Thread by @Nate_Cohn

Let's take a look at the turnout data so far in North Carolina, where counties worth one-third of the electorate have now updated their vote history data

These counties lean a little bit left, with a disproportionate chunk of the white liberal vote (Wake, Buncombe, Durham all in). So I think these numbers could be rosier than the final tallies for Democrats statewide, but I think the patterns will largely hold up

Let's start with party registration.
In these counties, 77% of Democrats turned out v. 82% of GOP. That's a 7 pt increase over 2016 in both cases.
As a result, the electorate by party reg in these counties is D+5.8, v. D+7.4 among all registered voters.

As I mentioned, I think it's possible that it doesn't wind up quite that good for Democrats in the end. But if we do suppose for a second that these turnout rates hold statewide, then you get a D+3.5 electorate v. D+5.3 registration.

By race, we have a story fairly similar to Georgia.
Over all, 79.4% of white registered voters turned out v. 69.5% of Black voters.
That's a 7 pt. increase in white turnout rate, and a 4 pt increase in black turnout

On all of these turnout rate statistics, the denominator is the registered voter population. And as in Georgia, the black share of registered voters has declined since 2016, so the decline in the black share of the electorate is greater than these turnout rates suggest

In these counties, Black voters made up 17.7 percent of the electorate v. 19.4 percent of registered voters.
In 2016, Black voters were 19.9% of the electorate and 21.4% of registered voters

Another similarity to Georgia is that there are a growing number of 'unknown' race voters, who now represent 10% of RVs in these counties. We don't know about their actual racial composition, but if many are Black then the decline in the Black share could be a bit overstated

One thing we can do in NC that we can't do in GA is look at age. We can't look at age in GA because I don't have a totally final voter file there, and I'm worried that would bias our findings against late registrants, who are disproportionately young.

The NC data in these counties shows no increase in the youth share of the electorate.
18 to 25 year old voters represent 9.4 percent of the electorate, the exact same as in 2016.

The turnout rate *among* 18 to 25 year old registrants, however, went up from 53.9 to 60.3 percent.
But the youth share of registrants went down, so there's no increase in the youth share of the electorate, whether as a consequence of coronavirus or an aging population

The only age group with an increase in its share of the electorate? Voters over age 65. They represented 23.7 percent of the electorate in these counties, v. 20.8 percent of the electorate in 2016.
I'd guess the aging population is a big factor here

So stepping back: in these counties, we have an older electorate, the black share of the electorate dipped, and no increase in Democratic turnout v. Republican turnout.
But Biden gained a net-3.5 points over Clinton in these counties, more than the state/nationwide

The above average swing toward Biden does raise the possibility that the turnout data will be even worse for him elsewhere, though it may just reflect more 'persuasion.' It'll be interesting to revisit this when more of the data is in.

I should add that I think it's genuinely surprising that the turnout increased so much without increasing the young/black/dem share of the electorate
Nonvoters in 16 were disproportionately young/black/dem. But higher turnout just drew another disproportionately old/white/rep set