Nate Cohn

Nate Cohn @Nate_Cohn

The turnout may offer additional upside to Warnock. According to our estimates, Warnock would have won the '20 general electorate by 4 percentage points, and that wasn't a great Dem turnout either -- see runoff. He would have won '16 and '18 as well (and nearly '14) — PolitiTweet.org

Nate Cohn @Nate_Cohn

I'll be honest, I think it's pretty hard to tell a story about why Walker wins this election. That's not to say it's impossible -- anything can happen. I think it would just be hard to explain how and why, given the candidates and the hard results last month — PolitiTweet.org

Nate Cohn @Nate_Cohn

Happy Election Day A few morning thoughts on the Georgia Senate race, which is close and yet seemingly fairly clear https://t.co/iiDzuNKmjY — PolitiTweet.org

Nate Cohn @Nate_Cohn

@davidshor i'm not sure that means what you're implying it means — PolitiTweet.org

Nate Cohn @Nate_Cohn

It's true. https://t.co/VOGCgMkHL2 — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @pollhannes: Over the past few weeks we've been digging into our errors in our 2022 polling. You can read about our initial findings her… — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @jon_m_rob: .@Nate_Cohn last April after the passage of SB 202 vs. @NYTnickc piece recapping turnout in GA this cycle. @tbonier says it'… — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @PollsAndVotes: At @kkondik request, here are results by college grad/non-grad. Note ns are modest so large MOE for the crosstab. https:… — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @PollsAndVotes: When asked the question: If it were a choice between just the two of them, who would you prefer as the Republican nomine… — PolitiTweet.org

Nate Cohn @Nate_Cohn

I'm only willing to support this if they start publishing their election results by precinct and method of voting statewide https://t.co/BTs6qJj7Pi — PolitiTweet.org

Nate Cohn @Nate_Cohn

@tbonier i've got 28.7, less than 20 — PolitiTweet.org

Nate Cohn @Nate_Cohn

Over in the newsletter this AM, we've got some interesting data on fairly low Black turnout in GA, NC, LA and elsewhere. (Tough news for Democrats in general, though oddly a decent reason for Democratic optimism in the GA runoff) You can sign up here https://t.co/vfLG2cPxwD — PolitiTweet.org

Nate Cohn @Nate_Cohn

https://t.co/pQekYVvGK0 — PolitiTweet.org

Nate Cohn @Nate_Cohn

@mattyglesias maybe i'm misinterpreting, but it sounds like it didn't wind up politically consequential, then? — PolitiTweet.org

Nate Cohn @Nate_Cohn

@PatrickRuffini Like, an instant run off? — PolitiTweet.org

Nate Cohn @Nate_Cohn

@tbonier @CookPolitical if excluding uncontested races means what it sounds like, i do think that's pretty clearly worse than including — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @UpshotNYT: https://t.co/ISKgGfCohf — PolitiTweet.org

Nate Cohn @Nate_Cohn

Now down to 52-48. “Yes” was over 60 when I saw this first — PolitiTweet.org

Nate Cohn @Nate_Cohn

Yes with the early lead, but “no” has been closing — PolitiTweet.org

Nate Cohn @Nate_Cohn

@cwarshaw @CERVASJ @SeanTrende @databyler sure, i mean, that's my position in this discussion. that said, traditionally the popular vote is not reported with imputation, whether it's an NYT results page, a wikipedia article, a table in a book article, etc. — PolitiTweet.org

Nate Cohn @Nate_Cohn

@cwarshaw @CERVASJ @SeanTrende @databyler sure, i mean, that's my position in this discussion. that said, traditionally the popular vote is not reported without imputation, whether it's an NYT results page, a wikipedia article, a table in a book article, etc. — PolitiTweet.org

Nate Cohn @Nate_Cohn

@cwarshaw @CERVASJ @SeanTrende @databyler but i don't think he would propose to look at, say, the mean-median gap without imputation — PolitiTweet.org

Nate Cohn @Nate_Cohn

@cwarshaw @CERVASJ @SeanTrende @databyler i'm a little more hesitant about that, since most of the reason for citing the national popular vote is for some kind of analytical purpose where imputation may be desirable or outright necessary — PolitiTweet.org

Nate Cohn @Nate_Cohn

@cwarshaw @CERVASJ @SeanTrende @databyler right, and i don't think he's contesting that. i think the extent of his position is that it's acceptable to report the current national vote as R+4, without imputation — PolitiTweet.org

Nate Cohn @Nate_Cohn

@CERVASJ @SeanTrende @databyler he’s referring to simple references to the national popular vote in an election — PolitiTweet.org

Nate Cohn @Nate_Cohn

@SeanTrende @databyler in the 'just reporting' category, i could even advance a principled argument against imputation — PolitiTweet.org

Nate Cohn @Nate_Cohn

@SeanTrende @databyler for me, the purpose is very relevant. if we're doing something sensitive, like the mean-median gap or a state popular vote tally or something, it feels like a must. if we're just reporting the popular vote outcome, no imputation is fine — PolitiTweet.org

Nate Cohn @Nate_Cohn

@SeanTrende @databyler i know you know this, but the differences are often pretty material! it probably does make a difference whether we should think of this as an R+2 or R+4 year for a lot of purposes — PolitiTweet.org

Nate Cohn @Nate_Cohn

@SeanTrende @databyler yeah, i mean, the literature is still out there using unadjusted ANES and so on lol. anyway, i'm not trying to start a movement. but i do think it's unequivocally better to impute the uncontested races if it's a sensitive analysis, like PV-v-tipping point — PolitiTweet.org

Nate Cohn @Nate_Cohn

@databyler @SeanTrende i do think 2 is pretty plainly the right choice for any serious analytical purpose. we certainly did in '18. there were people touting dems winning the FL house popular vote based on uncontested seats! — PolitiTweet.org