E.J. Fagan
@ejfagan
Associate Professor of Political Science, University of Illinois Chicago. Author of The Thinkers: The Rise of Partisan Think Tanks and the Polarization of American Politics. Also, baseball.
A smaller but still important gap has opened up over the last year between polls with live interviewers and polls conducted with address-based online probability samples.
Added Echelon, TIPP, HarrisX non-probability panels. Look at the consistency of the y-intercept on approval.
Some thoughts. First, if I were running a real polling aggregator, I would exclude non-probability samples entirely. This choice erases most of the weighting variation. The top line doesn't change a ton in my dashboard, but I only included two non-probability samples. Others include many more.
Okay, now weighting methodology. Smaller but consistent differences here as well. Unlike sampling, I don't have any real opinion on the best choices. Trump/Rs do 2-3 points better if you weight for Party ID or recalled 2024 vote. Some of this variation is correlated with sampling methodology.
But the real story are methodological differences. Let's start sampling. There's a pretty big difference in both series by methodology. Non-probability online panels are strongest for Trump/R by a large margin. On the generic ballot, phone RDD-ish polls are ~D+8 versus ~D+4 for non-prob.
I/Claude created a polling average tracker of high quality polls using a variety of methodologies. Ignores A/RV/LV differences. Whether you think 2026 will be a bad for cataclysmic year for Republicans depends entirely on polling. methodology. Topline: Approval 37.6%. Generic ballot at D+6.3 1/
CNN poll series: Does the president have the right priorities or has he not paid enough attention to the most important problem? Trump’s overall approve hit a new low at 34%, but is down just a few points this year. Much bigger decline on priories. Voters notice that he doesn’t care about inflation
Even on the low end, the data are super clear for undergraduate degrees. BAs make far more than less educated groups, and the BA opens the door to even higher income professional degrees. The income from a typical plumbing or carpentry job looks great when you are 24. It doesn’t scale though.
Here’s a count of workers in the trades from a building trade association. In a country of 350 million, even doubling the trades does little to replace our higher education system. The reality is that cutting off college will push the marginal student toward a low wage service job.
When candidates are named, Democrats are running 3-4 points ahead of the polling average in the marginal races (OH/AK/TX/IA/NE).
A masked vigilante in Mexico is attacking motorcycle thieves and leaving them taped up to telephone polls. www.yahoo.com/news/world/a...
My credentials: I know a thing or two about how Heritage (and CAP) have done this in the past.
Good news for Democrats: only 5% of Maine voters say they haven't heard of Platner's scandals. Any impact is mostly baked in and he's still ahead.
People are freaking out that Democrats might nominate El-Sayed in Michigan. But he's polling even or better than the other candidates pretty consistently.
Here is the distribution of voter fraud cases in the Heritage Foundation database. Most of the cases are either people casting inappropriate absentee ballots (a lot of these are double-voting in two different states I think) or people voting with eligibility issues.
Cannibalism actor comes back from cancellation as a conservative slop guy.
After a slow-ish 2025, Democratic small dollar fundraising (via ActBlue) is way up in 2026.
For context, here’s what Chicagoland garbage trucks (private contractor) look like right now. We like Chicago Pope.
Weird things happen when you're staring down a wave election. Democrats are within striking distance for governor in Nebraska.
Dems now up to +6.5 on the @fiftyplusone.news average. Finally starting to pull away into the landslide territory that you would expect given Trump’s approval rating.
Trump hits a new low in the @fiftyplusone.news approval rating tracker at 36.5%. Democrats also have their biggest generic ballot lead at 5.7%.
@gelliottmorris.com Is it normal for a party's vote share to decline over the course of the election year? I figured that the growth of the in/out party gap was due to undecideds resolving for the out party, but the Republicans are also losing some people who said they would vote for them last year.