ah well,,,

gotta wonder how accelerationist you could get here. get some state to draw a new map every odd-numbered year, getting more aggressive in cycles that look favorable for the governing party and more defensive in cycles that look bad for them.

oh my god, he admit it x.com/Acyn/status/...

I try to highlight the distinction between "reviewing" the literature (to prove to the reader that you've read) and "using" the literature (to teach the reader something). Each paragraph should make a point, those points should themselves fit together, etc. But it's hard to get out in front of!

my impression is that most of her audience is other journalists

so you don't want to maximally stretch all of your voters across as many R+5ish districts as you can (which Indiana just proposed), because in a D+7 year you lose all of them at once.

that's why you keep it R+10 or more, why maps like Indiana's newly proposed one (or the one that didn't make it through in Utah) are super risky, and another reason why mid-decade redistricting is bad form. the "game" is to try and maximize seats over a ten-year period w/ varying national moods

Like, re: OP's complaint here, you could draw maps to maximize the competitiveness at the seat level...but then really small fluctuations in the national mood will generate huge volatility in the chamber. That also seems bad!

I'm sure someone has written an impossibility theorem for single-member plurality district maps. They are simply bad ways to select legislatures, given all of the different things we want individual representatives and whole chambers to do.

in this sample, the lowest-income and highest-income students get accommodations at about the same rates while everyone else is less likely drive.google.com/file/d/1NPoR...

the audience for all of this stuff (and tbh, a lot of her 2024 writing) seems to mostly be other journalists

seriously