@lxeagle17 @SplitTicket_ I think what bothers me about this is just the attitude that "We're from academia and we're here to help!" (even though lots of people with >= qualifications have done better work on this question) and it turns out for the 812th time in a row there's Great News for Democrats.

RT @JerryEngelmann: Which NBA teams have the brightest futures?. Here are my takes on the 15 teams I ranked in the TOP half of @NateSilver5….

This was a good one!.

RT @lxeagle17: Had fun talking with @NateSilver538 about the WAR wars. I was a bit surprised at what I thought were elementary mistakes in….

What I'm less ambivalent about: the recent Bonica and Grumbach post on this subject was awful. Pulling rank and rude to @lxeagle17 and @SplitTicket_ while making a lot of obvious mistakes. Made me a little depressed about the state of academic political science tbh.

Me on the WAR wars. Having studied this a LOT, I do think moderate candidates win more often, but it's getting more complicated in an era of hyper-partisanship.

Starting soon!.

We're chatting at 1pm (Eastern) tomorrow!.

RT @AlecMacGillis: Really striking to what degree NY has become an outlier in recovering from the remote-work shift, especially given that….

They're also often also sycophantic because they're captured by an audience who don't know or don't care that its BS, typically because they like the implications of the BS.

It takes a special type. But certain humans act like hallucinating LLM when out of their depth. They use all the right words but not in the right order. When confronted with logical flaws they're like "oops" and reconfigure to a slightly better answer but still asymptote at BS.

RT @JerryEngelmann: Which NBA teams have the bleakest futures?. Here are my takes on the 15 teams I ranked in the bottom half of @NateSilve….

Whole thing here. We're probably due for another, more zoom-out-y take on redistricting soon, too.

Are the headlines right that Texas restricting would cost Democrats 5 seats? Could it be more? Could it turn into a dummymander?. @emckowndawson ran the numbers at the newsletter:

Maybe I'm a weird outlier, but the peer review process is obviously broken. Maybe it's better in the fields I *don't* know well. But I'd be surprised if that's true.

Join Hillsdale College President Larry Arnn, U.S. Secretary of Defense Pete Hegseth, and other leaders to remember and honor the stories of great Americans in our new video series:

Well, I know a lot about statistical inference, have been doing it for 25 years, have faced a lot of public scrutiny, and in the fields where I also have a lot of domain knowledge, probably half of published papers have obvious fatal flaws that render them unfit for publication.

You ran some cool regression analysis OK great. Make some nice graphics and put it on a Substack. Engaging headline, 1500-2500 well-written words. That's literally 100x faster than trying to publish in a journal and it's better peer review anyway.

The most important research paper of the past 10 years is the Google transformer paper ("Attention Is All You Need") and it was written by non-academics and published in an open-access journal.

Academic journals might be a lost cause but they'd probably be better if you had some non-academic practitioners serving as reviewers. Journalists have their problems too but they have much better bullshit detectors, for instance.

The AI's are incredibly helpful at debugging code, I think maybe their single best use case including *writing* code. But half the time the problem they (correctly) detect is like "you misspelled 'if' as 'uf' in line 672".