Nate Cohn

Nate Cohn @Nate_Cohn

Now, granted, the race was so close that it could have been decisive--even without a discernible contribution! But it does make it hard to write-up when that's in the range of possibilities — PolitiTweet.org

Nate Cohn @Nate_Cohn

Since maybe I didn't link it all together clearly: if the 'unknowns' are relatively white and backed Trump (as they did in Times/Siena polling), then changes in the racial composition of the electorate all canceled out and made no discernible contribution to Biden's gains — PolitiTweet.org

Nate Cohn @Nate_Cohn

There's some upside for Democrats in this graphic, btw. There's room for the Black share of the electorate to increase in the Senate runoff, and maybe make up for a bit of the deficit Dem Sen. candidates faced on Election Day https://t.co/3BdzaQtYW4 — PolitiTweet.org

Nate Cohn @Nate_Cohn

To reiterate what I said earlier: there's no way around concluding that the white share of the electorate was lower than 2016--maybe around 2 pts--while Hispanic/Asian/Other were up about the same. But nailing this down is a real mess, given the size of these groups v. 'unknown' — PolitiTweet.org

Nate Cohn @Nate_Cohn

This is a pretty small sample, but the stakes are pretty high: there are many more 'Unknown" voters than Hispanic, Asian, Other combined in this state, and so how you think about this group is pretty important — PolitiTweet.org

Nate Cohn @Nate_Cohn

In our Times/Siena polling, voters with 'unknown' race category were actually slightly likelier to identify as 'white' in the survey than the rest of the state's electorate. — PolitiTweet.org

Nate Cohn @Nate_Cohn

But the size of these trends--and therefore, the contribution of demographic shifts to Biden's win--is a bit dependent on how you handle the "unknown" race category — PolitiTweet.org

Nate Cohn @Nate_Cohn

No matter how you cut it, the data shows a secular decline in the white share of the electorate in Georgia and an increase in the Hispanic and Asian share of registered voters — PolitiTweet.org

Nate Cohn @Nate_Cohn

Analysis of other nonwhite groups in Georgia has another kind of arcane, confounding factor: the growing number of voters with 'unknown' race — PolitiTweet.org

Nate Cohn @Nate_Cohn

If we did go down the road of analyzing age, based on this data, I'm fairly confident it would be at least a little biased against youth turnout, and potentially meaningfully for the way the story is written. So we're just going to wait and see on that one — PolitiTweet.org

Nate Cohn @Nate_Cohn

The age problem is induced on our end: we do not have the final Georgia SOS voter file from before the election (though we've had a request in). This doesn't make a material difference on race, based on public state figures. It might on age, given that late registrants are young — PolitiTweet.org

Nate Cohn @Nate_Cohn

A few people have asked about some other demographic groups in Georgia, namely young and other nonwhite groups. For some arcane reasons, it's a little more complicated to think about those groups--but let's go into it — PolitiTweet.org

Nate Cohn @Nate_Cohn

So while analysis of the national electorate won't be complete for a while, I think we can pencil in that the final account will show a decline in the Black share of the electorate, as Black turnout ticked up but not nearly as much as non-Black turnout — PolitiTweet.org

Nate Cohn @Nate_Cohn

So while analysis of the national electorate won't be... final for a long time to come, I think we can pencil in that the final account will show that Black turnout ticked up, but not nearly as much as non-Black turnout, yielding a decline in the Black share of the electorate — PolitiTweet.org

Nate Cohn @Nate_Cohn

So while the final analysis of the election won't be... final for a long time to come, I think we can pencil in that the final account will show that Black turnout ticked up, but not nearly as much as non-Black turnout, yielding a decline in the Black share of the electorate — PolitiTweet.org

Nate Cohn @Nate_Cohn

We also have a lot of data at the county/precinct level elsewhere in the country suggesting the story will hold just about everywhere. TBH, Georgia was one of the very few places where it seemed like this story might not be true — PolitiTweet.org

Nate Cohn @Nate_Cohn

Georgia's the first of these states to fully update its vote history file. But we do already have some data from about 10% of N.C., and it has the exact same story: the Black share of the electorate drops notably as white turnout surges and Black turnout ticks up a few points — PolitiTweet.org

Nate Cohn @Nate_Cohn

The Georgia voter registration form has a question about race/ethnicity, and more than 90% of registered voters fill it out. So we know the racial composition of the Georgia registered voters--and the electorate--far better than most elsewhere in the country — PolitiTweet.org

Nate Cohn @Nate_Cohn

A few notes on this newly updated Georgia vote history data by race, which is among the best in the country https://t.co/3BdzaQtYW4 — PolitiTweet.org

Nate Cohn @Nate_Cohn

As a result, Biden's gains were concentrated in a ring around Atlanta https://t.co/nY8yg9NblE — PolitiTweet.org

Nate Cohn @Nate_Cohn

Biden compensated with huge gains in affluent and well-educated precincts https://t.co/ZC93lsNWLX — PolitiTweet.org

Nate Cohn @Nate_Cohn

Georgia is the first state where we have fully updated vote history data, and it shows the Black share of the electorate falling to its lowest level since 2006 https://t.co/kcquppig1I — PolitiTweet.org

Nate Cohn @Nate_Cohn

How Georgia turned blue: Huge gains in the Atlanta suburbs make up for a decline in the Black share of the electorate https://t.co/GdEN57oNeZ — PolitiTweet.org

Nate Cohn @Nate_Cohn

@jon_m_rob https://t.co/eOQVLebPUj — PolitiTweet.org

Nate Cohn @Nate_Cohn

Biden now leads by a full percentage point in Pennsylvania, as absentee and provisional ballots continue to add his tallies — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @mcimaps: I can confirm the Three Florida districts where Biden underperformed Clinton the most #FL27 - Went from 19% win to 3% win #FL… — PolitiTweet.org

Nate Cohn @Nate_Cohn

@MattGrossmann nope — PolitiTweet.org

Nate Cohn @Nate_Cohn

In all of these places, the GOP turnout advantage appears wider than it was in 2016--at least by comparing party registration of voters to all RVs. It'll be interesting to see how 'deep' the turnout edge runs when we get the final data, and how it plays out in less diverse places — PolitiTweet.org

Nate Cohn @Nate_Cohn

One thing that's quite clear in the places where we're already up to A-minus level turnout data (most of FL, NV, a couple of NC counties): the GOP enjoyed a pretty significant turnout advantage, even though most polls/analysts expected no side to enjoy a big advantage — PolitiTweet.org

Nate Cohn @Nate_Cohn

And it is not at all clear that the GOP will pursue every strategic part of the Trump message. I'd guess they will on cultural issues. The economic stuff? I think it'll be hard for someone else to be as good at bashing China, for ex. They won't be getting credit for stimulus — PolitiTweet.org