Nate Cohn

Nate Cohn @Nate_Cohn

RT @Neil_Irwin: Payrolls rise by 528k, unemployment rate edges down to 3.5%. Wow. This job market is still humming. — PolitiTweet.org

Nate Cohn @Nate_Cohn

@seeglazek evangelicals, Hispanics — PolitiTweet.org

Nate Cohn @Nate_Cohn

@jon_m_rob @davidshor @lxeagle17 @PiSomerandomguy it's really about mental health honestly. it's just not worth it to tweet — PolitiTweet.org

Nate Cohn @Nate_Cohn

@jon_m_rob @davidshor @lxeagle17 @PiSomerandomguy in all seriousness, my job isn't to make great graphics for twitter and i'm far too busy to do more, so if you're all going to complain about it it's just another reason to leave twitter — PolitiTweet.org

Nate Cohn @Nate_Cohn

@jon_m_rob @davidshor @lxeagle17 @PiSomerandomguy you'll miss the base R plots when they're gone, since the alternative is that you all don't get anything! — PolitiTweet.org

Nate Cohn @Nate_Cohn

@davidshor should have reported that with a hair more thought, as i just tossed a state dummy in there and the intercept was Alabama. taking the other four together, KS coefficient was .037 — PolitiTweet.org

Nate Cohn @Nate_Cohn

@davidshor coefficient for a KS state dummy? 0.07 — PolitiTweet.org

Nate Cohn @Nate_Cohn

Of course, there are other differences between theses states, but abortion's overperformance in Kansas survives controls for race, education, Evangelicals, etc. — PolitiTweet.org

Nate Cohn @Nate_Cohn

How abortion results compared to the 2020 presidential result in Kansas (blue) and pre-Dobbs votes in other states AL, LA, TN, WV (red) with similar ballot measures ('no' preserved state abortion right in all states) https://t.co/LNqdPCrboY — PolitiTweet.org

Nate Cohn @Nate_Cohn

This may not seem like much, but it's pretty rare. Reg. Democrats just don't really enjoy meaningful turnout advantages like this in high turnout elections, like this one. You could go back 30 years in Iowa or NC, where the data is accessible, and not find something quite like it — PolitiTweet.org

Nate Cohn @Nate_Cohn

Here's one hard turnout reading: in Johnson County, outside of Kansas City, 67 percent of registered Democrats turned out, compared to 60 percent of registered Republicans. https://t.co/EIxRH6yct7 — PolitiTweet.org

Nate Cohn @Nate_Cohn

@DKarol @jon_m_rob @ElectProject @mattyglesias i don't understand what you're saying, exactly. i'm saying that 'no' in KS just did better than 'no' did in those SD initiatives. if the dem turnout was more favorable in those years, that makes it all the more striking — PolitiTweet.org

Nate Cohn @Nate_Cohn

@DKarol @jon_m_rob @ElectProject @mattyglesias even the SD referendum in '06 that criminalized virtually all abortion did better than this KS referendum — PolitiTweet.org

Nate Cohn @Nate_Cohn

@DKarol @jon_m_rob @ElectProject @mattyglesias similarly, if you model the KS result based on county-level data from abortion referedums where 'no' was the pro-abortion option, including LA, you wind up with ~ a tie in Kansas, like the national polling — PolitiTweet.org

Nate Cohn @Nate_Cohn

@DKarol @jon_m_rob @ElectProject @mattyglesias obviously. but this was result was consistent with state-level estimates based on national polling data about abortion rights, which accounts for the racial composition of the state and so on, while this Kansas result is a net-20 pts to the left — PolitiTweet.org

Nate Cohn @Nate_Cohn

@jon_m_rob @ElectProject @mattyglesias i think that latter clause is entirely possible -- after roe, these votes have consequences that they didn't before. but i don't think this result simply follows some preordained pre-roe pattern — PolitiTweet.org

Nate Cohn @Nate_Cohn

@jon_m_rob @ElectProject @mattyglesias i don't think this can be reasonably interpreted as status quo bias -- or, if it is, it represents a change in status quo bias from prior votes https://t.co/3aJni6Hmmi — PolitiTweet.org

Nate Cohn @Nate_Cohn

It doesn't seem like there was a *huge* split in the early v election day vote, as expected, even though Ds were ~50% likelier to turn out in the early vote than Rs. That raises the possibility Ds had the turnout edge on election day as well--and a rare, big edge overall — PolitiTweet.org

Nate Cohn @Nate_Cohn

It'll be a while until we get voter file data on this election, but I think there's a chance this was the largest partisan turnout advantage in a high turnout election that Democrats have enjoyed as long as I've followed election data — PolitiTweet.org

Nate Cohn @Nate_Cohn

We don't make calls--the Times generally just takes race calls from the AP. But I'd guess they're thinking hard about it at this point https://t.co/p7uV65UcVa — PolitiTweet.org

Nate Cohn @Nate_Cohn

We've now got 12 Kansas counties--almost all rural GOP--where the vote appears to be completely wrapped up, and 'yes' is running is a net-14 pts (using margin to avoid confusion) behind our benchmarks for a tied election in Kansas, based on abortion referendums elsewhere. — PolitiTweet.org

Nate Cohn @Nate_Cohn

@bendreyfuss @Bearcat50 it was wrong. go listen to arnon's interview on the network at like 1am or 5am or whatever. they thought the remaining vote was democratic because it was mail. it was R+20. — PolitiTweet.org

Nate Cohn @Nate_Cohn

It was bad https://t.co/ZiM4DOzaen — PolitiTweet.org

Nate Cohn @Nate_Cohn

And for clarity: '6 pts' was in terms of vote share (like 56 v. 50), not margin (56-44 = 12) — PolitiTweet.org

Nate Cohn @Nate_Cohn

*cxn: 7 counties over 90 percent counted, not 75 percent — PolitiTweet.org

Nate Cohn @Nate_Cohn

Some of you may find waiting for largely completed data to be a little tedious, but this is the only sure-fire way to avoid making a bad Arizona call in the absence of data by method of voting. Sorry if it's slower than you'd like! But there's clearly a big early-eday split — PolitiTweet.org

Nate Cohn @Nate_Cohn

We're now up to seven Kansas counties where at least 75 percent of precincts are counted, and 'no' is running about 6 pts ahead of our pre-election benchmarks for a tied election, based on recent referendum results — PolitiTweet.org

Nate Cohn @Nate_Cohn

@NapervillePol well it would be pointing about 63% 'no,' but we wouldn't be publishing it since it doesn't have a great way to handle early voting unless it gets election results by method of voting or can borrow statistical power from states where the vote is already complete — PolitiTweet.org

Nate Cohn @Nate_Cohn

One thing I will say: no one should be surprised that this is close. Estimates based on national polling have long shown Kansans fairly divided over abortion, and that's basically confirmed by recent abortion referendums where abortion runs well ahead of Dems in white, GOP areas — PolitiTweet.org

Nate Cohn @Nate_Cohn

Based on recent abortion referendums, I had Phillips giving 72% of the vote to 'yes,' v. 65% so far (and there's a little election day vote left). It's one tiny county, so don't read much into it, except to say it'll clearly tighten up a lot — PolitiTweet.org