Nate Cohn

Nate Cohn @Nate_Cohn

@jonathanchait not strictly speaking, but it could make biden the winner on election night without waiting for PA — PolitiTweet.org

Nate Cohn @Nate_Cohn

With more than 80% of 2016 voters already in, you might think it's too late for this. I don't really think so. Trump's strength in our poll is among Latino voters who haven't yet voted, probably aren't too engaged, and probably have conflicting views that ads can exploit — PolitiTweet.org

Nate Cohn @Nate_Cohn

A familiar move for Bloomberg, who made big last minute plays in red CDs in 2018 https://t.co/vCI8NxhytY — PolitiTweet.org

Nate Cohn @Nate_Cohn

A familiar move for Bloomberg, who made big, last minute plays in red target CDs https://t.co/vCI8NxhytY — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @alexburnsNYT: New: Bloomberg funding a last-minute TV blitz in TEXAS & OHIO to boost Biden in the biggest red states. ~$15 million in… — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @kwcollins: The paradox of campaign effects: As the election grows closer, the proportion of the electorate that is undecided goes down,… — PolitiTweet.org

Nate Cohn @Nate_Cohn

could you pick gary peters out of a crowd? https://t.co/pf9FrQB8dY — PolitiTweet.org

Nate Cohn @Nate_Cohn

@2020vbnmw just undecided — PolitiTweet.org

Nate Cohn @Nate_Cohn

Pct. undecided in NYT/Siena Senate polls of MI/NC, by wave: June: 21% Early Oct.: 14% Late Oct. (partial data): 7% — PolitiTweet.org

Nate Cohn @Nate_Cohn

Pct. undecided in MI/NC Senate polling, by wave: June: 21% Early Oct.: 14% Late Oct. (partial data): 7% — PolitiTweet.org

Nate Cohn @Nate_Cohn

Most early voting is just one party banking votes it was going to get anyway. In TX, we don't really have a great sense of what kind of electorate we were going to get 'anyway' https://t.co/ZKYh7qssqj — PolitiTweet.org

Nate Cohn @Nate_Cohn

And this is the kind of thing that gives me pause about where we're headed here in Texas. No one can honestly tell you what this electorate is going to look like. In a lot of states, I feel comfortable saying we're just watching cannibalization when Dems run up the score. TX? IDK — PolitiTweet.org

Nate Cohn @Nate_Cohn

In our poll (the one that shows Trump+4), these folks do lean Republican by voter file based measures of partisanship, but they back Biden by 7 https://t.co/5IynFNWbik — PolitiTweet.org

Nate Cohn @Nate_Cohn

That's mainly based on data that was conducted before the debate, at least in our own average and I'd guess 538 as well. There was certainly indication of 'tightening' in surveys conducted early last week, as we wrote about that the time https://t.co/pTwOLvFpd6 — PolitiTweet.org

Nate Cohn @Nate_Cohn

I'm a little surprise we don't have a bit more post-debate data, though we might get some tomorrow. That said, this is a case where patience will pay. Every day without 'final' polls means that we'll get them later in the week, closer to the election. That's a trade I'd take — PolitiTweet.org

Nate Cohn @Nate_Cohn

We don't have a lot of polling that's fully post-debate, but we do have quite a bit that's at least partly afterward and I don't really see any hint or indication of a shift. — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @Redistrict: A great read on the complexities of TX's Hispanic vote (h/t @amyewalter). https://t.co/cMMi7nbAxm — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @Garrett_Archer: Here's the just-pulled-seconds-ago Maricopa county Early Ballot update: 2020 Total: 1,157,731 R: 417,221 D: 452,025 O… — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @carlosodio: @Nate_Cohn That's exactly where we were with Hispanic men a month ago: 43-43. We had the Latinas at 61-22, so *maybe* there… — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @PatrickRuffini: UPDATE to our Multi-Party Democracy numbers. Only 1 point separates a center-left Labor Party from a Conservative Par… — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @Cmkahn: Coming today: Reuters/Ipsos polls in WI, PA (previous wks...) PA Oct. 13-19: Biden+4 Oct. 6-11: Biden+7 Sept. 29-Oct.6: Biden… — PolitiTweet.org

Nate Cohn @Nate_Cohn

@conorsen one framing of our poll might be that Trump gets the incumbency/Abbott numbers with nonwhite voters and the Beto numbers with white voters — PolitiTweet.org

Nate Cohn @Nate_Cohn

@conorsen one framing of our poll might be that Trump gets the incumbency/Abbott numbers with nonwhite voters than the Beto numbers with white voters — PolitiTweet.org

Nate Cohn @Nate_Cohn

Looking under the hood, quite a big gender gap among Hispanics in our Texas poll https://t.co/R0FSdtXauR — PolitiTweet.org

Nate Cohn @Nate_Cohn

@carlosodio sorry--wrong hispanic race variable there, but still similar 70-24 among women; 43-44 among men — PolitiTweet.org

Nate Cohn @Nate_Cohn

@carlosodio Or if you prefer, Trump at 43 with men, 22 with women — PolitiTweet.org

Nate Cohn @Nate_Cohn

@carlosodio We also have a pretty big difference there--Biden at 70 with women, 45 with men — PolitiTweet.org

Nate Cohn @Nate_Cohn

But I do think this is why our AZ polls, for ex, are suddenly quite a bit better for Democrats. We know it'll be high turnout and there's nothing in our data holding us back from it — PolitiTweet.org

Nate Cohn @Nate_Cohn

This year, I think there's far less room for that kind of turnout error, since we start on the assumption of a high turnout in presidential elections. I do think there's still some risk in this area in TX--no one knows what this electorate will look like — PolitiTweet.org

Nate Cohn @Nate_Cohn

Take TX23, for instance, where we missed most badly. 18 to 29 year olds were only 6% of the electorate. Hurd was only up 4 among RVs. — PolitiTweet.org