E.J. Fagan
ejfagan.com
E.J. Fagan
@ejfagan.com
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.
Republicans won California all but once between 1948 and 1992, including the close 1976 election. Two of the four elected Rep presidents were from CA, plus one VP and the Chief Justice. They held Governor from 1983-1999 too. Flipped in 1992.

Makes me think about Democratic possibilities in Texas.
November 9, 2025 at 5:56 PM
Just wow
November 6, 2025 at 12:59 PM
This is one hell of a paragraph:
November 6, 2025 at 12:54 PM
Ecological fallacy caveat aside, I think that the clear story here is that the non-white working class gains that Trump managed to win and make NJ look like a swing state are reversed, but Ciattarelli held white voters pretty well. Consistent with AP-NORC national polling crosstabs.
November 5, 2025 at 4:10 AM
Middlesex is 95% in. Merrill is running 9 points ahead of Harris. Super diverse, educated. 20% South Asian, 40% White, 20% Hispanic. New Brunswick, Rutgers, Edison. Very dense suburbs.
November 5, 2025 at 3:16 AM
Hudson is 78% in. Huge county and one of the densest in the country anchored by Jersey City and Hoboken. Super diverse immigrant communities and people who commute into New York. Sherrill running 11 points over Harris. 40% Hispanic, 17% Asian

Pretty clear that the working class realignment is off
November 5, 2025 at 3:09 AM
Running close to even with Harris in Ocean, the NJ Republican Party's best county by far. White, very rich area mixed with Springsteen working class. Upper Jersey Shore. Old Money.
November 5, 2025 at 2:59 AM
She's only running even with Harris in Hunterdon, a less dense but relatively educated and wealthy county on the PA border with 95% of the vote in.
November 5, 2025 at 2:54 AM
Bergen, my ancestral county, has 64% of the vote in. She's doing five points better than Harris. Mix of very, educated rich suburbs and immigrant communities. There's a bunch of white educated towns here that haven't flipped to Democrats mostly because of age demographics.
November 5, 2025 at 2:52 AM
Here's Passaic County, a mix between big immigrant cities and whiter, less educated (for NJ at least) suburbs. It was shocking to see Democrats lose the county in 2024.

Only 36% of the vote in, but looking good for Democrats so far.
November 5, 2025 at 2:50 AM
Austin, TX in November:
November 4, 2025 at 7:16 PM
How do people define political polarization for their classes? Here's how I've been introducing the concept. I think it works well.
November 3, 2025 at 7:58 PM
Shifts in the national environment are never uniform, but here's a bunch of Cook PVIs. 2018 was D+8. They held (with incumbency advantages) Montana, Ohio, West Virginia. Lost Florida by a fraction, Texas/Indiana by 2 points, Missouri by 3 points.
November 2, 2025 at 7:51 PM
US Patent #10262281B1, the Dominos PizzaTracker. Some of our finest work promoting the Progress of Science and useful Arts.

patentimages.storage.googleapis.com/13/9a/21/e8d...
October 30, 2025 at 1:02 AM
The heuristics they use are fun:
October 28, 2025 at 4:34 PM
AP-NORC has Trump's favorability with Hispanic voters inching closer to his approval with Black voters. He was almost breaking even this time last year.
October 24, 2025 at 10:42 PM
Bawn et al but in 1947:
October 21, 2025 at 8:14 PM
I mean if we're going deep:
October 21, 2025 at 7:59 PM
October 21, 2025 at 6:59 PM
In 2021, Van Orden decided to confront a kid working at a local library over this book:

This is a Trump+8 (and Hovde +6) district, but Van Orden beat Cooke by just 3 in 2024.
October 20, 2025 at 3:34 PM
Here’s how they define it:
October 19, 2025 at 9:27 PM
The spectrum of the party network, according to me.
October 13, 2025 at 9:27 PM
Just to clarify how dumb and political it is to bring in the Texas National Guard to allegedly protect a federal facility in Broadview, here is the distance from the Illinois National Guard base nearby compared with Texas.
October 8, 2025 at 3:02 PM
The gap is very stable over time though. In fact, really stable between left-right parties across countries.

It's a big reason why I believe spatial models are a bit oversold. It matters more which issues are salient, often outside of a party's control.

journals.sagepub.com/doi/full/10....
October 7, 2025 at 1:44 PM
Very skeptical of this question wording. My guess is that the party out of power will always look worse when asked about a plan, because one does not reasonably exist until at least the presidential nomination campaigns.

“Trust to handle” is the classic wording.
October 7, 2025 at 12:58 PM