FLORIDA: As of yesterday, 944k voters had voted, 18x higher than the 52k who’d voted at this point in 2016. Vote share by party, vs. this point in 2016:
Dem 52.6% (vs. 36.8%)
GOP 28.4% (vs. 35%)
Unaffiliated 17.9% (vs. 22.6%)
2/
NORTH CAROLINA: As of yesterday, 418k voters had voted. At this point in 2016, 32k voters had voted. Vote share by party, vs this point in 2016:
Dem 51.9% (vs. 36.3%)
GOP 17.3% (vs. 37%)
Unaffiliated 30.5% (vs. 26.4%)
3/
PENNSYLVANIA: As of yesterday, 90k voters had cast ballots. At this point in 2016, 38k had voted. Vote share by party, vs this point in 2016:
Dem 76.5% (vs. 45.2%)
GOP 15.3% (vs. 44.1%)
Unaffiliated 7.6% (vs. 6.4%)
4/
Most states don’t register voters by party. In those states, African Americans, Hispanics, college whites make up disproportionate share of voters so far. Some examples…
5/
GEORGIA share of electorate that has voted so far, by race, compared to 2016:
African American 36.3% (vs. 19.7%)
White college 23.2% (vs. 23.8%)
White non college 35.3% (vs. 52.4%)
6/
VIRGINIA share of electorate that has voted so far, by race, compared to 2016:
African American 13.3% (vs. 8.5%)
White college 37.1% (vs. 34.9%)
White non college 42.7% (vs. 49.5%)
7/
Repubs will say, rightly, that votes Clinton’s camp banked were regular voters who would have voted anyway. This year, Dems are paying attention to new and infrequent voters who are showing up at the polls. So far, ~21% of ballots cast come from new or infrequent voters.
8/
Among new voters: 20% are registered Dems, 8.8% are registered Repubs, 69% are unaffiliated.

In 2016, those early voters (to date) were: 17.9% Dem, 15.4% Repub, 64.8% unaffiliated.
9/
Among infrequent voters: 23.8% are registered Dems, 9.4% are Repubs, 65.6% unaffiliated.

In 2016, those infrequent voters who voted early (to date) were 20.2% Dem, 16% Repub, 62.2% unaffiliated.
10/
Important: Just because Repubs and white non-college voters are declining shares of electorate does NOT mean they’re not turning out. EVERY demographic group is turning out at hugely higher rates than at this point in 2016.
11/
Examples, using raw numbers, 2020 votes cast to date vs. to this point in 2016:
18-29: 383k in ’20, 62k in ’16.
65+: 3.1m in ’20, 365k in ’16
Blacks: 541k in ’20, 40k in ’16
Hispanics: 153k in ’20, 14k in ’16
12/
More examples, raw numbers, 2020 votes cast to date vs. to this point in 2016:
Non-col white: 2.9m in ’20, 438k in ’16
Col white: 2.06m in ’20, 227k in ’16
Suburb wom: 1.46m in ’20, 158k in ’16
13/
These Dem advantages will narrow, as early voting opens in more states and as more people go to the polls. But edge is significant. When polls open on Election Day, Pres. Trump is likely to be substantially behind Joe Biden.
14/
And, for more early voting data, keep tabs on @ElectProject’s work here: https://electproject.github.io/Early-Vote-2020G/index.html 16/15 and this time really done
You can follow @PoliticsReid.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: