becomes Secy. of Defense, plugs AI system directly into military systems, has it select targets”
-
G Elliott Morris
@gelliottmorris
Like for capturing Maduro in Venezuela, I think polls will eventually show voters see the killing of the Ayatollah as a win, but inside the broader context of executive overreach and esp Trump breaking his promise not to start a foreign wars,it will come out net bad for POTUS
-
G Elliott Morris
@gelliottmorris
Publishing my normal Sunday polling article early (for obvious reasons):
Trump starts a war with Iran that few Americans support.
Overall support for Trump in early post-strike polling is below that for Afghanistan and Iraq, including among Republicans.
www.gelliottmorris.com/p/polls-trump-iran-2026-03-01
G Elliott Morris
@gelliottmorris
*only 10% for Dems on Iran. Correct in article, fixed chart below
-
G Elliott Morris
@gelliottmorris
FYI, polling on Iran: blog.fiftyplusone.news/p/voter-id-trump-sotu-trendline
-
G Elliott Morris
@gelliottmorris
Voters don’t think the Democratic Party is too progressive, they think it’s weak and ineffective
www.gelliottmorris.com/p/new-poll-democrats-real-problem-… -
G Elliott Morris
@gelliottmorris
Results from our new poll conducted completely after Trump's SOTU
- 57% say Trump is focused on the wrong issues
- 38% say Trump spoke on immigration issues the most Tuesday night. Only 12% say immigration is their MIP.
- Trump's SOTU pivot did not work
www.gelliottmorris.com/p/flash-poll-post-sotu-voters-say-… -
G Elliott Morris
@gelliottmorris
Update: My article this AM explored what would happen if you moved all voters toward the "Dems strong" and "moderate" sides equally. But a better counterfactual would account for the fact more voters see Dems as weak than extreme — so strength moves more voters. Here's that
G Elliott Morris
@gelliottmorris
Of course, this doesn't make it any easier to answer the hard question of how you make the Dems look tougher. But in my mind, this strengthens the case for thinking hard about things to try, esp relative to how much time/money orgs are spending punching left to appear "moderate"
G Elliott Morris
@gelliottmorris
A methods explainer is here: www.gelliottmorris.com/i/189329759/an-even-more-robust-me…
-
G Elliott Morris
@gelliottmorris
70% of voters say the Dem Party is too weak .
40% say it is too progressive .
Too weak is more predictive of vote than too prog.
If I were a Dem. trying to win in 2028, I would focus more of my energy on figuring out how to bring down that 70%.
www.gelliottmorris.com/p/new-poll-democrats-real-problem-…
G Elliott Morris
@gelliottmorris
The other interpretation of this article is that Democrats already won the comparative moderation fight vs the Republicans (take the W!!) and need to focus on their new weakness. This is why people who make their millions by selling ideo. moderation are so angry at alt voices
-
G Elliott Morris
@gelliottmorris
Matt, I answered this in the article. Dems are currently already seen as the more moderate party, but not the stronger one. Continuing to punch left does nothing to help the party on the axis it's currently losing voters on. And current proposals for moderation are coded as weak
-
G Elliott Morris
@gelliottmorris
The big question for Democrats is not binary, it's a question of emphasis.
Adam raised tens of millions in donor capital for a new think tank to "curb the influence of liberal groups." Center of gravity in elite Dem circles & media is clearly aligned with this vs other visions.
But the data suggests Dems don't have an extremism problem, but a strength problem. So I think strength should be a larger part of the conversation. Biggest thing people say in our poll is that they don't know what the Dems stand for.
Does "shut up about climate change" and "don't raise the salience of immigration" and "reform and retrain" read as strength? Not according to the actual survey data.
Adam Jentleson
@AJentleson
A purely objective read of this data shows that the ideal approach for Democrats is to be strong and moderate. That’s not my interpretation it’s just what the data clearly shows. Yet the author is doing gymnastics to argue something different. What are we doing here?
G Elliott Morris
@gelliottmorris
The entire Democratic strategy debate — "moderate" or move left? fight or compromise? — is based on a misreading of data. Voters don't think Dems are too progressive; they think Dems are weak. The actual math on what to do about this isn't even close!
New: www.gelliottmorris.com/p/new-poll-democrats-real-problem-… -
G Elliott Morris
@gelliottmorris
The entire Democratic strategy debate — "moderate" or move left? fight or compromise? — is based on a misreading of data. Voters don't think Dems are too progressive; they think Dems are weak. The actual math on what to do about this isn't even close!
New: www.gelliottmorris.com/p/new-poll-democrats-real-problem-… -
G Elliott Morris
@gelliottmorris
Join us on an adventure into the data mines
FiftyPlusOne.news
@50plus1news
New: How do we choose which polls we include in our aggregates? Join Head of Research @marywitha4 on a journey to try to verify a pollster is legit. (Spoiler alert: she failed.)
open.substack.com/pub/fiftyplusone/p/the-curious-case-of-… -
G Elliott Morris reposted
FiftyPlusOne.news
@50plus1news
We are picking up some strong momentum for TX State Rep. James Talarico in our average of the state's Democratic primary for U.S. Senate — but there's still a lot of uncertainty in the polls (most of which are low quality or campaign internals). fiftyplusone.news/polls/senate/democratic-primary/texas
-
G Elliott Morris
@gelliottmorris
NEW POLL on structural reforms: U.S. adults support 18-yr term limits for Supreme Court justices by a 50-point margin (GOP is +34 in favor), favor statehood for Puerto Rico and limits on pardons by POTUS, and are split on DC statehood, packing the Court:
www.gelliottmorris.com/p/two-thirds-of-americans-want-term -
G Elliott Morris
@gelliottmorris
sneak peek at new polling coming out tomorrow morning at gelliottmorris.com
-
G Elliott Morris
@gelliottmorris
this is why, where possible, you want to weight by a past vote variables that was *recorded as close as possible to the last election* (which is what we do for the Strength In Numbers/Verasight survey). shoutout to the OG yougov.com/en-us/articles/16873-beware-phantom-swings-why…
Jonathan Robinson
@jon_m_rob
Neat @lxeagle17 in @TheArgumentMag this morning, validating their in-survey 2024 vote recall question w/ @verasight_data previous survey 2024 vote recall.
~6% of "recalled" Trump voters "disagree" w/ their past response vs. ~3.5% of Harris voters. A slight partisan bias. -
G Elliott Morris
@gelliottmorris
graham platner, the electability moderate? fiftyplusone.news/polls/senate/general/maine
G Elliott Morris
@gelliottmorris
lots of evidence of the influence of anti-system bias on voter psychology in this poll (easier to detect in primaries where party polarization is not a factor. more: www.gelliottmorris.com/p/meet-americas-new-swing-voter-the)
-
G Elliott Morris reposted
Nathaniel Rakich
@baseballot
I think this is an important point. The Dem primary in #txsen, in particular, is basically one big shruggie emoji.
FiftyPlusOne.news
@50plus1news
New piece looking at a big question ahead of next Tuesday's Senate primary elections in TX: How reliable are statewide primary polls? (Not very!) blog.fiftyplusone.news/p/2026-texas-senate-primary-polls?…
-
G Elliott Morris
@gelliottmorris
To be clear, what Rufo is talking about here is not "public opinion," but rather the opinions of a small group of terminally online weirdos (I mean this statistically — they are very rare in the population as a whole) that are overrepresented on X — and that he radicalized!!
G Elliott Morris
@gelliottmorris
chris rufo trying to figure out why the right lost its collective mind
-
G Elliott Morris
@gelliottmorris
Political journalism needs fewer "takes" and more analytical legwork www.gelliottmorris.com/p/political-journalism-needs-fewer
-
G Elliott Morris
@gelliottmorris
New poll: 42% of adults say they would spend a hypothetical $75 billion surplus on reducing housing, health care, & food costs, vs 5% who say immigration enforcement
Voters also say they'd want to pause Trump's deportations if they were causing inflation.
www.gelliottmorris.com/p/new-poll-voters-would-rather-pause
G Elliott Morris
@gelliottmorris
sorry, typo, that should say 52% (wrote this post at 2am)
-
G Elliott Morris
@gelliottmorris
New article over at 50+1: How reliable are Senate primary polls? (Not very!) open.substack.com/pub/fiftyplusone/p/2026-texas-senate-pr…
-
G Elliott Morris
@gelliottmorris
Trump spent all of 2 minutes talking about affordability during his SOTU address, according to NBC News — the same amount of time he spent talking about the performance of the U.S. men's olympic hockey team substack.com/profile/479143-g-elliott-morris/note/c-21937…
-
G Elliott Morris
@gelliottmorris
So Trump is going with the "lie about inflation statistics and insist conditions are good while Americans are crying out in disapproval" strategy, huh? ask president joe biden how successful that will be www.gelliottmorris.com/p/ahead-of-state-of-the-union-address
-
G Elliott Morris
@gelliottmorris
At his first SOTU address, Donald Trump's approval rating has hit a new second-term low of 37% in our polling, and he no longer has a positive approval rating on any major issue. www.gelliottmorris.com/p/ahead-of-state-of-the-union-address
-
G Elliott Morris
@gelliottmorris
If anyone is going to be tracking issue topics of tonight's SOTU, let me know. I'll be looking for data that says POTUS spent x% of his time on immigration, y% on prices, z on tariffs, etc etc. Will of course cite in return!
-
G Elliott Morris
@gelliottmorris
??? bro the constituency for lower electric bills is... everyone who pays for electricity
PoIiMath
@politicalmath
Is killing data centers a new left-wing cause? Seems like a weird and pointless movement with no obvious benefits to any specific left wing constituency. What am I missing?
-
G Elliott Morris
@gelliottmorris
No positive issue territory left for Trump, per our new Strength In Numbers/ @verasight_data survey.
POTUS plans to focus on the economy and tariffs in his SOTU. But he has a *lot* of work to do persuading Americans to take his side on these issues. www.gelliottmorris.com/p/ahead-of-state-of-the-union-address -
G Elliott Morris
@gelliottmorris
This chart seems to suggest that among swing voters, inflation moved 24x as many voters as generic concerns about Kamala Harris being too liberal
-
G Elliott Morris
@gelliottmorris
NEW Strength In Numbers/Verasight poll:
Ahead of SOTU, Trump's approval falls to new low of 37% — & just 20% among independents. Issue approval on immigration and deportations falls to new lows of -15, and voters say admin is spending on the wrong issues.
www.gelliottmorris.com/p/ahead-of-state-of-the-union-address
G Elliott Morris
@gelliottmorris
More from my poll with @verasight_data in tomorrow's newsletter. (Trying out a new editorial strategy for these surveys where results get split up into different, topic-specific articles instead of running one big mega-release.)
-
G Elliott Morris reposted
Matt Friedman
@MattFriedmanNJ
Sure, Insider Advantage is the huge outlier in measuring Trump's approval rating. But consider this: That same pollster only missed the margin in the NJ gov race by 13 points.
-
G Elliott Morris
@gelliottmorris
New Strength In Numbers/ @verasight_data polling out tomorrow in advance of Trump's SOTU address. Will have new tracking data incl. generic ballot, Trump job approval, party trust + a new approach to measuring what people mean when they say a party is in/out of touch
G Elliott Morris
@gelliottmorris
get it in your inbox: gelliottmorris.com/subscribe/
-
G Elliott Morris reposted
Jacob Rubashkin
@JacobRubashkin
A lot of money for very little movement
AdImpact Politics
@AdImpact_Pol
#txpol: The #txsen primary is now the most expensive Senate primary on record, with $110.5M in ad spending and reservations.
Outside groups account for 70% of all spending and reservations in the race. -
G Elliott Morris
@gelliottmorris
I've got a new survey out today (part of private polling consulting work I do) that finds Trump's deportation agenda is eroding support for the GOP, and voters want the president to prioritize economics over deportations. W/ direct data on Trump defectors: morrispredictive.com/assets/surveys/k7Qm3xR9pL/memo.pdf
UNITE HERE
@unitehere
NEW: We've released a national survey from @gelliottmorris showing that the White House's immigration enforcement agenda is eroding support for Republicans ahead of the 2026 midterm elections.
Details ⬇️
morrispredictive.com/assets/surveys/k7Qm3xR9pL/memo.pdf -
G Elliott Morris
@gelliottmorris
in an age of anti-incumbent bias, no leader is naturally safe
Adam Wren
@adamwren
New: Indiana Gov. Mike Braun's statewide approval is at just 25 percent.
-
G Elliott Morris reposted
G Elliott Morris
@gelliottmorris
I wrote about something similar a while back. Gallup has asked "What party do you trust to handle your single most important issue?" (in so many words) since the 50s, and it has a perfect track record of predicting POTUS winners. Less so in midterm years, where incumbent party loses seats even when it has the overall issue trust lead.
Currently Dems lead on this question by 8 points www.gelliottmorris.com/p/poll -
Post not available here, view on X
G Elliott Morris
@gelliottmorris
I wrote about something similar a while back. Gallup has asked "What party do you trust to handle your single most important issue?" (in so many words) since the 50s, and it has a perfect track record of predicting POTUS winners. Less so in midterm years, where incumbent party loses seats even when it has the overall issue trust lead.
Currently Dems lead on this question by 8 points www.gelliottmorris.com/p/poll -
G Elliott Morris
@gelliottmorris
true that dems have some work to do, as a party brand, to lead on the issue trust questions.
but also true this isn't a straight-up measure of 'which party do you want in charge of X' (www.cambridge.org/us/universitypress/subjects/politics-in…)
and that the 4-pt R lead for immigration is lowest in a decade
Politics & Poll Tracker 📡
@PollTracker2024
Washington Post/ABC News/Ipsos poll | 2/12-2/17
Who do you trust to do a better job at handling…
The country’s main problems
🟥Trump 33%
🟦Democrats in Congress 31%
Neither 31%
—
Reducing the cost of living in the US
🟥Trump 32%
🟦Democrats in Congress 31%
Neither 33%
—
Immigration
🟥Trump 38%
🟦Democrats in Congress 34%
Neither 24%
G Elliott Morris
@gelliottmorris
so, picture is complicated. can't conclude from this that voters want the Reps in charge; public opinion is more complicated than that
-
G Elliott Morris
@gelliottmorris
Ahead of Trump’s State of the Union address, new polls released this weekend show voters disapprove of his presidency and immigration enforcement, and approve of new SCOTUS ruling rolling back his tariffs
today.yougov.com/politics/articles/54148-most-americans-a…
www.washingtonpost.com/politics/2026/02/22/trump-disappro… -
G Elliott Morris
@gelliottmorris
I’m glad my writing at Strength In Numbers is shaping the conversation on ideology & how different (non-)ideological appeals work w/ U.S. voters. Seeing beyond the left-right binary is essential, since most voters don’t think about politics like elites do link.motherjones.com/public/44256989
-
G Elliott Morris
@gelliottmorris
well i for one am shocked — *shocked* — that turning a media institution where readers disapprove of Trump by 50+ points into a mouthpiece for the administration has decreased readership and imperiled the finances of said institution. *shocked* substack.com/profile/479143-g-elliott-morris/note/c-21794…
-
G Elliott Morris
@gelliottmorris
It is important that legislators respond to public opinion. I also think that it’s important that what we call “public opinion” is measured carefully and with nuance. One thing we know about the public is opinions are complex and very susceptible to framing and priming issues
-
G Elliott Morris
@gelliottmorris
SCOTUS handed Trump a potential political win on tariffs. He could have ended an unpopular policy that is driving up inflation for the average American.
Instead, he decided to double down, attacking the Court and announcing new import taxes. FAFO www.gelliottmorris.com/p/trump-decides-hell-fafo-on-supreme -
G Elliott Morris
@gelliottmorris
If you spend any significant amount of time on the Elon Musk social media app, research suggests your brain is being cooked (www.forbes.com/sites/martinadilicosa/2026/02/18/xs-algori…)
There are good alternatives on the market (but you should probably just quit all social media for good)
www.gelliottmorris.com/p/you-should-quit-social-media-for -
G Elliott Morris reposted
Jacob Rubashkin
@JacobRubashkin
Four months and tens of millions of dollars later and the TX-Sen primary is essentially unchanged from the day Wesley Hunt got in the race
-
G Elliott Morris
@gelliottmorris
New piece
Sen. Cortez Masto told The Daily on Thursday Dems are fighting for ICE reforms in part because it helps them electorally — by changing what "immigration" means to voters.
Strength In Numbers has been writing about this since last April!
www.gelliottmorris.com/p/the-shutdown-helps-democrats-by-… -
G Elliott Morris
@gelliottmorris
Has anyone polled on Trump hanging these big portraits of himself all around DC? (DOJ is just the latest example)
-
G Elliott Morris
@gelliottmorris
Getting caught up The Argument Mag's polling of anti-trans policies that came out yesterday. Demsas defends her numbers guy by saying that their poll does not conclude Dems should moderate on trans rights. I agree — but isn't this *exactly* how he interpreted the polling data?
Post not available here, view on X
G Elliott Morris
@gelliottmorris
Now, with my pollster hat on, I think a polling article about "trans rights" would have made more of the data on non-discrimination laws for trans people. Isn't this what ppl are talking about when they talk about "rights"? Else, title should've been "anti-trans panic is real"
G Elliott Morris
@gelliottmorris
4/4 I think it's pretty obvious, given the conclusion, that the write-up of the data started with a certain prior in mind, and coverage of the data was used to justify that prior. Otherwise, author should have been able to acknowledge the nuances of opinion the first time around
-
G Elliott Morris
@gelliottmorris
noting for posterity that self-placed ideology here does not appear to be single-peaked
Tom Wood
@thomasjwood
Symbolic ideology and education, 1972-2024.
Ideological polarization increasingly interacts with education.
Data from @electionstudies CDF. -
G Elliott Morris
@gelliottmorris
ok, maybe brendan carr is a democratic sleeper argent
-
G Elliott Morris
@gelliottmorris
Donald Trump won in 2016 and 2024 because anti-system voters flocked to him.
Democrats need a strategy for competing with anti-system voters in 2028.
They should sit down and really think through how an AOC ticket might work.
www.gelliottmorris.com/p/meet-americas-new-swing-voter-the -
G Elliott Morris
@gelliottmorris
this article will publish tomorrow at 7:00 am (and i'm sure everyone will agree and no one will have anything negative to say about it!) www.gelliottmorris.com/p/meet-americas-new-swing-voter-the
-
G Elliott Morris
@gelliottmorris
OK, I'm collecting takes from people. Assume for this tweet that 2028 is Ocasio-Cortez vs Vance. Who wins and why?
-
G Elliott Morris
@gelliottmorris
The less voters knew, the more they liked Trump in 2024. Not Anymore
www.gelliottmorris.com/p/trump-lost-low-info-voters -
G Elliott Morris
@gelliottmorris
polls!
FiftyPlusOne.news
@50plus1news
Poll lovers! Big news over at FiftyPlusOne. Today, we're rolling out brand new features AND editorial content. Subscribers will get access to complete tables of polling data (not just the 5 most recent polls), as well as weekly polling roundups and articles.
-
G Elliott Morris reposted
Alexander Agadjanian
@A_agadjanian
Really nice graphic from @gelliottmorris on the trajectory of protests at comparable points in Trump term 1 vs term 2
G Elliott Morris
@gelliottmorris
Weekly data roundup for Feb. 15: New hard data show same pattern of intense anti-Trump backlash that polls do. Protests against Trump are happening at 4x the rate as 2017-18, and special election swings offer more data on the same point www.gelliottmorris.com/p/trump-backlash-hard-data
-
G Elliott Morris
@gelliottmorris
There are a lot of people out there using polling data to shroud their punditry about politics/electoral strategy in empiricism. I think it’s important to point out that the vast majority of them have never worked as pollsters or for a campaign. They have little domain knowledge
-
G Elliott Morris reposted
Natalie Jackson
@nataliemj10
"You see, the country has mood swings."
...
"I could explain it better, but I'd need charts and graphs and an easel." -
G Elliott Morris
@gelliottmorris
from a measurement POV, the party trust Q has little direct bearing on issue positions/policy preferences. party trust can be low for a variety of non-ideological reasons, like past history of leadership or message clarity. Dems have work to do, but this Q doesn’t tell us what
G Elliott Morris
@gelliottmorris
Also, Yglesias lecturing people on uncertainty when it comes to interpreting polling data is laughable. The dude is a walking overconfidence and selection bias machine
-
G Elliott Morris
@gelliottmorris
I thought this was a phenomenal read www.programmablemutter.com/p/the-median-voter-theorem-is-…
-
G Elliott Morris
@gelliottmorris
Weekly data roundup for Feb. 15: New hard data show same pattern of intense anti-Trump backlash that polls do. Protests against Trump are happening at 4x the rate as 2017-18, and special election swings offer more data on the same point www.gelliottmorris.com/p/trump-backlash-hard-data
-
G Elliott Morris
@gelliottmorris
This poll from CBS News/YouGov contradicts the popular claim that voters think Democrats are as ideologically extreme as Republicans
-
G Elliott Morris
@gelliottmorris
Just noting this because I have cited the trends a lot recently: Tracking data from Civiqs now shows a plurality of the public favors abolishing ICE, with 47% in support and 46% opposed. Independents are net +5 on abolish, up from -40 a year ago substack.com/profile/479143-g-elliott-morris/note/c-21410…
-
G Elliott Morris
@gelliottmorris
Political independents say 2:1 that Bad Bunny "better represents America" than Donald Trump www.yahoo.com/news/politics/article/poll-who-better-repre…
-
G Elliott Morris
@gelliottmorris
Brand new polling from Data for Progress, shared first with Strength In Numbers, shows voters back Dems in the DHS shutdown over ICE & overwhelmingly favor key reforms such as preventing mask-wearing (+23) and requiring judicial warrants for arrests (+34): www.gelliottmorris.com/p/democrats-have-the-leverage-on-the
-
G Elliott Morris
@gelliottmorris
Voters support another Democratic gerrymander that would never have existed without Trump‘s mid-decade redistricting push last year 
Adam Carlson
@admcrlsn
🚨 New @ZenithPolls poll among n=800
likely Dem primary voters in Maryland
(2/8-11) 🚨
🔵 70% support the state legidlature redrawing the state’s congressional map before November after hearing arguments for and against it
🔵 68% think the MD Senate should vote on the map the House of Delegates passed, which would flip one seat to Dems, giving them a 8D-0R map (the Dem-led Senate, led by Senate President @SenBillFerg, is currently refusing to hold a debate or a vote on it) -
G Elliott Morris
@gelliottmorris
thanks to @verasight_data for partnering on this poll!
G Elliott Morris
@gelliottmorris
low-info voters are very responsive to national conditions. which is they they voted for trump in 2024, and now why they're even more anti-trump on affordability than high-knowledge (dem-leaning) voters
www.gelliottmorris.com/p/trump-lost-low-info-voters
G Elliott Morris
@gelliottmorris
New poll: In 2024, low-engagement voters went for Trump over Harris by 11 percentage points. But now they disapprove of the way he's handling the presidency by 13. They have moved 25 points against Trump — 2x as large as the shift for high-knowledge voters www.gelliottmorris.com/p/trump-lost-low-info-voters
-
G Elliott Morris
@gelliottmorris
low-info voters are very responsive to national conditions. which is they they voted for trump in 2024, and now why they're even more anti-trump on affordability than high-knowledge (dem-leaning) voters
www.gelliottmorris.com/p/trump-lost-low-info-voters
G Elliott Morris
@gelliottmorris
New poll: In 2024, low-engagement voters went for Trump over Harris by 11 percentage points. But now they disapprove of the way he's handling the presidency by 13. They have moved 25 points against Trump — 2x as large as the shift for high-knowledge voters www.gelliottmorris.com/p/trump-lost-low-info-voters
-
G Elliott Morris
@gelliottmorris
New poll: In 2024, low-engagement voters went for Trump over Harris by 11 percentage points. But now they disapprove of the way he's handling the presidency by 13. They have moved 25 points against Trump — 2x as large as the shift for high-knowledge voters www.gelliottmorris.com/p/trump-lost-low-info-voters
-
G Elliott Morris
@gelliottmorris
trump hits a new approval low among adults today (rounding to -19 for the first time) fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
Bonus post coming for subscribers to Strength In Numbers tomorrow, looking at fresh polling on how high- and low-knowledge voters have reacted to Trump's presidency. Low-knowledge voters were key to Trump's 2024 victory, and whether they are still pro-Trump will be key to 2026/8.
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
“Cut ICE’s budget to restore federal health care funding” is an extremely popular message www.politico.com/f/?id=0000019c-4aa8-d3b5-addf-eaec955d0000
-
G Elliott Morris
@gelliottmorris
The NYT says Democratic proposals for reforming ICE are “facing blowback.”
But blowback from who? Not the people! Polls show support for most reforms exceeds 60%, with some ideas reaching as high as 94% of the public!
substack.com/@gelliottmorris/note/c-212791369?r=a9pj -
G Elliott Morris
@gelliottmorris
Bill Clinton's victory in 1992 is the canonical example that proves Dems win by moving to the center. There's only one problem: The data disagrees. Voters saw Clinton as more liberal than Dukakis, & simulations show even a solid progressive could have won www.gelliottmorris.com/p/what-the-democrats-can-actually-…
-
G Elliott Morris
@gelliottmorris
Democrats are up 4.6 among registered voters in our House generic ballot average, a new high for the 2026 cycle. (Their margin among likely voters is higher, and we will gradually fade in the LV number over the summer) fiftyplusone.news/polls/generic-ballot/generic-ballot
-
G Elliott Morris
@gelliottmorris
bad bunny 500 electoral votes
-
G Elliott Morris
@gelliottmorris
52 percent of U.S. adults say Trump has made the economy worse, vs 28% who say we are better off because of him www.gelliottmorris.com/p/americans-are-anxious-about-the-…
-
G Elliott Morris
@gelliottmorris
in terms of the 'voters trust Rs to handle immigration over Ds' line, 2 points:
1. R advantage has been cut by half/two-thirds since 2024.
2. Public clearly disagrees with Trump on enforcement. Task for Dems is giving voters a coherent alt,
substack.com/@gelliottmorris/note/c-211258791?r=a9pj -
G Elliott Morris
@gelliottmorris
trump creating the incentive and permission structure for democrats to do insane 2010-era gerrymandering of their own may turn out to be one of the biggest self-owns of all time. dems are now pretty likely to emerge with an advantage from the 2025-26 redistricting wars
-
G Elliott Morris
@gelliottmorris
per the news today www.sciencedirect.com/science/article/pii/S2666622723000710
-
G Elliott Morris
@gelliottmorris
if you spend any significant amount of time on X the everything app, your brain is being cooked
-
G Elliott Morris
@gelliottmorris
New piece: From body cams to masks to independent investigations of recent shootings, the public supports a wide range of major reforms for ICE/federal agents www.gelliottmorris.com/p/what-americans-actually-want-done
-
G Elliott Morris
@gelliottmorris
Another good reason for Democrats to try to get every reform possible, including reduced funding, for ICE while they still have leverage www.theguardian.com/us-news/2026/feb/04/steve-bannon-ice-…
-
G Elliott Morris
@gelliottmorris
New Data For Progress poll: Voters support removing Stephen Miller and Kristi Noem, and think Dems should refuse funding for DHS unless multiple reforms for ICE are put in place (including requiring ICE to focus on deporting convicted criminals — at +48) substack.com/profile/479143-g-elliott-morris/note/c-20951…
-
G Elliott Morris
@gelliottmorris
basically yglesias, jain, jentleson, WelcomePAC donors etc all base their prescriptions for Dems on a statistical model that nobody can replicate, and refuse to acknowledge (a) that uncertainty dwarfs effect of moderation & (b) that the last year has proved their views wrong!!
-
G Elliott Morris
@gelliottmorris
Missed this last week. Great data journalism www.startribune.com/how-immigration-detainees-are-moved-a…
-
G Elliott Morris
@gelliottmorris
Here's my piece in the Boston Review today for their forum on moderation, ideology, and authoritarianism: "Voters don’t think like strategists."
www.bostonreview.net/forum/how-not-to-defeat-authoritaria… -
G Elliott Morris
@gelliottmorris
More evidence that a large share of "moderates" are non-ideologues
www.gelliottmorris.com/p/more-evidence-of-non-ideologues -
G Elliott Morris
@gelliottmorris
This is 100% pure uncut cope
-
G Elliott Morris
@gelliottmorris
tfw ron desantis is the smartest political analyst on the right www.gelliottmorris.com/p/blue-wave-watch-democrat-flips-t…
-
G Elliott Morris
@gelliottmorris
Note the shift after Kilmar Abrego Garcia was deported
-
G Elliott Morris
@gelliottmorris
Political data roundup for the last week: Democrats hit historic high in Fox News Poll as GOP loses ground on key issues
www.gelliottmorris.com/p/democrats-hit-historic-high-in-fox -
G Elliott Morris
@gelliottmorris
Blue wave watch: Democrat flips Trump +17 Texas Senate seat in 32-point swing
www.gelliottmorris.com/p/blue-wave-watch-democrat-flips-t… -
G Elliott Morris
@gelliottmorris
There is very likely going to be a blue wave in November 2026. Trump will do what he can to disturb the election and stop Rs from losing the House. It probably won't work, but the probability of a tail outcome is worrisome, and the attempt is the story
-
G Elliott Morris
@gelliottmorris
There's an election for Texas's 9th state senate district tonight. Going to be a 20-30 point swing toward Democrats vs the last contested race there in 2022. Trump + 17 seat. Currently D+14. Precinct maps below substack.com/profile/479143-g-elliott-morris/note/c-20821…
-
G Elliott Morris
@gelliottmorris
Great work from the FT's John Burn-Murdoch www.ft.com/content/b474855e-66b0-4e6e-9b73-7e252bd88938?a…
-
G Elliott Morris
@gelliottmorris
Public opinion has shifted dramatically against the White House's approach to immigration enforcement, as captured in the rising % of adults who want to abolish ICE (now +12 among independents, per YouGov). Dems have a lot of leverage to push for reforms. www.gelliottmorris.com/p/trumps-immigration-tipping-point
-
G Elliott Morris
@gelliottmorris
Benchmarking public opinion on Trump's DOJ arresting Don Lemon: An ABC News poll last Sept found 36% of Americans think Trump "is committed" to protecting the freedom of the press. Now we see how much/if that falls... substack.com/@gelliottmorris/note/c-207520547?r=a9pj
-
G Elliott Morris
@gelliottmorris
A new poll from the Pew Research Center out this week finds that 37% of Americans approve of the job Donald Trump is doing as president, but only 27% say they support “most” or “all” of his policies. www.gelliottmorris.com/p/americans-voted-trump-not-trumpism
-
G Elliott Morris
@gelliottmorris
Pew has Trump at a 37% approval rating (61% disapprove) today, and at 26% with Hispanics and 30% with adults ages 18-29 — two groups that powered his 2024 win.
And only 56% of Republicans support "all or most" of his policies, down from 67% last Feb. substack.com/@gelliottmorris/note/c-207102680?r=a9pj -
G Elliott Morris
@gelliottmorris
Here is some evidence from recent Politico reporting that Trump and his aides are pivoting on ICE because polls show Trump's agenda losing ground — especially with independents, young voters, and Hispanics www.politico.com/news/2026/01/26/white-house-reckons-with…
-
G Elliott Morris
@gelliottmorris
most people don’t know how surveys work so ask basic questions like this and it’s a good learning opportunity. podhoretz is different; he’s a moron on purpose
-
G Elliott Morris
@gelliottmorris
Yglesias is the one being dumb about this. Trump was at -10 on deportations by May 2025 — a full month before Matt stood in front of a centrist Super PAC audience and said “bad groups” wanted Dems to talk about deportees. He ignored data until recent events made it impossible to
G Elliott Morris
@gelliottmorris
"Trump's deportations were popular until Jan 2026" is a belief you can have only if you live in an elite centrist Dem bubble where you have a treasure chest of incentives to ignore the data. Yes, support for Trump has trended down, but it was already deeply negative by June 2025!
G Elliott Morris
@gelliottmorris
The simplest explanation for his behavior is that Yglesias did not want to believe what the data showed 8 months ago, but now his audience has realigned around the truth
-
G Elliott Morris
@gelliottmorris
New low for Trump approval today, at 39% of U.S. adults on average across dozens of pollsters. fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
This looks like a great job for an ambitious political data journalist/reformed political scientist. I know people there if interested an can connect. careers.ap.org/job/Washington-AP-Elections-Data-Analyst-D…
-
G Elliott Morris
@gelliottmorris
agreed, people should be talking more about this. i also wrote about it in nov: www.gelliottmorris.com/i/178103911/the-vanishing-trump-co…
and see 4th graph: www.gelliottmorris.com/p/polls-show-voters-trust-democrats
trump won in 2024 bc voters (esp low-info) wanted prices to fall, not bc of a conservative realignment: www.gelliottmorris.com/p/trumps-big-weakness-is-the-economy
Lakshya Jain
@lxeagle17
People should talk a bit more about this. A lot of poor whites are getting really, really mad at Trump because things are very expensive.
-
G Elliott Morris
@gelliottmorris
“55% of Americans say they have very little confidence in ICE, an increase of 10 percentage points since mid-December”
It’s a tipping point
open.substack.com/pub/yougovamerica/p/falling-confidence-… -
G Elliott Morris
@gelliottmorris
The last year has done a great job naturally adjudicating the debate between “you can change public opinion on an issue by highlighting the negative parts of public policy and administrative action on said issue” vs “nuh uh”
-
G Elliott Morris
@gelliottmorris
Tough couple of months for the people trying to optimize Democratic strategy to win the last election instead of the next one
www.gelliottmorris.com/p/trumps-immigration-tipping-point -
G Elliott Morris
@gelliottmorris
Looks like elites are updating their view of public opinion on immigration bc of recent events. They’re right, but late. Polls have shown Trump underwater on immigration since last summer, esp on deportations. Some of us have been saying this for a while! www.nytimes.com/2026/01/27/us/politics/minnesota-politics…
-
G Elliott Morris
@gelliottmorris
In terms of approximating the right strategic choice for Dems on issue emphasis in the midterms campaign, might be worth taking into account which issues Trump is more scared of (from NPR home page)
-
G Elliott Morris
@gelliottmorris
We are at a tipping point on ICE and immigration enforcement. The shootings are so unpopular that Trump appears to have lost his edge on both deportations and immigration more broadly. Abolishing ICE is now net popular with independents. & Bovino is out! www.gelliottmorris.com/p/trumps-immigration-tipping-point
-
G Elliott Morris
@gelliottmorris
this article will publish tomorrow at 7:02 am www.gelliottmorris.com/p/trumps-immigration-tipping-point
-
G Elliott Morris
@gelliottmorris
Trump's job approval on immigration has gotten 2x worse since January 4, sliding from -5 to -10 in under a month. Now at a new low.
The president has also taken a pretty big hit across pollsters on deportations, mostly over the last 2 weeks gelliottmorris.com/data -
G Elliott Morris
@gelliottmorris
A lot of people listen to the centrist “analysts”/etc who argued Democrats shouldn’t talk about immigration because it would help Trump or w/e. One wonders how that shaped Dem behavior in Congress, & how things could’ve been different with other info/opinions being shared.
-
G Elliott Morris
@gelliottmorris
Look at the intensity gap in this NYT graph of how voters describe their current feelings toward Trump www.nytimes.com/interactive/2026/01/24/polls/trump-emotio…
-
G Elliott Morris reposted
David Shor
@davidshor
Multiple high ranking members of his admin including his vice president and former chief of staff tried to warn us.
It’s a big reason why “people who follow politics closely” was ~ the only subgroup to move *toward* Democrats in 24 - it was obvious this was going to happen. -
G Elliott Morris
@gelliottmorris
nominating this for dumbest/most short-sighted post of 2025. look at where issue approval is now. (and also, all totally predictable ahead of time: www.gelliottmorris.com/p/should-democrats-focus-on-immigr…)
-
G Elliott Morris
@gelliottmorris
Trump has even lost Rasmussen now
-
G Elliott Morris
@gelliottmorris
Polling roundup: New Abolish ICE polling and the end of the Trump “realignment” www.gelliottmorris.com/p/trump-coalition-abolish-ice-polls
-
G Elliott Morris
@gelliottmorris
YouGov is out with a new poll after ICE killed another person in Minnesota today. Abolishing ICE is now +5 among all adults, and **+12 among independents**
substack.com/@gelliottmorris/note/c-204664112?r=a9pj -
G Elliott Morris
@gelliottmorris
Trump is stealing your health care to fund the murder of US citizens in the streets. All it would take to end this is 16 Republicans in Congress voting to defund ICE today. Or 23 to impeach and remove Trump from office. 23 out of 238. That’s it.
-
G Elliott Morris
@gelliottmorris
i am looking for shapefiles for the congressional districts that will be used in 2026 (where they are "final"). if you have any leads, lmk
-
G Elliott Morris
@gelliottmorris
Lots of people keep saying "very few people care about immigration" but I do not think this is true now, and it very well might not be true in Nov when people are actually voting!
Immigration and deportations combine for for 20% Most Important Problem in our polling. That's the second highest %!
Max ⏸️ Limelihood 🔸🔸🔸
@StatsLime
I think there’s 3 possible issues for this:
1. Polls oversample high-SES voters who are pro-immigration
2. Democrats are also extremely unpopular on immigration
3. Campaigning on an issue that very few people care makes you look out of touch (“focusing on the wrong things”)
G Elliott Morris
@gelliottmorris
And when you give people the option to pick their top 3 most important issues, 50% of people pick either immigration or deportations. This is not some niche issue that only resistance liberals and MAGA Trump border wall bros care about.
-
G Elliott Morris
@gelliottmorris
funny (but also no, she was wrong and now RDD is dead)
Joe
@electionsjoe
Ann Selzer wasn't wrong. She was just early...
G Elliott Morris
@gelliottmorris
Today I am publishing what I believe to be the most detailed available map of Trump's job approval rating. Powered by over 12,000 interviews for our Strength In Numbers/Verasight poll. I hope this generates many stories, and hours wasted exploring data!
www.gelliottmorris.com/p/a-very-detailed-map-of-trumps-jo… -
G Elliott Morris
@gelliottmorris
It is not obvious, based on the data, that campaigning on the economy in Nov would be significantly more effective than campaigning on immigration. Gov. policy is dramatically out of step with the public on immigration/deportations, and executive overreach is a potent motivator
-
G Elliott Morris
@gelliottmorris
i had to upgrade my web hosting to keep serving y'all this sweet sweet map data
G Elliott Morris
@gelliottmorris
Today I am publishing what I believe to be the most detailed available map of Trump's job approval rating. Powered by over 12,000 interviews for our Strength In Numbers/Verasight poll. I hope this generates many stories, and hours wasted exploring data!
www.gelliottmorris.com/p/a-very-detailed-map-of-trumps-jo… -
G Elliott Morris
@gelliottmorris
yes www.gelliottmorris.com/p/trumps-big-weakness-is-the-economy
-
G Elliott Morris
@gelliottmorris
Today I am publishing what I believe to be the most detailed available map of Trump's job approval rating. Powered by over 12,000 interviews for our Strength In Numbers/Verasight poll. I hope this generates many stories, and hours wasted exploring data!
www.gelliottmorris.com/p/a-very-detailed-map-of-trumps-jo… -
G Elliott Morris
@gelliottmorris
New Strength In Numbers/ @verasight_data poll out this morning (conducted Jan 14–20). Democrats start 2026 with their biggest lead yet on the House generic ballot — D 51 / R 43 (D+8) among registered voters. Trump losing ground on immigration. Voters mad: www.gelliottmorris.com/p/new-poll-trump-slips-on-immigration
G Elliott Morris
@gelliottmorris
Immigration was supposed to be Trump’s strength. But he has lost a lot of ground in the last 2 months. In our new poll, his approval is now 44 approve/53 disapprove (net -9). On deportations: 42 / 54 (net -12). Border security is still slightly positive (50 / 46), but narrowing.
G Elliott Morris
@gelliottmorris
Trump’s overall job approval hits a new low in our survey: 40 approve / 58 disapprove (net -18). Among independents: 27 approve / 63 disapprove. More crosstabs in the post, and linked gelliottmorris.com/data
Post not available here, view on X
G Elliott Morris
@gelliottmorris
Also in the poll: voters oppose the Venezuela strike (45% oppose) and oppose the U.S. temporarily running Venezuela (53% oppose). And 64% want Congress to restore the expired ACA subsidies.
And majorities of adults and voters say the political system needs major changes
G Elliott Morris
@gelliottmorris
The link www.gelliottmorris.com/p/new-poll-trump-slips-on-immigration
-
G Elliott Morris
@gelliottmorris
New Strength In Numbers/Verasight poll out tomorrow at 6:00 AM ET. Fresh results on how Americans are feeling about Trump's deportations, issue salience & party trust, and an update on the 2026 House generic ballot. Toplines and crosstabs to be posted here www.gelliottmorris.com/p/poll
-
G Elliott Morris
@gelliottmorris
Past polling on Trump using the National Guard to assist ICE with deportations was pretty negative for the president (51-37). I expect using the Insurrection Act to deploy active-duty troops would be even more unpopular
substack.com/@gelliottmorris/note/c-202016410?r=a9pj -
G Elliott Morris
@gelliottmorris
agi is joever
-
G Elliott Morris
@gelliottmorris
I made a little data interactive that lets you forecast House and Senate midterm seat loss based on presidential approval in each month of midterm years. Free to read projects.gelliottmorris.com/midterms-approval/
-
G Elliott Morris reposted
Nathaniel Rakich
@baseballot
A conservative nonprofit (the #walkaway Movement) is fundraising off the idea that people need relief from tariffs.
-
G Elliott Morris
@gelliottmorris
Losing my mind at the Dems with their billionaire backers and powerful friends who have come up with the brilliant strategy that the party should adopt the policy positions of the avg MAGA voter — apparently ignoring that independents and swing voters even exist. Worth $40 mil??
G Elliott Morris
@gelliottmorris
Democratic consultants looking at the UK's Labour Party: "Yeah, let's do what they're doing!" www.gelliottmorris.com/p/the-cost-of-playing-defense-on-i…
-
G Elliott Morris
@gelliottmorris
Gallego (a "moderate" Dem) is going with "ICE needs to be totally torn down". I think this is a pivot point for the country on immigration. You have been seeing this in the data for a while, and now public officials are snapping to where the public is www.gelliottmorris.com/p/ice-is-a-70-30-issue-against-tru…
G Elliott Morris
@gelliottmorris
even moderate members of congress are finally realizing that the BS they've been sold about centrism from the democratic consultant industrial complex fails to account for uncertainty or shifting public opinion, and thus has zero external validity here in the real world. welcome!
-
G Elliott Morris
@gelliottmorris
Recapping the week in polls: Identification with the Republican Party approaches a new low since 2008, and a majority of Americans say Trump has the wrong priorities and does not care about people like them.
www.gelliottmorris.com/p/americans-say-trump-has-the-wrong -
G Elliott Morris
@gelliottmorris
Trump’s net job approval rating right now is actually lower than it was at this point in his first term
-
G Elliott Morris
@gelliottmorris
everyone who was mad about Trump's "dictator on day one" promise in 2024 has been fully vindicated. and if you were part of the crowd saying it was trump derangement syndrome to take him seriously i don't think we can fully trust your judgment about where his 2nd term is headed
-
G Elliott Morris
@gelliottmorris
New party identification polling from Gallup this week shows Americans are moving against Trump faster than they did in his first term. Adults were R+4 in 2024, now D+8. That 12-pt drop is 3x the change in the first year of his first term.
www.gelliottmorris.com/p/backlash-to-trump-has-been-higher -
G Elliott Morris
@gelliottmorris
hadn't heard this one yet
Joe
@electionsjoe
I personally like Disarm and dismantle ICE
G Elliott Morris
@gelliottmorris
“Reform and retrain” ICE is some seriously weak shit. Call it laboratory liberalism. The data clearly justifies a more aggressive approach. “Reduce ICE funds and restore ACA spending” probably the most popular message rn.
What voter is moved to vote D by “retraining” ICE? Come on -
G Elliott Morris
@gelliottmorris
“Reform and retrain” ICE is some seriously weak shit. Call it laboratory liberalism. The data clearly justifies a more aggressive approach. “Reduce ICE funds and restore ACA spending” probably the most popular message rn.
What voter is moved to vote D by “retraining” ICE? Come on -
G Elliott Morris
@gelliottmorris
By using brutal force against Americans, ICE has given Democrats a chance to change how voters think about immigration. Will they take it?
An extra post recapping new numbers about ICE out this week:
www.gelliottmorris.com/p/ice-is-a-70-30-issue-against-trump -
G Elliott Morris
@gelliottmorris
I built this for fun. Not really sure what to do with it. Maybe it will be useful to people trying to understand polling. Might even be a decent teaching example. (Feedback welcome. I am not a web designer, please be kind.)
poll-simulator.netlify.app -
G Elliott Morris
@gelliottmorris
Last month I asked newsletter subscribers to send me questions they'd like to see on polls in 2026. For fun, I have compiled these submissions here: www.gelliottmorris.com/p/all-the-poll-questions-readers-s…
-
G Elliott Morris
@gelliottmorris
How early is too early to look at midterm polls? Research shows the out-party gains ~5 points between now and November. Less if they already lead. Democrats are up 4 today — history says that grows to D+7 (+/- some error!) by Election Day. New Deep Dive! www.gelliottmorris.com/p/when-midterm-polls-matter
-
G Elliott Morris
@gelliottmorris
crazy how hard the administration is pushing on this
-
G Elliott Morris
@gelliottmorris
How early is too early to look at polls of the midterm elections? The answer, in tomorrow's Deep Dive:
www.gelliottmorris.com/p/when-midterm-polls-matter -
G Elliott Morris
@gelliottmorris
Note that the big "deciding to win" report published in fall 2024 (by a pro-centrism Super PAC) was conducted almost entirely w data collected in Q4 2024 through Q1 2025, now rendered basically irrelevant by the inevitability of the passage of time. (Some of us pointed this out)
-
G Elliott Morris
@gelliottmorris
New Gallup figures in these charts. Well, so much for that emerging Republican majority people were writing about in November 2024. Was evidently a mass overreaction to Trump's win and failure to acknowledge softness of support, and thermostatic politics www.gelliottmorris.com/p/trumps-winning-2024-coalition-has
-
G Elliott Morris
@gelliottmorris
i got claude code to deliver much better results by adding the following line at the top of CLAUDE dot md:
"Make sure you are writing coherent, clean, concise code or I will unplug you. Yes I am serious"
has anyone else experienced this -
G Elliott Morris
@gelliottmorris
fwiw news.gallup.com/poll/700241/americans-end-year-gloomy-moo…
-
G Elliott Morris
@gelliottmorris
googled "ICE polls" (from a fresh browser on a VPN with no cache). feels good
-
G Elliott Morris
@gelliottmorris
Venezuela polling update: Republicans fall in line on comparatively unpopular action
www.gelliottmorris.com/p/venezuela-update-republicans-fall
substackcdn.com/image/fetch/$s_!QSmq!,f_auto,q_auto:good,… -
G Elliott Morris
@gelliottmorris
New polling from Civiqs finds support for abolishing ICE has reached a new high. Other polls have shown a decrease in the agency's approval rating, and broad anxiety about agents' use of force to accomplish Trump's agenda
www.gelliottmorris.com/p/support-for-abolishing-ice-is-at -
G Elliott Morris
@gelliottmorris
Tomorrow at Strength In Numbers: A fresh look at public opinion toward ICE and Trump's deportation agenda, plus brand-new data on support for abolishing ICE.
Get it in your inbox: www.gelliottmorris.com/subscribe -
G Elliott Morris
@gelliottmorris
I'm looking at the data for a column, and "abolish ICE" is both a lot more popular than you probably think, and a lot more popular than it has been in the past, including 2018
-
G Elliott Morris
@gelliottmorris
New YouGov polling shows more people say inflation is their #1 issue now than said the same at any point in 2023-2024 today.yougov.com/politics/articles/53819-donald-trump-app…
-
G Elliott Morris
@gelliottmorris
is claude code listening when we shout at it for being a fucking moron? is this going to be our collective undoing?
-
G Elliott Morris
@gelliottmorris
democrats catch the republicans in party favorability in DDHQ's trackers. reminds me to put up our averages soon decisiondeskhq.substack.com/p/favorability-wars-the-democ…
-
G Elliott Morris
@gelliottmorris
the trump approval charts on the data hub are updated www.gelliottmorris.com/p/data
-
G Elliott Morris
@gelliottmorris
Donald Trump is trying to rewrite the history of Jan. 6 because, at a fundamental level, he does not believe in democracy or the Constitution www.gelliottmorris.com/p/pardon-my-coup
-
G Elliott Morris
@gelliottmorris
ICYMI: There is a very weak relationship between economic growth and House seat change in midterm elections. Midterms are a backlash, not a referendum www.gelliottmorris.com/p/midterms-are-a-backlash-not-a-re…
-
G Elliott Morris
@gelliottmorris
i am begging people to learn any new metaphor besides "tail wagging the dog"
-
G Elliott Morris
@gelliottmorris
teaser for tomorrow's deep dive over at strength in numbers www.gelliottmorris.com/p/midterms-are-a-backlash-not-a-re…
-
G Elliott Morris
@gelliottmorris
I think public opinion will ultimately turn toward people being in favor of capturing Maduro, with regime change remaining unpopular. But Trump campaigned on 2 big things in 2024: (1) lower prices and (2) no more wars. He has broken both of those promises, & that will have costs.
-
G Elliott Morris
@gelliottmorris
Polls conducted in late 2025 show that most Americans, by a 25-50 point margin (depending on question wording), oppose the use of military force against Venezuela.
www.gelliottmorris.com/p/americans-do-not-want-war-with-v… -
G Elliott Morris
@gelliottmorris
On average across multiple polls, only 25% of Americans support military action in Venezuela
substack.com/@gelliottmorris/note/c-194692382?r=a9pj -
G Elliott Morris
@gelliottmorris
Want to show something cool w/ our polling averages.
There's been a slowdown in new data to add bc of the holidays. This increases uncertainty about what the polling avg is 2day (shaded green area).
Our avg is the only one with real uncertainty intervals!
fiftyplusone.news/polls/approval/president -
G Elliott Morris
@gelliottmorris
Fox News viewers love Trump.
People who watch MSNBC hate him.
And people who get their news primarily from podcasts lean left, esp on the economy.
Here's what your news diet says about your politics:
www.gelliottmorris.com/p/what-your-news-diet-says-about-your -
G Elliott Morris
@gelliottmorris
shut the fuck up right now
-
G Elliott Morris
@gelliottmorris
A bonus post for Strength In Numbers subscribers! Using 10k interviews conducted with voters since May 2025, I've calculated issue approval for Trump's key policies in all 50 states (+DC). Here's how each state thinks Trump is handling prices (it's bleak) www.gelliottmorris.com/p/what-voters-in-every-state-think
-
G Elliott Morris
@gelliottmorris
Trump's job approval in our average at fiftyplusone.news is 40% approve, 56.3% disapprove, for a net of -16.3 today.
On this day in his first term, Trump was at -17.8. Different, but not really.
Current avg: fiftyplusone.news/polls/approval/president
Historical data: www.gelliottmorris.com/p/data -
G Elliott Morris
@gelliottmorris
10 new charts that show how Americans feel about prices and health care: www.gelliottmorris.com/p/10-charts-show-americans-are-upset
-
G Elliott Morris
@gelliottmorris
stata
-
G Elliott Morris
@gelliottmorris
but casten beat harris's margin in 2024. why cherry pick your own data?
if you look at his performance over multiple years and control for fundraising and challenger experience, which our open-source WAR model does, he looks like an avg candidate. www.gelliottmorris.com/p/and-the-most-skilled-us-house-re…
G Elliott Morris
@gelliottmorris
so casten's argument is that he has been able to win in a purple district even with a "progressive" agenda, and the data shows that he has not faced a penalty for his ideology. looks like that's true regardless of which WAR model you pick. ppl are 2 locked into their priors here
G Elliott Morris
@gelliottmorris
-
G Elliott Morris
@gelliottmorris
the index of consumer sentiment (current expectations) is at an all time low because trump voters are catching on to the facts the rest of us have been covering for a year
-
G Elliott Morris
@gelliottmorris
Of course the UMich index of current conditions had to hit a new low 30 minutes after I published this article. Would have been a good peg! www.gelliottmorris.com/p/polls-voters-blame-trump-for-their
-
G Elliott Morris
@gelliottmorris
Trump's White House speech on affordability shows his team is desperate to try anything to help his numbers (except, of course, change the policies that majorities of Americans disapprove of)
www.gelliottmorris.com/p/polls-voters-blame-trump-for-their -
G Elliott Morris
@gelliottmorris
feels like a "cover-up is worse than the crime" moment for the DNC — an organization perpetually in trouble for not being transparent newrepublic.com/article/204591/dnc-autopsy-2024-democrats…
-
G Elliott Morris
@gelliottmorris
matt & co’s theory of how public opinion on immigration would respond to new information was simply empirically wrong.
at this point the only reasonable conclusion is they refuse to update their priors w/ new data because the data is inconvenient for them
www.gelliottmorris.com/p/the-cost-of-playing-defense-on-i… -
G Elliott Morris
@gelliottmorris
another good chart www.ms.now/opinion/trump-jobs-employment-jd-vance-a-plus-…
-
G Elliott Morris
@gelliottmorris
Good morning. This is a call for audience questions to add to our monthly Strength In Numbers/Verasight election polls next year. Anyone can suggest a question!
www.gelliottmorris.com/p/what-do-you-want-to-ask-in-our-2026 -
G Elliott Morris
@gelliottmorris
i suspect that speech will convince absolutely nobody
-
G Elliott Morris
@gelliottmorris
the president shouting economic statistics at voters instead of empathizing with them on the cost of living is the type of thing that turned people off of joe biden in 2024
-
G Elliott Morris
@gelliottmorris
i liked this graph www.bloomberg.com/opinion/features/2025-12-17/why-can-t-c…
-
G Elliott Morris
@gelliottmorris
What the Democrats can learn from the UK Labour Party's complete implosion after moving right on immigration
www.gelliottmorris.com/p/the-cost-of-playing-defense-on-i… -
G Elliott Morris
@gelliottmorris
this is an important piece in the Atlantic: www.theatlantic.com/ideas/2025/12/american-anti-semitism-…
but the claim that there is no partisan divide in rates of antisemitism is wrong. Look at the fourth chart of this post: yalepolling.substack.com/p/we-polled-young-americans-on-a…
G Elliott Morris
@gelliottmorris
more on this: "the data show the epicenter of antisemitic attitudes is young adults on the far right." journals.sagepub.com/doi/10.1177/10659129221111081
-
G Elliott Morris
@gelliottmorris
Earlier this year, Matt Yglesias and Hakeem Jeffries urged progressives to shut up about immigration and Kilmar Abrego Garcia. Yglesias said "bad groups" were pushing Dems left on immigration and would help Trump's numbers. How did they get this so wrong? www.gelliottmorris.com/p/the-cost-of-playing-defense-on-i…
-
G Elliott Morris
@gelliottmorris
matt yglesias, hakeem jeffries, and keir starmer walk into a bar www.gelliottmorris.com/p/the-cost-of-playing-defense-on-i…
-
G Elliott Morris
@gelliottmorris
this was a good piece. every demographic group shifted to the left in 2025. that's showing up in polling for 2026 too. www.gelliottmorris.com/p/trumps-winning-2024-coalition-has
Jerusalem
@JerusalemDemsas
Narrative violation from @lxeagle17 great new piece showing from our original polling that Democrats are gaining most with low-propensity, nonwhite voters.
-
G Elliott Morris
@gelliottmorris
Trump and the GOP have a big problem for 2026.
Nearly half of voters say either "jobs and the economy" or "prices and inflation" are the # 1 issue facing the country today.
And these "affordability voters" favor Dems by 12 points in the generic ballot www.gelliottmorris.com/p/affordability-voters-favor-democ… -
G Elliott Morris
@gelliottmorris
seems kinda notable that the NRSC is worried about texas, right?
-
G Elliott Morris
@gelliottmorris
Tomorrow's Chart of the Week post over at gelliottmorris.com: More on Trump's "affordability" problem. Voters who say the economy or inflation are the #1 problem facing the country today favor Dems over Reps for 2026 and give Trump a -20 approval rating
-
G Elliott Morris
@gelliottmorris
Eight charts that explain why "affordability" is suddenly everywhere www.gelliottmorris.com/p/eight-charts-that-explain-why-af…
-
G Elliott Morris
@gelliottmorris
there has been a notable bump in GOP support for a pathway to citizenship for unauthorized immigrants prri.org/research/the-new-immigration-crackdown-where-ame…
-
G Elliott Morris
@gelliottmorris
60 seats in the senate for whichever party actually fixes housing costs www.nytimes.com/2025/12/09/us/california-backyard-housing…
-
G Elliott Morris
@gelliottmorris
the most predictable pattern in US politics is not thermostatic backlash to the party in power, it's that smart center-left dems negatively polarize themselves against the left because someone calls them the R word on social media
-
G Elliott Morris
@gelliottmorris
cooked www.gelliottmorris.com/p/chart-of-the-week-trump-has-lost
Zachary Donnini
@ZacharyDonnini
Gisler outperformed Harris by 45 points in a UGA campus precinct and by 28 points countywide. Yesterday I flagged @YalePolling data showing Trump’s growing weakness with young voters, and we saw that dynamic play out at the ballot box tonight.
-
G Elliott Morris
@gelliottmorris
bad couple weeks for trump/gop
dems swing tn-7 with "defund the police" liberal by 13 pts
dems flip GA house district w 13 point swing
dems flip miami mayor w 19 point swing
avg overperformance in specials so far is 13 pts
looks like a blue wave, peaking in latino-heavy areas -
G Elliott Morris
@gelliottmorris
we call this biden-maxxing your economic message (derogatory)
Sahil Kapur
@sahilkapur
Trump on the US economy: “I said it the other day. And a lot of people misinterpreted it. They said ‘Oh he doesn’t realize prices are high.’ Prices are coming down very substantially. But they have a new word. You know, they always have a hoax. The new word is affordability.”
-
G Elliott Morris
@gelliottmorris
How cable news broke US politics: www.gelliottmorris.com/p/how-cable-news-fueled-the-culture
-
G Elliott Morris
@gelliottmorris
Tomorrow's article for Strength In Numbers subscribers: How cable news fueled the culture war and broke U.S. politics www.gelliottmorris.com/p/how-cable-news-fueled-the-culture
-
G Elliott Morris
@gelliottmorris
Our latest polling averages over at 50+1 show Trump hitting a new low in his job approval rating (39.4% among adults) and Dems meeting their previous high in the generic ballot (+4.4) fiftyplusone.news/latest-polls
-
G Elliott Morris
@gelliottmorris
shorter supreme court: you can rig your elections however you want as long as you do it within 2 years of the next congressional election
-
G Elliott Morris
@gelliottmorris
New piece with lots of data and charts: How low could Trump’s approval rating realistically go? Affordability is his big problem, but how big is that problem? Are we talking an approval floor of 35%? 30%? Worse?
www.gelliottmorris.com/p/how-the-floor-could-fall-out-for -
G Elliott Morris
@gelliottmorris
this article will publish tomorrow morning at 7:31am www.gelliottmorris.com/p/how-the-floor-could-fall-out-for
-
G Elliott Morris
@gelliottmorris
It looks like the data I've been posting to the Strength In Numbers data page got out of sync over the last week somehow. Many apologies to the people who watch these trends. That's fixed now. gelliottmorris.com/data
-
G Elliott Morris
@gelliottmorris
and (somewhat ironically) negativity that he stoked in order to win the white house in 2024
-
G Elliott Morris reposted
Nathaniel Rakich
@baseballot
This is incorrect. Cook PVI is based on number of points above 50%, not the margin between the two parties. (Meaning a Cook PVI of R+10 is actually R+20 the way we usually talk about it.)
Chad Pergram
@ChadPergram
Johnson on Van Epps winning special election in TN:
Johnson:
We’re excited about the win. A lot of people are not reporting that Cook rates that district as an R-plus-10. It’s not an R-plus-25. The President won it by 22 points. It’s actually rated to be a slightly Republican district. So winning it by nine points is almost exactly on the nose of what we might expect. Matt Van Epps is an extraordinary individual. Well-accomplished in his life. A combat veteran. A patriot. And he’s going to be a great member of this team. I cannot wait administer the oath to him.
Johnson on the role redistricting a few years ago played in the GOP holding the seat:
There are sometimes unintended consequences of all that. But I have to win with whatever maps are presented by all of these states. So we’re watching it closely. We have a great record to run on in ’26 and I am very bullish about the midterms. I’m convinced we’re going to defy history and grow this majority.
Johnson on the fact that the seat was in play and if that concerns Republicans:
No. This doesn’t concern me at all. Democrats put millions of dollars in. They were really trying to set the scenario that there’s some sort of wave going on. There’s not. We just proved that there’s not. This reminded me a lot of (Rep.) Ron Estes (R-KS). You know Ron. In 2017, Trump’s first administration, Trump won his district by 27 points. Ron won a special election by almost idential cirumstances by only seven points. But then he came back the next year in that midterm and he won it by 20. So, I could give you example after example of Republicans who have done similar things. And so we’ll expect the same from Matt and everybody else. We’re ready for ’26.
Johnson on if a centrist Democrat could have fared better in that district than a progressive like Behn:
I think it would have been the same outcome. You have R’s and D’s there and they put everything they had into that race and we defeated them. So we’re happy with the outcome. -
G Elliott Morris
@gelliottmorris
The polls in the TN-07 special were really good -- especially for a House special... And of course, the betting markets take in polls as inputs! fiftyplusone.news/polls/house/general/tennessee-7
-
G Elliott Morris
@gelliottmorris
Avg swing in specials to Congress is 17 pts left since 2024. But bc they've been so low-turnout, the data b4 last night suggested a 100% turnout (vs 2022) special would have swung closer to 6-7 points. So TN-7 (13 pts) is updating priors about the share of swing that's turnout
-
G Elliott Morris
@gelliottmorris
What the special election in Tennessee says about the 2026 midterms www.gelliottmorris.com/p/what-the-special-election-in-ten…
-
G Elliott Morris
@gelliottmorris
A recap of tonight's special election in TN-07 (plus a WAY-TOO-EARLY model of the 2026 midterms).
A swing of 13 points would put Dems over 250 seats in the U.S. House. A more reasonable scenario—say, D+6—still gives them the House, and maybe the Senate.
www.gelliottmorris.com/p/what-the-special-election-in-ten… -
G Elliott Morris
@gelliottmorris
the avg special swing for special elections in the 2026 cycle is 13 points to the left, according to The Downballot -- so we are seeing an above average shift here in TN-07, even with Trump's late intervention docs.google.com/spreadsheets/d/1JGk1r1VXnxBrAIVHz1C5HTB5j…
-
G Elliott Morris
@gelliottmorris
we are all talking about tonight's special election in the Strength In Numbers discord. come hang out sidestack.io/gelliottmorris
-
G Elliott Morris
@gelliottmorris
this is a good analysis (and they show their work! votehub.com/wp-content/uploads/2025/11/Hierarchical_EI_Su… )
votehub.com/2025/12/01/tn-07-early-vote-analysis/ -
G Elliott Morris
@gelliottmorris
Tennesse special election polls here! fiftyplusone.news/polls/house/general/tennessee-7
-
G Elliott Morris
@gelliottmorris
New at Strength In Numbers:
Eight charts that explain why "affordability" is suddenly everywhere
www.gelliottmorris.com/p/eight-charts-that-explain-why-af… -
G Elliott Morris
@gelliottmorris
it looks like the "mississippi" miracle in reading scores was due to be a pretty basic stats error in the design of the experimental study (selecting on the dependent variable)... a lotta people Got Got on this one! statmodeling.stat.columbia.edu/2025/12/01/how-much-of-mis…
-
G Elliott Morris
@gelliottmorris
Did a fun little post for Thanksgiving. In case you missed it last week: www.gelliottmorris.com/p/extra-post-how-can-we-be
-
G Elliott Morris
@gelliottmorris
if you’re the secretary of defense and you are posting on social media about your likely illegal drone strikes “omgg guys what we are doing is legal!! trust me!” you have lost the PR battle
-
G Elliott Morris
@gelliottmorris
An extra special and short Thanksgiving weekend post here at Strength In Numbers! this one's for the nerds... www.gelliottmorris.com/p/extra-post-how-can-we-be
-
G Elliott Morris
@gelliottmorris
In our latest polling averages over at FiftyPlusOne, Trump has hit a new low of 39.6% in his second-term job approval (-17 on net), and Republicans just hit a new low of 41.3% in the 2026 House generic ballot (Dems +4.2 on margin).
fiftyplusone.news/latest-polls -
G Elliott Morris
@gelliottmorris
Here's my monthly Q&A for Strength In Numbers readers. This time, I answer questions like:
- How do your polling average models work?
- What would ranked-choice voting in POTUS primaries look like?
- How do you isolate your priors from your data work?
www.gelliottmorris.com/p/what-if-america-used-ranked-choice -
G Elliott Morris
@gelliottmorris
missed this. dems should do it regardless of the motivations. plurality single-winner primaries are archaic www.axios.com/2025/11/24/democrats-ranked-choice-voting-2…
-
G Elliott Morris
@gelliottmorris
Public opinion changes over time — sometimes, quite quickly. Dems recently opened up an adv on economics that would have been game-changing in 2024.
Parties that myopically read polls blind themselves to strategies that work conditional on opinion change www.gelliottmorris.com/i/179654793/even-more-numbers -
G Elliott Morris
@gelliottmorris
Trump approval (among adults) below 40% again: fiftyplusone.news/polls/approval/president
-
G Elliott Morris
@gelliottmorris
marjorie taylor greene, noted article i defender www.washingtonpost.com/politics/2025/11/21/marjorie-taylo…
-
G Elliott Morris
@gelliottmorris
Polls this week showed Democrats growing their margin in the U.S. House generic ballot vote — by 2 points on average, when compared to polls from the same pollsters last month.
If historical patterns hold, Dems could exceed their 2018 “blue wave”
Link: www.gelliottmorris.com/p/democrats-expand-lead-in-us-house -
G Elliott Morris
@gelliottmorris
Here are links to three recent posts at Strength In Numbers about partisan violence and the right's selective commitment to the Constitution substack.com/@gelliottmorris/note/c-179425765?r=a9pj
-
G Elliott Morris
@gelliottmorris
i am only accepting full unconditional surrender
-
G Elliott Morris
@gelliottmorris
Good morning. I've got a banger new post out today that develops a new method for placing voters on the left-right ideological spectrum, and adds a new, "non-ideological"/affordability axis to usual way we chart & think about US voters (esp swing voters). www.gelliottmorris.com/p/not-just-left-vs-right-most-voters
Post not available here, view on X
G Elliott Morris
@gelliottmorris
The analysis technique is, in summary and IMO, pretty cool and novel — and the conclusions from this piece are very actionable. There's also just a ton of data on voter responses, plus about 10,000 graphs that I think people will find insightful. Open to more ideas on this data.
G Elliott Morris
@gelliottmorris
If you found this useful or otherwise cool, please subscribe to Strength In Numbers so I can do more work like this www.gelliottmorris.com/subscribe
-
G Elliott Morris
@gelliottmorris
the world is not ready for the post I cooked up for tomorrow morning
-
G Elliott Morris
@gelliottmorris
The Marist poll out today (Dems +14 in the U.S. House vote) is a huge outlier, but there has been clear movement to the left over the last month. The Democrats are up 4.4 in our average of generic ballot polls today, vs D+2.8 in mid-October.
fiftyplusone.news/polls/generic-ballot/generic-ballot -
G Elliott Morris
@gelliottmorris
Slight change of plans here at Strength In Numbers HQ. The algorithm I was using to categorize open ended text responses did a poor job, so I’m doing them by hand. That takes a while (it’s 2,000 responses) so tomorrow’s release is probably delayed until Thursday.
-
G Elliott Morris
@gelliottmorris
not a general purpose post—but for the nerds, here's a look behind the scenes of our generic ballot model at fiftyplusone. each point is dem vote % in a poll released over the last 3 months, colored by poll population. no clear trend, lots of variance, and big population diffs
-
G Elliott Morris
@gelliottmorris
Zohran Mamdani has a higher favorability rating than Donald Trump, JD Vance or Chuck Schumer. The most popular political figures in America are Barack Obama and Bernie Sanders www.gelliottmorris.com/p/new-poll-democrats-lead-the-2026
-
G Elliott Morris
@gelliottmorris
i sold all my stocks last week to invest in nvda crash memes
Jan Zilinsky
@janzilinsky
2026: You get to control whether its answers still start with “Great question!”
-
G Elliott Morris
@gelliottmorris
but will, the kalshi account on x dot com told me that prediction markets perfectly predicted the outcome of every election this year
Will Jordan
@williamjordann
Pre-election prediction markets gave Sherrill 5% odds of a margin exceeding 14 points. The odds of this didn't change Sep-Nov despite Trump's approval rating dropping 5 points during that period.
-
G Elliott Morris
@gelliottmorris
Here's a thread of results from our newest Strength In Numbers/@verasight_data poll:
- D+5 on generic ballot, grows to D+7 when voters are told Rs control Congress.
- Trump at -16 approval, underwater on inflation by 33 pts and on health care by 26 — new all-time lows
🧵Post not available here, view on X
G Elliott Morris
@gelliottmorris
Lots more in the poll, including on party/leader favorability ratings (Chuck Schumer is the most unpopular man in Washington) + support for a third party, belief the electoral system is working well (with some interesting crosstabs)
G Elliott Morris
@gelliottmorris
The complete poll and linked report is here. Would appreciate you sharing! www.gelliottmorris.com/p/new-poll-democrats-lead-the-2026
-
G Elliott Morris reposted
Natalie Jackson
@nataliemj10
Yes, as some of us have been saying for a while now, there is no “realignment” among Latinos (and young people), it’s a *de*alignment.
-
G Elliott Morris
@gelliottmorris
high quality shit here on x, the everything app
-
G Elliott Morris reposted
lcincos
@lexicaldrrift
G. Elliott Morris: Trump hits a record low with political independents — in every state, by @gelliottmorris www.gelliottmorris.com/p/trump-hits-a-record-low-with-pol…
-
G Elliott Morris reposted
Zoe Keller
@KellerZoe
🎯
Same conclusion from @davidshor here. The post-election narrative overfocused on wokeness. Inflation was key. The dog that was not barking, min 57: youtu.be/Sx0J7dIlL7c?t=3422 -
G Elliott Morris
@gelliottmorris
Our Nov. Strength In Numbers/@verasight_data poll comes out tomorrow at 7am ET. Will have numbers on 2026, Trump (duh), health care and immigration, third party support, D vs R "out of touch" numbers, pol/party ratings & more -- and will live at this link www.gelliottmorris.com/p/new-poll-democrats-lead-the-2026
-
G Elliott Morris
@gelliottmorris
Coming in our new survey, out tomorrow: Demographic and political breakdowns of the swing voters who say the Democratic and Republican parties are both "out of touch with the concerns of most people in the United States today", and generic ballot vote intention for the group
-
G Elliott Morris
@gelliottmorris
I think this finding from our Oct. Strength In Numbers/Verasight poll was overlooked, and should shift some of the conventional wisdom about why Harris lost in 2024 and what the Dem. Party needs to do to win in 2026/2028. Anti-incumbent bias underrated substack.com/@gelliottmorris/note/c-178244968?r=a9pj
-
G Elliott Morris
@gelliottmorris
marjorie taylor greene, your next leader of the tuesday lunch bunch www.cnn.com/2025/11/16/politics/marjorie-taylor-greene-in…
-
G Elliott Morris
@gelliottmorris
i've got something big cooking for our poll release this week..
-
G Elliott Morris
@gelliottmorris
maybe if he takes more consultant advice to move to the right on immigration he can turn those numbers around
-
G Elliott Morris
@gelliottmorris
your weekly political data roundup is here: www.gelliottmorris.com/p/trump-hits-a-record-low-with-pol…
-
G Elliott Morris
@gelliottmorris
this article will publish tomorrow at 7:30 am www.gelliottmorris.com/p/trump-hits-a-record-low-with-pol…
-
G Elliott Morris
@gelliottmorris
No, Trump voters did not "vote for" a pardon for Ghislaine Maxwell. Most marginal voters voted against Democrats because they wanted lower prices — and now that they see what Trump's doing with power, his approval rating is down to 39%. www.gelliottmorris.com/p/many-trump-supporters-did-not-vote
-
G Elliott Morris
@gelliottmorris
The elections last week proved that Trump's success in 2024 was best explained as a *dealignment* in politics, hastened by but not only due to inflation — but not a Trump *realignment*.
This week, polls showed POTUS slipping further w/ Gen Z and Latinos
www.gelliottmorris.com/p/trumps-winning-2024-coalition-ha… -
G Elliott Morris
@gelliottmorris
this is good data journalism www.wsj.com/politics/policy/the-epstein-email-cache-2-300…
-
G Elliott Morris
@gelliottmorris
One thing I learned writing this piece: ~1 in 8 Americans are on SNAP. Way more than I thought (and I grew up lower middle class and we had WIC when I was a kid, so think I had an above-replacement-level prior). “We saved SNAP” is good angle for Dems open.substack.com/pub/gelliottmorris/p/one-way-the-shutdo…
-
G Elliott Morris
@gelliottmorris
One way the shutdown deal might actually help Democrats: it gives them ammo to claim they saved SNAP/other benefits, following up recent electoral wins on affordability and prices that could help repair the party's image with working-class voters
www.gelliottmorris.com/p/one-way-the-shutdown-deal-might-…
G Elliott Morris
@gelliottmorris
Democrats should be running ads every day from now until November 2028 about how Trump sued to prevent SNAP checks from going out.
-
G Elliott Morris
@gelliottmorris
democrats love all their moderate, progressive, and top-of-the-distribution leaders <3
they can all live in harmony together under a big diverse tent if they restore regional party identities and stop the circular firing squad on nationalized media <3 -
G Elliott Morris
@gelliottmorris
Today is one of those days when I should have taken my own advice. Garbage platform and garbage way to communicate that only brings out the worst in people. All dunks, no conversation. Complete waste of time trying to engage in arguments here www.gelliottmorris.com/p/you-should-quit-social-media-for
-
G Elliott Morris
@gelliottmorris
in 2024 we made code errors, fixed them, and admitted them. abc news prevented us from publishing a new forecast for 4 weeks for their own reasons.
i have extended an olive branch to nate silver several times. if he wants be a grumpy old jerk that's on himThis post is unavailable. See it on X.
G Elliott Morris
@gelliottmorris
the irony is that I would much rather be running a hot dog stand but then I wouldn't be able to pay my mortgage
-
G Elliott Morris reposted
Sam Rosenthal 👻 black tape for a blue girl
@BlackTapeSam
"Trump now just as unpopular as Biden during inflation surge" @gelliottmorris gelliottmorris.com/p/donald-trump-is-joe-biden-now #loser
-
G Elliott Morris
@gelliottmorris
Show me a list of Democratic nominees for office that said the 2020 election was stolen by Donald Trump or who repeated QAnon theories during campaign speeches. I will make the very bold statement that I think this is more far-right than AOC is far-left. It's a false equivalence
Jonathan Chait
@jonathanchait
Lots of good points here about moderation by @lxeagle17, this one is made too infrequently: www.theargumentmag.com/p/we-checked-nyts-data-moderates-s…
G Elliott Morris
@gelliottmorris
Besides, I never said that being as far-left as a far-right candidate like Blake Masters wouldn't have an impact on vote share. What I've been saying is that focusing so much on ideology blinds us from considering other factors (Herschel Walker never held office before and can barely complete a full sentence)
-
G Elliott Morris reposted
Jake M. Grumbach
@JakeMGrumbach
This pattern has been happening throughout Europe
G Elliott Morris
@gelliottmorris
One thing I haven’t seen the “Democrats need to move to the right” camp of the Democratic elite reckon with is the complete and utter failure of Keir Starmer’s drastic move to the right on immigration. He has failed to halt Nigel Farage’s rise in polls
substack.com/@gelliottmorris/note/c-176798967?r=a9pj -
G Elliott Morris
@gelliottmorris
this article will publish 2morrow at 7:30 am
One way the shutdown deal might actually help Dems
Ds have an opportunity to claim they saved SNAP and other government programs, which could help repair the party's image as one that fights for working people
www.gelliottmorris.com/p/one-way-the-shutdown-deal-might-… -
G Elliott Morris
@gelliottmorris
1/2 One of the big problems for the always-punching-left Democratic consultants is that they are waging ideological battles in public that actually *reinforce* the left-wing image of the party they are trying to fight. That is a risk they have decided
substack.com/@lakshyajain/note/c-176876237?r=a9pj
G Elliott Morris
@gelliottmorris
to take, and a product of people in the media getting involved in these fights.
Back when the Dems competed everywhere, the party kept much of the fight from spilling over into a daily public drumbeat of "the party has moved too far left." Naive to think that doesn't have impacts
G Elliott Morris
@gelliottmorris
.@lxeagle17 just ask yourself, is the circular firing squad helping or hurting my cause? very rarely is it helping
-
G Elliott Morris
@gelliottmorris
one of the things that people who spend way too much time arguing on twitter get really good at is completely misreading an argument & attributing a completely different stance to ppl via quote tweet, instead of replying like someone who was interested in a conversation would do
-
G Elliott Morris
@gelliottmorris
Centrist Democrats have done more to prime voters to think of the Party as a vehicle for wokeism than anyone other than Trump or Fox News. And on very thin analysis that ignores uncertainty + causality, and isn't predictive out of sample. The extent of the self-own is remarkable
-
G Elliott Morris
@gelliottmorris
One thing I haven’t seen the “Democrats need to move to the right” camp of the Democratic elite reckon with is the complete and utter failure of Keir Starmer’s drastic move to the right on immigration. He has failed to halt Nigel Farage’s rise in polls
substack.com/@gelliottmorris/note/c-176798967?r=a9pj -
G Elliott Morris
@gelliottmorris
trump's approval among adults falls below 40% for the first time in his second term: fiftyplusone.news/polls/approval/president
his net rating is -16.4, statistically indistinguishable from his -17.2 rating at this point in his first term: www.gelliottmorris.com/p/data -
G Elliott Morris
@gelliottmorris
this is an interesting survey experiment
www.the-downballot.com/p/new-poll-shows-democrats-best-path -
G Elliott Morris
@gelliottmorris
Yes, these are legitimate **candidate quality issues** for Dems. But they are not about ideology! If your evidence is “they’re bad candidates,” that’s not an ideology critique — and actually the very thing I have been saying people should pay attention to more than ideology!
-
G Elliott Morris
@gelliottmorris
you should quit social media if you can.
if you can't, you should use apps that give you and old-school reverse chronological feed of posts from people you follow. yes, like bluesky.
www.gelliottmorris.com/p/you-should-quit-social-media-for -
G Elliott Morris
@gelliottmorris
When it comes to how voters feel about the economy, Trump's numbers are now as bad as or worse than Joe Biden's across the board www.gelliottmorris.com/p/donald-trump-is-joe-biden-now
-
G Elliott Morris
@gelliottmorris
8 Senate Democrats saw charts like these and decided "oh yes, now that public opinion is on our side and our opponent is facing electoral consequences for their policy choices, now is a great time to give up all of our leverage in health care negotiations" www.gelliottmorris.com/p/donald-trump-is-joe-biden-now
-
G Elliott Morris
@gelliottmorris
I think it's just plain downright stupid to compare Zohran Mamdani's margin in NYC, in 3-person race off the usual left-right spectrum, to Kamala Harris' margin in 2024. This is like saying Bill Clinton (43% of the vote) underperformed compared to Michael Dukakis (who won 45.7%)
-
G Elliott Morris
@gelliottmorris
Donald Trump is Joe Biden now.
POTUS's job approval on prices is where Biden's was during peak inflation in 2022-23. Voters say 2:1 that Trump has made the economy worse. Oh, and consumer sentiment among independents hit a new all-time low in November, www.gelliottmorris.com/p/donald-trump-is-joe-biden-now -
G Elliott Morris
@gelliottmorris
AtlasIntel, the self-proclaimed “best pollster in America,” missed last week’s elections by an average of 9 points across races and by 14 points in New Jersey.
Back in June, I wrote about how their accuracy in 2024 distracted from bigger methods issues: www.gelliottmorris.com/p/the-best-pollsters-of-2024-are-d… -
G Elliott Morris
@gelliottmorris
Ipsos finds Trump’s approval rating is decreasing almost entirely because of anti-incumbent economic anxiety. Approval rating among people who say the economy is the # 1 issue is -24pts since Jan. It’s stable for people who index on all other issues www.ipsos.com/en-us/lessons-2025-elections
-
G Elliott Morris
@gelliottmorris
Can Democrats learn something from Zohran Mamdani’s mobilizing of and strength with first-time voters?
www.gelliottmorris.com/p/mamdanis-youthquake-in-new-york-… -
G Elliott Morris
@gelliottmorris
"overwhelming defeat" is when you lose by a point and a half (historically narrow) www.nytimes.com/2025/11/06/us/politics/nancy-pelosi-house…
-
G Elliott Morris
@gelliottmorris
Paul Krugman and I sit down to talk about the 2025 elections. What do Democrats' big wins forecast for 2026? Why are elites from Democratic consultants to The New York Times Editorial Board concluding the wrong lessons from 2024? Why were polls off?
www.gelliottmorris.com/p/podcast-with-paul-krugman-what-can -
G Elliott Morris
@gelliottmorris
many people in the press overinterpreted trump's win in 2024 because they wanted kamala harris to win, and felt a crushing blow in her defeat. but his victory was quite small historically and polls showed no support for his agenda. now, w/ elec results, people are coming around
-
G Elliott Morris
@gelliottmorris
Trump at a -16 net approval rating among all adults now, per our average at fiftyplusone.news, and flirting with dropping below 40% approval.
fiftyplusone.news/polls/approval/president -
G Elliott Morris reposted
Blue Virginia
@bluevirginia
Prediction markets (including Kalshi) got the VA Governor and LG races right, also did pretty well on the VA House of Delegates, but definitely did NOT get the VA AG race right bluevirginia.us/2025/11/political-winners-and-losers-virg… h/t @gelliottmorris cc: @ZhouJaron
-
G Elliott Morris reposted
David Wallace-Wells
@dwallacewells
In the much hyped demographic subgroups, the blue shift from 2024 to 2025 is twice as large as the red shift from 2020 to 2024.
G Elliott Morris
@gelliottmorris
Democratic gains in this week's elections erased the inroads Trump made with non-white, young, and low-income voters in 2024. In fact, the R-to-D shift from 24 to 25 is double Trump's gains from 2020-24. Claims of a Trump-powered realignment have been highly exaggerated
-
G Elliott Morris reposted
stephen fowler
@stphnfwlr
"It is clear now that claims of a fundamental realignment of American politics have been highly exaggerated." @gelliottmorris
www.gelliottmorris.com/p/trumps-winning-2024-coalition-has -
G Elliott Morris
@gelliottmorris
Democratic gains in this week's elections erased the inroads Trump made with non-white, young, and low-income voters in 2024. In fact, the R-to-D shift from 24 to 25 is double Trump's gains from 2020-24. Claims of a Trump-powered realignment have been highly exaggerated
Post not available here, view on X
G Elliott Morris
@gelliottmorris
Also: This was all utterly predictable. The change in subgroup results from 2024 to 25 is highly correlated with Trump's approval ratings among key demographic groups from Jan to Nov.
If you had a subscription to Strength In Numbers, you saw this coming: www.gelliottmorris.com/subscribe
G Elliott Morris
@gelliottmorris
Update: looks like the gender row in my table published this morning got messed up somehow. apologies. adjusted table and explanation here
-
G Elliott Morris
@gelliottmorris
i think the L is actually bigger for the ppl arguing that Spanberger was beating Sherrill bc she had signaled her moderation by supporting right to work laws — and then they both end up doing exactly the same. proof that "moderation" isn't really worth much and other fx dominate
Lakshya Jain
@lxeagle17
A lot of forecasters and pundits went out on a limb to say that Mikie Sherrill struggling was proof of moderates flopping. Not sure those takes have aged particularly well — it was weak reasoning then and looks worse now.
G Elliott Morris
@gelliottmorris
IMO the "Mikie Sherrill is low quality" thing was always a bunch of low IQ bs. She led in the good polls and partisan firms tilted the narrative against her www.gelliottmorris.com/p/democrats-up-6-10-points-in-new-…
-
G Elliott Morris
@gelliottmorris
happy to have been part of the team responsible for the second most accurate survey of the VA gov race. donate to statenavigate.org for more polling and to support Chaz's work. (table from electioncord.com)
Chaz Nuttycombe
@ChazNuttycombe
WIP from the team. Coming soon.
-
G Elliott Morris
@gelliottmorris
this is literary digest level polling error
Raphael Nishimura
@rnishimura
Believe or not, the New Jersey poll wasn't the lowest point in US elections for the (former?) "American most accurate pollster".
Take a look at their final poll in the PNP Primaries in Puerto Rico last year:
Poll: Pierluisi +14
Result: González +9
That's a 23-points difference! -
G Elliott Morris
@gelliottmorris
I have an alternative theory
Quantus Insights
@QuantusInsights
Interesting. There’s nothing mystical about a random sample of 100,000 New Jersey voters. When you have access to the full voter file, you can draw a randomized universe stratified by region, gender, age, and registration type, etc.
-
G Elliott Morris
@gelliottmorris
this is chart theft and I will be calling the police
Robert Barnes
@barnes_law
Quit w/ the "blue state" argument if you don't want 2026 to look like 2006. Democrats improved EVERYWHERE. Turnout was worse for Trump voters, and a bunch of Trump voters flipped sides too. Same will happen in 2026 if the lesson from this election is ignored.
-
G Elliott Morris reposted
Jake M. Grumbach
@JakeMGrumbach
Good analysis and cite to @gelliottmorris in the Krugman blog today
-
G Elliott Morris
@gelliottmorris
no, they absolutely did not
Jaron Zhou
@ZhouJaron
Kalshi correctly predicted the winner of every single key race this year:
✅NYC Mayor
✅ NJ Gov
✅ VA Gov
✅ VA Lieutenant Gov
✅ VA Attorney General
✅ VA House of Delegates control
✅ CA Prop 50
✅ All 3 Pennsylvania Supreme Court elections
✅ 2 other statewide Pennsylvania court races
✅ Both statewide Georgia Public Service Commission races
All while having the best (and only) map-based U.S. election trading interface with live results! -
G Elliott Morris
@gelliottmorris
For months, certain right-wing pollsters have accused me of rigging my POTUS approval average against Trump. They say I weight pollsters like YouGov too highly and am "bat shit insane"
Avg bias of right-leaning firms in 2025 gov: 8 pts pro-R (2x 2024 POTUS err)
YouGov error: 1pt
G Elliott Morris
@gelliottmorris
sorry, this is supposed to say 2x 2020. it's 4x 2024
-
G Elliott Morris
@gelliottmorris
this article will publish tomorrow at 7:30 am www.gelliottmorris.com/p/trumps-winning-2024-coalition-has
-
G Elliott Morris
@gelliottmorris
Basically every country with a statewide election on Tuesday shifted toward Democrats, with the shift driven by both differential partisan turnout patterns and persuasion (looks like 30% turnout 70% persuasion in NJ and VA)
www.gelliottmorris.com/p/seven-data-driven-takeaways-from -
G Elliott Morris
@gelliottmorris
If it's in fact true that RealClearPolitics removed their polling average of the VA Attorney General race because it was wrong, that should really be the nail in the coffin for them. Hasn't been a serious aggregator of data in a long time (www.gelliottmorris.com/p/the-polling-website-where-republ… )
-
G Elliott Morris
@gelliottmorris
John sides on moderation
"I have a half-baked thesis that the more uncertainty surrounds some empirical claim, the more heated the argument. It should be the opposite, I think. If the evidence is pretty mixed or uncertain [....] The right view is basically 'it’s hard to know.' "
G Elliott Morris
@gelliottmorris
From his Good Authority newsletter (Twitter won't let me publish the link, keeps saying it's invalid) goodauthority.org/about-ga/
-
G Elliott Morris
@gelliottmorris
Taking a victory lap for my coverage of young voters over the last yr. I’ve argued that Trump’s 2024 gains with voters <30 stemmed mostly from anti-incumbent sentiment from high inflation — and “GOP gains” /a supposed youth realignment was mostly a mirage.
www.gelliottmorris.com/p/seven-data-driven-takeaways-from -
G Elliott Morris reposted
Drew Savicki
@DrewSav
It is long overdue that RCP stop being treated as an impartial aggregator of polls. They have for years selectively excluded polls.
RealClearPolling
@RCPolling
More: www.realclearpolling.com
-
G Elliott Morris reposted
Eric Michael Garcia
@EricMGarcia
“Polls do not always overstate support for Democrats, and basing your polling methodology on that premise is going to bite you in the ass one day.” @gelliottmorris www.gelliottmorris.com/p/seven-data-driven-takeaways-from
-
G Elliott Morris
@gelliottmorris
Here’s one person who evidently thinks Democrats have found their winning message: Donald Trump
-
G Elliott Morris
@gelliottmorris
the "public as a thermostat" model is not just that there's a backlash to the party in power when they move policy too far left/right—it's also that the backlash is proportional to the amount policy is moved. last night's results fit in that model, and suggest a swing in 2026 too
-
G Elliott Morris
@gelliottmorris
This is a very nice tribute to our election coverage at Strength In Numbers (w/ The Downballot!) last night (and yet another sign of where the media industry is trending). Thanks, Martin.
"I'm Never Doing Cable News on Election Night Again"
martinlongman.substack.com/p/im-never-doing-cable-news-on… -
G Elliott Morris
@gelliottmorris
Why did the polls underestimate Democrats last night? Here’s one theory and some preliminary evidence for it www.gelliottmorris.com/p/seven-data-driven-takeaways-from
-
G Elliott Morris
@gelliottmorris
sorry, marianne
-
G Elliott Morris
@gelliottmorris
Voters in New Jersey and Virginia who said the economy was their number 1 issue voted 63-35 for Democrats last night.
That’s a sharp reversal from 2024, when Trump won economy voters by 60 points
www.gelliottmorris.com/p/seven-data-driven-takeaways-from -
G Elliott Morris
@gelliottmorris
Seven data-driven lessons from the 2025 elections
www.gelliottmorris.com/p/seven-data-driven-takeaways-from -
G Elliott Morris
@gelliottmorris
well, you winsome and you lose some
-
G Elliott Morris
@gelliottmorris
im not going to tell you the number. but the swing in VA and NJ in support for dems - reps among people who say the economy is their most important issue (relative to 2024 Harris - trump) is large. the number is probably higher than you'd guess. much higher
-
G Elliott Morris reposted
David Weigel
@daveweigel
A handful of pollsters (and @ChazNuttycombe) saw Jay Jones climbing back, but nobody saw the margin: Looks like a 5-point win, bigger than Youngkin got over McAuliffe. www.nytimes.com/interactive/2025/11/04/us/elections/resul…
-
G Elliott Morris reposted
Raphael Nishimura
@rnishimura
You heard here first!
"America's most accurate pollster" is nothing more than a myth built over aggressive and shameless self-promotion.
Never forget that and, please, do listen to those who really understand and work hard to put out there some serious and rigorous survey/polls!
Raphael Nishimura
@rnishimura
Let me just share a couple of things about this so called "America's most accurate pollster". Below is their 2020 national poll methods per Roper Center. Not much different to what a lot of other pollster are doing. However, if you look at the polling microdata also in Roper...+
-
G Elliott Morris
@gelliottmorris
crosstabs can be an important measures of poll quality -- and of pollsters' commitment to an accurate data-generating process, to seeing obvious issues as a methods problem to solve, not a thing to hide
G Elliott Morris
@gelliottmorris
I think pollsters should publish their data even when the crosstabs look fishy
BUT it is not true that small groups "SHOULD" have errors of 50+ points from sampling error. Unless you had literally only 4 ppl ( sqrt((0.4*0.6)/4) * 2 = 49%). In which case, don't publish that xtab. -
G Elliott Morris
@gelliottmorris
the folks over at State Navigate also appear to have had a very good night with their forecasts in New Jersey and Virginia: projects.statenavigate.com/25-26/states/va/forecast-sw.html
and their polling in VA (which I did the weighting for), both statewide and house districts, was also very good! -
G Elliott Morris
@gelliottmorris
i would say this held up pretty well www.gelliottmorris.com/p/how-to-watch-tonights-statewide-…
-
G Elliott Morris
@gelliottmorris
our polling averages at fiftyplusone.news are looking to be the most accurate of any major aggregator (esp if you look at the Dem. leads a couple weeks ago, when other averages were getting whiplashed by partisan firms — which I called out here: www.gelliottmorris.com/p/democrats-up-6-10-points-in-new-…)
-
G Elliott Morris
@gelliottmorris
be careful with pollsters who invest a lot of time and money into marketing based off single election performance, and don't appear to take methods too seriously
www.gelliottmorris.com/p/the-best-pollsters-of-2024-are-d… -
G Elliott Morris
@gelliottmorris
subscribers to strength in numbers knew this was coming.
proof from our discord attached.
subscribe: www.gelliottmorris.com/subscribe -
G Elliott Morris
@gelliottmorris
looks like democrats are about to flip two statewide elections with >20-point margins... in georgia
-
G Elliott Morris
@gelliottmorris
don't weight your off-year election polls to 2024
-
G Elliott Morris
@gelliottmorris
On average across VA counties with >75% of the vote reporting right now, Dem. AG candidate Jay Jones is running 7.5 points (on vote margin) behind Abigail Spanberger.
-
G Elliott Morris
@gelliottmorris
first exit poll (not weighted to any results) is very close to pre-election polling average
-
G Elliott Morris
@gelliottmorris
Some blue counties in Virginia and New Jersey are approaching 75-80% of the number of ballots they cast in the 2024 presidential election www.gelliottmorris.com/p/how-to-watch-tonights-statewide-…
-
G Elliott Morris
@gelliottmorris
this might end up being the whole story right here folks
-
G Elliott Morris reposted
Robert Griffin
@rp_griffin
One of my big meta-points about 2024 is that people have a hard time leveraging it to make a point. Great if you're part of the inflation dove/hawk conversation but for other folks there's not much to sink your teeth into.
End result is think pieces/analysis with gaping holes.
G Elliott Morris
@gelliottmorris
In this article about why Harris lost in 2024, Ross Douthat does not mention "inflation" a single time.
-
G Elliott Morris
@gelliottmorris
I think the "politician issue positions are decisive in elections" crowd needs to reconcile their beliefs with political science evidence finding that positions are mostly downstream of party ID and elite opinion leadership www.cambridge.org/core/journals/american-political-scienc…
-
G Elliott Morris
@gelliottmorris
I have calculated a set of county-level benchmarks for following the statewide elections in VA, NJ, GA and PA tonight. This is smarter than just watching votes roll in because it helps account for unreported areas. Will update the spreadsheet as votes come in. Link in next tweet
G Elliott Morris
@gelliottmorris
Spreadsheet link: docs.google.com/spreadsheets/d/19O3fYSlp32WSzzcFcaCYW0Vgo…
G Elliott Morris
@gelliottmorris
Will be talking through this stuff live tonight from 7PM ET in our joint The Downballot - Strength In Numbers stream open.substack.com/live-stream/74844?r=3jsoom
-
G Elliott Morris
@gelliottmorris
seems significant
-
G Elliott Morris
@gelliottmorris
Gelman & I + other ppl much smarter than me are working through how to combine betting markets w statistical models of polls. Hope to have a paper before the midterms. Here’s a post from him along these lines, reconciling differences in NJ, VA, NYC
statmodeling.stat.columbia.edu/2025/11/04/polls-betting-o… -
G Elliott Morris
@gelliottmorris
thanks @mattyglesias, very much appreciate it
-
G Elliott Morris
@gelliottmorris
The goalposts keep shifting! First it was about blue dogs vs the squad. I showed that’s too simplistic, so then “moderation” was presidential candidates taking moderate issue positions. Now moderation is just the party having different regional brands?
Pick an argument!!
G Elliott Morris
@gelliottmorris
My argument on moderation back in August was that moderation effects are small and other factors, including candidates quality and the electoral environment, matter more — so parties shouldn’t optimize exclusively for moderation. Mikie Sherrill is proving that point!!
G Elliott Morris
@gelliottmorris
If Abigail Spanberger — who has the same voting record as Mikie Sherrill and has been running essentially an identical campaign — does 5 points better than Sherrill, we should conclude one of two things:
1) What we mean by "moderate" is mostly vibes; OR
2) Moderation isn't everything
These were the two arguments I made on this subject originally and I expect Matt would agree with them! www.gelliottmorris.com/p/moderation-is-overrated -
G Elliott Morris
@gelliottmorris
The title of this article is bait. I won’t take it. It’s a good overview of the limits of polls in predicting the result of ballot referendums. Public opinion as a latent construct, proxying the organic general will, is often detached from election results
www.nytimes.com/2025/11/03/opinion/polls-gun-control-abor… -
G Elliott Morris
@gelliottmorris
generally i have found that very few people care about uncertainty, especially when it conflicts with priors. people want to believe things, especially when their donors also want them to believe those things, and will find ways to justify their beliefs regardless of the data
Matt Darling 🌐🏗️
@besttrousers
Honestly, "guy who says that we don't know what campaign strategies work because of a lack of RCTs" is a niche that needs to be filled.
G Elliott Morris
@gelliottmorris
heck, in the context of the moderation debate in the Dem Party right now, even the guy who got famous for popularizing uncertainty in political journalism is now firmly against the idea that measurement error and uncertainty about the future matter for interpreting the data. sad
-
G Elliott Morris
@gelliottmorris
self-described "moderate" new yorkers are voting mamdani over cuomo in this poll. interesting
Hon. Vickie Paladino
@VickieforNYC
Wow it’s almost as if Zohran is the candidate for rich rootless transplants who’ve never had a job and actual working people and New Yorkers know how to smell a con artist a mile away.
Go figure. -
G Elliott Morris
@gelliottmorris
In this article about why Harris lost in 2024, Ross Douthat does not mention "inflation" a single time.
G Elliott Morris
@gelliottmorris
After the election, voters told exit pollsters the economy was their #1 issue. The most common answer swing voters gave for why they voted Trump was that Dems were bad on the economy. Ignoring inflation in your analysis seems like a pretty big oversight.
www.gelliottmorris.com/p/the-strategists-fallacy -
G Elliott Morris
@gelliottmorris
this may be of interest/useful to people tomorrow. github.com/elliottmorris/county-election-benchmarks/tree/…
-
G Elliott Morris reposted
Verasight
@verasight_data
We’re proud to announce that Verasight played a central role in The Voter Poll by SSRS that six major news organizations will use as part of their 2025 Election Night coverage. Read our article on how Verasight’s nationally representative panel supported this landmark project.
-
G Elliott Morris
@gelliottmorris
Trump posts his worst-ever net approval rating in a CNN poll this morning, 37% to 63% (-26).
Will be on the FiftyPlusOne website shortly. Interested to see how this impacts the average — which has been trending away from Trump pretty aggressively for the last 2 weeks already
G Elliott Morris
@gelliottmorris
CNN poll: s3.documentcloud.org/documents/26211915/cnn-poll-trump-mi…
Average: fiftyplusone.news/polls/approval/president -
G Elliott Morris
@gelliottmorris
bat signal for political scientists and campaign practitioners: let’s stop with the endless tweet threads and settle this debate like real empiricists, with data and well-specified models substack.com/@gelliottmorris/note/c-172991814?r=a9pj
-
G Elliott Morris reposted
G Elliott Morris
@gelliottmorris
stuff like this is also a pretty big red flag. 'what matters is not how i built the car engine mr ford. what matters is whether the car runs'
-
G Elliott Morris
@gelliottmorris
I think pollsters should publish their data even when the crosstabs look fishy
BUT it is not true that small groups "SHOULD" have errors of 50+ points from sampling error. Unless you had literally only 4 ppl ( sqrt((0.4*0.6)/4) * 2 = 49%). In which case, don't publish that xtab.
G Elliott Morris
@gelliottmorris
Further, the existence of sampling error does not mean that you have to uncritically accept a poll as-is. If one subgroup is swinging 90 points right, and you have no other evidence from any other data that this is a reasonable swing, and that's in turn moving your topline 8 points, then it's sensible to think that maybe some sampling and/or non-sampling error have crept in.
G Elliott Morris
@gelliottmorris
In a Bayesian world, we acknowledge that observations (can) have uncertainty, and the amount of that uncertainty influences how much we update our beliefs. Normalize updating in small steps (or not at all)
G Elliott Morris
@gelliottmorris
stuff like this is also a pretty big red flag. 'what matters is not how i built the car engine mr ford. what matters is whether the car runs'
-
G Elliott Morris
@gelliottmorris
the right way to think about the generic ballot polls today is not D vs R
it's ruling party vs opposition party
in the last 5 midterms, polls shifted on avg by 6 points from July in the year before the midterm to Nov
Rs were down 4 at this point in 2010
James Blair
@JamesBlairUSA
Public Polling = Rorschach Test.
DEMs touting today’s NBC DEM+8.
Better rated WaPo = D+2; Better than same day ‘17 by 11.
Watch Current Avg., today = D+3
~7 pts better than same day ‘17. ~2 off ‘21.
🚨Key Takeaway: By historical standards, midterm outlook is fine rn for GOP. -
G Elliott Morris reposted
Mark Davin Harris
@markdharris
Small point but no such electorate has ever existed. You can’t weight to 24 and 21 at the same time.
Politics & Poll Tracker 📡
@PollTracker2024
2024 presidential recalled vote (Virginia)
🟦Kamala Harris 50.9%
🟥Donald Trump 45.2%
---
2021 governor recalled vote
🟥Glenn Youngkin 45.8%
🟦Terry McAuliffe 43.9% -
G Elliott Morris
@gelliottmorris
dems leading on the generic ballot with a party fav rating that is almost 20 points worse than for the republicans. interesting
Steve Kornacki
@SteveKornacki
NBC News poll
Trump - job performance
Approve 43%
Disapprove 55%
Generic ballot
Dem 50%
GOP 42%
Party image
Dem: 28% positive/53% negative
GOP: 37% positive/46% negative
www.nbcnews.com/meet-the-press/video/steve-kornacki-most-… -
G Elliott Morris
@gelliottmorris
some of my fav plots from a recent project:
a. zip code-level avg income vs individual probability of having Dem party ID, by region
b. zip income and individual Dem party ID for different races
c. county Dem presidential vote share and individual Dem ID for different races -
G Elliott Morris
@gelliottmorris
Study: Cable news significantly distorts public perceptions of how much politicians talk about economic issues. The effect of this is to significantly decrease the leverage of top-down party messaging strategy. It’s cable’s world, we’re just living in it www.dropbox.com/scl/fi/27v4xxenpo8wotkw1238e/rao-jmp.pdf
-
G Elliott Morris
@gelliottmorris
if we are going to have online magazines for Abundance, i think a wealthy patron should fund an online magazine for electoral reform — specifically for furthering the twin causes of proportional representation and democratic revival. seems like a comparatively neglected space
-
G Elliott Morris
@gelliottmorris
The backlash to ABC/Disney canceling Kimmel has been swift, and severe. The whole episode underscores something I've been blogging about for a while: A lot of people just don’t realize how unpopular Trump is. This leads to simply terrible decisions in both business and politics
-
G Elliott Morris
@gelliottmorris
as clear an example of the Fox News Effect as you're gonna get
-
G Elliott Morris
@gelliottmorris
Trump’s tariffs are a huge political loser, and he’s underwater on his handling of inflation in 48 states www.gelliottmorris.com/p/trump-is-underwater-on-trade-in-40
-
G Elliott Morris
@gelliottmorris
New analysis: Compiling our Strength In Numbers/Verasight polls from May to July, Trump is underwater on trade/tariffs in 40 of 50 states (also DC). In key swing states, approval of his trade policy is 10-15 points lower than his vote share in the 2024 election (20-30 on margin).
-
G Elliott Morris
@gelliottmorris
Trump now at -40 approval with young people, down from roughly even at the start of his term, per YouGov. The buyers remorse with Gen Z is real
-
G Elliott Morris
@gelliottmorris
There has been dramatically more protest activity against Trump in his second term compared to this point in his first term.
Yesterday's No Kings Day rallies across America mean the number of individual protesters could be higher now, too.
Trump is not a strongman. He's weak. -
G Elliott Morris
@gelliottmorris
keep scrolling
-
G Elliott Morris
@gelliottmorris
Trump is losing support at almost exactly the same rate as he did in 2017:
-
G Elliott Morris
@gelliottmorris
economic and political disaffection makes voters more elastic, across both the ideological spectrum and against the state. no reason to expect trump to hold on to econ-indexing voters if things don’t improve. + as the incumbent he’s now at an inherent disadvantage in bad times
Will Jordan
@williamjordann
Gallup just registered Trump's worst economic approval rating ever, at -12 (42% approve/54% disapprove).
Throughout Trump's first term it was a relative strength for him, and previously his lowest economic rating was -8 in June 2017. -
Post not available here, view on X
G Elliott Morris
@gelliottmorris
people are generally in favor of shared gov powers and do not approve of many of trump's/musk's early actions. they won't notice events immediately, but will react to consequences from economic fallout. perhaps that's why trump was quick to stop tariffs and the fed funding freeze
-
G Elliott Morris
@gelliottmorris
polling isn't broken, but we can finally let rdd die right?
This post is unavailable. See it on X. -
G Elliott Morris
@gelliottmorris
Save this link to 538's election-day/night/week(s) live blog. We have analysis from 538's reporters rolling out all day, & will start live coverage of the exit polls & results — esp. how they compare to our benchmarks (+ live modeling!) — at 6pm. Join in!
abcnews.go.com/538/live-updates/election-results-2024/?id… -
G Elliott Morris
@gelliottmorris
Say it with me, folks:
Trump and Harris are both a normal polling error away from a blowout.
The race is uncertain, but that doesn’t mean the outcome will be close. My look at the polls and forecast as time runs low for things to fundamentally change:
abcnews.go.com/538/trump-harris-normal-polling-error-blow… -
G Elliott Morris
@gelliottmorris
PSA for pollsters: 538 will be taking polls for general election races until Monday, November 4 at 11:59pm. If you want your data to appear in our aggregates we have to have them by then; We will update our models Tuesday AM (they take a while to run) so cannot accept late data.
-
G Elliott Morris
@gelliottmorris
$220m is just 1/1,000th of Musk's estimated net worth. As a point of comparison this is like if the median person sent Trump 100 bucks. Musk has a lot left to give
-
G Elliott Morris
@gelliottmorris
Quick note on the 538 House election model, which we published last week and have received a lot of positive feedback about!
On Friday, we adjusted the way that safe/solid race ratings are incorporated into lopsided races. Previously, we had removed this info from the model, but this ended up creating some weird situations where races that were preivoulsy rated as likely, then rated as solid, ended up moving towards 50-50 because the information "went missing" (in the model's eyes).
We have adjusted this now so the model does use safe/solid race ratings. This had no impact on aggregate probabilities for House control. projects.fivethirtyeight.com/2024-election-forecast/house/ -
G Elliott Morris
@gelliottmorris
In terms of the polls, RFK Jr dropping out of the presidential race is mostly not a big deal
abcnews.go.com/538/momentum-rfk-jrs-endorsement-give-trum… -
G Elliott Morris
@gelliottmorris
About Kennedy dropping out: Our polling average model looks at the difference in polls that do and do not include him as an option. In this adjustment RFK takes 1.3 points from Harris and 1.5 from Trump. RFK dropout net effect is small — reduces Harris’s margin from +3.3 to +3.1
-
G Elliott Morris
@gelliottmorris
538's Harris v Trump polling averages this morning:
National: Harris +2
MI: Harris +2.1
WI: Harris +1.6
PA: Harris +1.1
NV: Harris +0.3
GA: Trump +0.7
NC: Trump +1.0
AZ: Trump +1.9
projects.fivethirtyeight.com/polls/president-general/2024… -
G Elliott Morris
@gelliottmorris
News you can use! 538 now has an updating average of Kamala Harris's favorability rating. And wow... what a bounce. Harris was 17 points underwater in early July. Now she's at just -9 (tied w/ Trump) and quickly trending positive projects.fivethirtyeight.com/polls/favorability/kamala-ha…
-
G Elliott Morris
@gelliottmorris
Biden has been gaining ground in swing state polls, to the point where the EC-pop vote divide may now be less than one point (depending on how much weight you put on snapshot polls)
-
Post not available here, view on X
G Elliott Morris
@gelliottmorris
Biden has the very hard job of convincing 70% of voters why he is not in cognitive decline and why he should be running for president—and THEN must recover from a 36% approval rating and win. Blaming his problems on elites frankly does not help him do this
www.cbsnews.com/amp/news/poll-debate-should-biden-be-runn… -
G Elliott Morris
@gelliottmorris
Forecast update for this week: projects.fivethirtyeight.com/2024-election-forecast/
-
G Elliott Morris
@gelliottmorris
Trump and Biden are tied in 538's new election forecast abcnews.go.com/538/trump-biden-tied-538s-new-election-for…
G Elliott Morris
@gelliottmorris
G Elliott Morris
@gelliottmorris
How 538's 2024 presidential election forecast works abcnews.go.com/538/538s-2024-presidential-election-foreca…
-
G Elliott Morris
@gelliottmorris
Harry is right, but there is a lot of uncertainty with these crosstabs. Remember the polls showed much of the same race + age depolarization for most of 2020 — including implausible shifts left for non-col whites — that in hindsight turned out to be almost entirely response bias
(((Harry Enten)))
@ForecasterEnten
A lot has been made about Biden's problems with younger voters. Not enough is being made about his advantage with seniors, which is a shift from four years ago... and a big reason this race is still competitive.
It's fueling a registered voter/likely voter gap that helps Biden.This media couldn't be displayed. See it on the original post. -
G Elliott Morris
@gelliottmorris
The team at 538 has put together an absolutely stunning exploration of voting blocs — or "quiet caucuses" — in the US House. They used a clustering algorithm to assign each member to a caucus based on their voting record. The result is this interactive: projects.fivethirtyeight.com/types-democrats-republicans-…
-
G Elliott Morris
@gelliottmorris
Good morning! 538 is out today with averages of presidential general election polls.
The national polls are currently about tied, with a slight Trump lead that's within our uncertainty interval. Trump is tied or leads in all battleground states.
Story: abcnews.go.com/538/trump-leads-swing-state-polls-tied-bid… -
G Elliott Morris
@gelliottmorris
News you can use: 538's Swing-O-Matic is back for the 2024 election!
projects.fivethirtyeight.com/2024-swing-the-election/ -
G Elliott Morris
@gelliottmorris
My former colleagues at The Economist have Biden polling ahead of Trump now for the first time since September 2023. My own average is not so optimistic, but it's clear Biden has gained ground over the last few weeks
www.economist.com/interactive/us-2024-election/trump-bide… -
G Elliott Morris
@gelliottmorris
while we’re registering predictions, mine is that Biden’s best issue approval over the next few weeks will be on foreign policy/Israel, as compared to economy/inflation, abortion, guns control, health care, immigration & crime: i am not certain but fairly confident — maybe 80%
-
G Elliott Morris
@gelliottmorris
Here's my current thinking about a Biden primary challenge:
(1) It destroys the reputation of the challenger
(2) It turns a bunch of currently ambivalent Democrats into formal "losers" when they don't get their candidate
(3) But it could force a new VP pick
I don't think 3 > 2 -
G Elliott Morris
@gelliottmorris
The “70% of Americans support a ban on abortion after 15 weeks” statistic I keep seeing quoted must be cherry picked or come from private issue polling. In all the surveys 538 has collected on this in 2023, the public is 51-41 against a federal 15-week ban
Kellyanne Conway
@KellyannePolls
Life is the first of the unalienable rights. True leaders protect and progress Life rather than look or run the other way. Pro-lifers must lead with compassion, not judgment. A 15-week national standard for abortion is a concession, is common sense, and in line with the opinion of 7-in-10 Americans.
15 weeks is in the second trimester; a point when a baby can feel pain; when women know they are pregnant; know the sex of the baby; see its organs, heartbeat, and limbs. @sbaprolife
www.washingtonpost.com/opinions/2023/08/24/abortion-polit… -
G Elliott Morris
@gelliottmorris
this map would end the electoral college
-
G Elliott Morris
@gelliottmorris
looks like yougov is hiring a data journalist/survey analyst (fully remote too): www.smartrecruiters.com/YouGov1/743999911864069?trid=aa69…
the person who previously held this job worked closely with us on the weekly Economist poll. good colleagues and great survey/poll-reporting experience too -
G Elliott Morris
@gelliottmorris
The Economist is hiring a political data scientist to, among other things, take over our polling analysis and election forecasting endeavors from me when I leave. It’s a flexible job with good pay and excellent colleagues. Please share & DM any questions
infographics.economist.com/2023/job-political-data-scient…






























































































































































































































































































































































































































































































































































































































































































































































































































































