New WP! Philip O'Sullivan, Kosuke Imai, and I generalize an earlier SMC sampler for redistricting plans, allowing it to scale better & be applied to multi-member districts, too, like the Dáil Éireann (Irish parliament) below. Look for a new 'redist' version soon(ish)! arxiv.org/abs/2603.22188

Atrocity upon atrocity. Every detail here is worse than the last. www.theguardian.com/world/2026/m...

Strength In Numbers is looking for a smart part-time survey research assistant (or a few) www.gelliottmorris.com/p/strength-i...

LA City Council’s City President Marqueece Harris-Dawson says he has been pulled over four times while a member of the council; the most recent was this week. He missed a committee meeting because of the stop, according to him. Among the questions: “how do you have this vehicle,” according to him.

I just cannot underscore enough how uselessly simplistic this WAR-based mode of thinking about elections is. And it’s not even accurate! In Jain’s own construction WAR explains 25% of future residual candidate performance, including fundraising, candidate skill, district fit, & general vibes-y stuff

v0.2 of `bases` is up on CRAN! `bases` brings nonparametrics into your favorite modeling functions—using random Fourier features is as easy as sticking `b_rff()` into your formula. v0.2 brings graph Fourier & random convolutional features, `mgcv` integration, & more! corymccartan.com/bases/

I'd very much like to return to a world where academia, the sciences, the media, etc. *can* be neutral because everyone accepts their right to flourish, but that's not the reality we're living in.

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-d...

Redistricting Reforms Reduce Gerrymandering by Constraining Partisan Actors is now forthcoming in the APSR! We use a game theoretic treatment and continuous DiD(iD) to show when redistricting reforms work with @corymccartan.com, @simko.bsky.social, Emma Ebowe, Michael Zhao, and Kosuke Imai

Specifically, the “issue” is/was my (non-)compliance with Regulation 9.016. A similar presentation I gave in Finland in October was also flagged for violating this same regulation because I had “transgender” in the title there too. www.flbog.edu/wp-content/u...

We need more people to know about this simple solution to gerrymandering ⬇️ Here's how proportional representation could make map manipulation impossible, and it’s actually achievable. Watch and share to spread the word: https://youtu.be/xN5V0au9Xms

finally: to the extent that it's possible safety isn't found in secrecy, it's found in numbers, and the former is at odds with achieving the latter.

🇺🇸 trifectas get <one> shot at doing something Obama chose policy (ACA). It didn’t fix the authoritarian problem Biden chose policy (ARP/IRA). It didn’t fix the authoritarian problem The platform must be <immediate> institutional reform next time. Right from the start, as @dhnexon.bsky.social said

Worth serious consideration and engagement, whether or not you end up agreeing with it. jacobin.com/2026/02/uaw-...

new newsletter post up about all those folks who insist they "can't have an open conversation about AI"

these people are sick

Anyway, I am skeptical of many suggestions of the reform movement (eg preregistration, registered reports, etc) because they do not address the underlying incentives, nor do they alter the purpose of the system. 9/N

At what point should universities that practice this kind of censorship lose their accreditation?

2nd chart is interesting - swing was not correlated with turnout. Consistent with a mostly persuasion effect

One thing I hope people take the time to acknowledge: The sea change in public opinion has been achieved because violence, though encourage by a lot of people here online, was never once employed by the movement. Violence was solely the purview of the state and that is why the state is losing.

The administration is boxed into a corner right now and its entirely the result of the incredible discipline of the people in the streets. The Insurrection Act angle was always predicated on ICE brawling with the black bloc in street. Without that, it's worthless.

The most succinct description of the murder I've seen.

This is when the Linz stuff comes to a head. Two competing branches of government with competing claims of legitimacy in open conflict.

“Everyone should be allowed to carry a gun for their safety but the government can murder you in the street if you have a gun” is an incredible endpoint for the conservative moment

There is always _a_ reason. Even in the most repressive state. It is usually "they didn't go when we told them to go"

I am trying not to post on these events in this state of mind but: I hope people understand what the observers are doing is brave and dangerous.

It’s currently operating totally outside of our constitutional rules. Habeas corpus , due process of law, equal protection, freedom of speech, 4th amendment protections against search and seizure, are all not being enforced in any common sense or meaningful way.

Good to see another aggregation of public poll results, but I wouldn't describe these averages as transparent at all! The methodology page doesn't share any details about how weights are determined or how they are "empirically grounded." Cf the methodology at fiftyplusone.news/methodology

This WSJ story about the unconstitutional secret ICE administrative warrants memo shows everything that's wrong with legal journalism under a lawless presidency. The TLDR is that it falsely presents unconstitutional conduct as law-creating. Very important thread 🧵🧵🧵 www.wsj.com/politics/pol...

#SCOTUS is likely to do the right thing in the Lisa Cook case. But it's worth bearing in mind that this idea that the Fed is some bespoke exception to the unitary executive theory is utterly incoherent as a matter of both law and history—and is really just proof that the UET is itself bollocks:

Last fall I shared new methods research with @shirokuriwaki.bsky.social on ecological inference—inferring individual relationships from aggregate data. Our new review WP frames past EI methods as linear models, and argues credible EI requires controlling for covariates arxiv.org/abs/2601.07668

it has been truly amazing to see certain people who would normally acknowledge things like this contort their positions in crazy ways just to avoid admitting something that might hurt their ability to win contracts (campaign messages don't matter as much as conditions and wouldn't have saved Harris)

Incredible research and really good summary thread. Hopefully we can take more intentional steps to foster success for FGLI students in the academy

Unpopular opinion: computer science has to publish fewer papers per year and rethink its peer review model

Study after study shows campaign ads barely move the needle. So where does money’s real power come from? I ranked the five ways money corrupts politics—from least to most corrosive. What I’ve learned from 15 years of tracking political money:

Interestingly, the TX gerrymander doesn't have any net impact in a D+12 environment Plots below show new effect of redistricting this cycle—cf left with the changes vs right if the 3-judge ruling holds

As a follow up to this: In the area where I hang out, I have big concerns about external validity. I see a lot of, "when a local official says X, people trust elex more!" That's important, sure, but... The house-on-fire emergency is the whole election denialism movement. Something about deck chairs

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-s...

This is an impressive project. My reaction to what it shows though is that survey experiments have gotten out of hand in polisci. I will blog more on this, but I do not think survey experiments are emblematic of the credibility revolution. Some are already interpreting as such, which is a problem.

Democrats fail to flip TN 7, but it's another big overperformance Blue shift of ~14 pts, which is nearly on par with shifts in other U.S. House specials this year Ds hit vote target in Nashville, but more rural counties didn't shift as much Turnout notable: ~180K total votes, matching '22 midterm

Interestingly, the TX gerrymander doesn't have any net impact in a D+12 environment Plots below show new effect of redistricting this cycle—cf left with the changes vs right if the 3-judge ruling holds

Davidson/Nashville swinging D+25 is wild Most swing districts next year won't look the way TN-07 does, with mainly rural counties that are shifting D by less

D+10-15 national environment would mean Dems end up ~100 seats over Rs in the House tinyurl.com/cmchousemodel

Only early vote so far (see www.elections.tn.gov/county-break...) and unclear how/if this model handles vote method, but notably Montgomery county is only D+3 in the early vote. My feeling is it would take combo of massive turnout in Nashville + e-day overperformance district-wide for Behn to win

New paper in press at JPSP! An adversarial collaboration focusing on a large-scale test of how strongly implicit racial attitudes predict discriminatory behavior. Pre-print here: osf.io/preprints/ps...

The guy they interview was also on Trump's data team! No hard questions asked about how "independent" this effort is

To me this is a depressing theme in modern academia. There is so much work being produced, and so many competing demands on our time, that people rarely seem able to just closely read work and frankly say "yes, I believe this" or "no, I don't." If we aren't doing this, what _are_ we doing?!

Interesting to think about the mechanism here Poetic language evidently signals that the preceding system prompt should be interpreted in a different, less literal, context. One could imagine learning the direction in model-space that leads to this reframing, and optimizing prompts to get there

"Before folks assume that one side of this fight clearly acted in bad faith and the other didn’t, it’s worth indulging the possibility that Purcell itself is the culprit—and that its standardless-ness creates perverse incentives for lower courts in election cases." Me on the TX redistricting case:

IF it holds, this would net Dems 1.6 seats, on average, due to mid-decade redistricting (D+1 seat in a Dem-favoring environment)