NYTimes/Siena: Biden+6 in AZ, +3 in FL, +6 in PA, +11 in WI

[afleitch]

Expected.  WI appears gone, but everything else is still in play.

With early voting it's a case of Schrodinger's Vote. Things may actually not or have never been in play on the inside even if outwards it looks like it could.

[Landslide Lyndon]

Biden winning seniors everywhere except Arizona (Trump 48/47) and independents by crushing margins.

[CellarDoor]

Happy to hear that they opted for a higher N for these polls, especially when compared to Selzer/Monmouth.

[Orser67]

The "don't know/refused" demographics look fairly favorable to Democrats; lots of non-whites and people in Democratic-leaning metro areas, not a lot of non-college educated whites. Looking more at the first three states than WI when I say that.

Also, Biden being down by just two points with non-college educated whites in WI is pretty impressive.

[philly09]

The "don't know/refused" demographics look fairly favorable to Democrats; lots of non-whites and people in Democratic-leaning metro areas, not a lot of non-college educated whites. Looking more at the first three states than WI when I say that.

Also, Biden being down by just two points with non-college educated whites in WI is pretty impressive.

A perfect fit for those states.

[CellarDoor]

The crosstabs on FL voting preferences are interesting.  56% of D's and 63% of R's say they are planning to vote in person on election day.  Crosstabs are not necessarily reliable (and these numbers are impossible given who has already voted in FL) but I wonder if D's will have better turnout than expected on Tuesday.

EDIT:  I misread the results.  the 56/63 election day split is for people who haven't voted yet.  Thanks to prag_prog for pointing this out.  I think the relatively small lead for R's here is good news for Biden.  I think most folks had suspected a heavier R skew on election day.

[philly09]

The "don't know/refused" demographics look fairly favorable to Democrats; lots of non-whites and people in Democratic-leaning metro areas, not a lot of non-college educated whites. Looking more at the first three states than WI when I say that.

Also, Biden being down by just two points with non-college educated whites in WI is pretty impressive.

https://twitter.com/RonBrownstein/status/1322758466641907712

[Pericles]

Expected.  WI appears gone, but everything else is still in play.

You aren't getting a 6 point error in Arizona for your side.

[Penn_Quaker_Girl]

To expound on the AZ-poll and the "Rest of Arizona besides Maricopa and Pima Counties", here's how the numbers looked in 2016 after some quick calculations (uh oh!):  

Rest of Arizona (Other than Pima and Maricopa) -- 2016

Trump  56%  337,612
Clinton  38%  233,869
Other  6.0%  36,351

And comparing that to this poll:

Rest of Arizona (Other than Pima and Maricopa) -- 2020 NYT/Siena

Trump 51% (-5)
Biden 42% (+4)
Other 2.0% (-4)

[Blair]

will we get 18 pages for this?

[Skye]

These Wisconsin numbers have been horrible for Trump. The state is borderline Likely D.

[Statilius the Epicurean]

[Penn_Quaker_Girl]

These Wisconsin numbers have been horrible for Trump. The state is borderline Likely D.

Which, combined with MI and Biden presumably holding all Clinton-2016 states, means that Trump MUST run a straight through AZ, FL, GA, NC, and PA. 

Not impossible by any means (nor is him winning in WI or MI), but Biden being loosely set-up with 258/259 (depending on NE-02) EC votes is a pretty spectacular position in which to be. 

[Podgy the Bear]

All of this points to a 320 EV and an 8-9 PV win for Biden (about a 12 million vote plurality).  And this doesn't include TX--which continues to intrigue and tantalize.

[prag_prog]

The crosstabs on FL voting preferences are interesting.  56% of D's and 63% of R's say they are planning to vote in person on election day.  Crosstabs are not necessarily reliable (and these numbers are impossible given who has already voted in FL) but I wonder if D's will have better turnout than expected on Tuesday.

no...you got that wrong. That question is for people who haven't voted yet.

[Skye]

These Wisconsin numbers have been horrible for Trump. The state is borderline Likely D.

Which, combined with MI and Biden presumably holding all Clinton-2016 states, means that Trump MUST run a straight through AZ, FL, GA, NC, and PA.  

Not impossible by any means (nor is him winning in WI or MI), but Biden being loosely set-up with 258/259 (depending on NE-02) EC votes is a pretty spectacular position in which to be.  

Yeah, it's pretty much the only scenario where Trump could possibly win that I'm paying attention to. The Biden camp (and doomers around here) should probably be concerned about the closeness of FL and NC but at least his margins in AZ and PA are a bit more favorable.

[prag_prog]

https://twitter.com/ThePlumLineGS/status/1322852329910116359

Also, from the summary of the polls -

"Many of the those who said they did not vote in 2016 said they had already voted this year. In Florida and Arizona, more than two thirds of nonvoters in 2016 who were identified as likely voters this year said that they had already cast a ballot. That figure was 56 percent in Wisconsin and 36 percent in Pennsylvania."

[Statilius the Epicurean]

^ This makes me wonder if some pollsters' likely voter screens and turnout models are missing a Democratic surge, especially in states like Texas which have been uncompetitive before. I think Nate Silver mentioned this as a possibility on the 538 podcast.

[CellarDoor]

The crosstabs on FL voting preferences are interesting.  56% of D's and 63% of R's say they are planning to vote in person on election day.  Crosstabs are not necessarily reliable (and these numbers are impossible given who has already voted in FL) but I wonder if D's will have better turnout than expected on Tuesday.

no...you got that wrong. That question is for people who haven't voted yet.

Well that makes complete sense.  Thank you.  I've updated my original post.  I think that split is good news for Biden.  It seems like a more favorable ED split for D's than most had projected.

[Stuart98]

Yeah this is what I expected. Selzer looks like an obvious outlier. Doomers chillax.

[philly09]

https://twitter.com/Nate_Cohn/status/1322867188630519810

https://twitter.com/Nate_Cohn/status/1322867839745859584

[prag_prog]

https://twitter.com/Nate_Cohn/status/1322867188630519810
https://twitter.com/Nate_Cohn/status/1322867839745859584

I think Nate meant AZ/WI/PA in the first tweet

[Splash]

Feeling good about Arizona and Wisconsin. Pennsylvania looks solid as well but I am glad that Biden is closing the chapter on his campaign there tomorrow. Florida is still a coin toss. 

[Catalyst138]

If Wisconsin and Michigan are basically safe then Biden only needs to win one of PA or AZ to win. This election is as close to over as it can be without being over.

[Crumpets]

If Wisconsin and Michigan are basically safe then Biden only needs to win one of PA or AZ to win. This election is as close to over as it can be without being over.

Or AZ + NE-2, which is my "break glass in case of emergency" scenario I'm keeping in mind if early results look meh.