Dem
50
GOP
50
We're going to take a look at the early voting numbers each Tuesday between now and the election, and see what's interesting.
This week, there are a couple of things worth noting. The first is that, at this time in 2020, 15,938,255 early ballots had been cast. This year, so far, 4,729,468 early ballots have been cast. Obviously, that is considerably fewer. A big part of the equation is that the pandemic has passed, and considerably fewer people regard early/absentee voting as the only safe option. There are also changes in the rules (e.g., red states getting less permissive) and also in reporting procedures. The general presumption is that early voting favors the Democrats, letting them pile up a tough-to-topple lead heading into Election Day. It looks like the blue team won't have as big a lead this year as they did in 2020. That said, there's still time, and the early-voting numbers could take a big jump.
The other thing worth noting is something we got wrong last week. As we noted, only five states break down their early ballots by gender: Colorado, Georgia, Idaho, Michigan and North Carolina. At the moment, the early votes from those states are 54.9% from women and 44.1% from men (the other 1.1% declined to state, or put a gender identity other than the two from the Garden of Eden).
No states broke down early voting by gender in 2020, so we can't compare directly. However, here is the breakdown of women's general election vote, rounded to the nearest whole number, in every election since 2000:
| Year | Voted D | Voted R | Dem. Adv. |
| 2000 | 52% | 48% | +4% |
| 2004 | 54% | 46% | +8% |
| 2008 | 53% | 47% | +6% |
| 2012 | 53% | 47% | +6% |
| 2016 | 52% | 48% | +4% |
| 2020 | 52% | 48% | +4% |
Obviously, with a lead greater than 10 points, Kamala Harris would outperform every Democrat of the last 20 years, if that were to hold.
Because of the abortion issue, Harris probably is doing better with women than most or all recent candidates. That said, the numbers are also skewed by the fact that the early votes are disproportionately Democratic (53.9% to 29.0% so far), and women are more likely to be Democrats than Republicans. However, we can try to correct for that, at least a little.
Taken as a whole, American women are 51% Democratic/lean Democratic, 44% Republican/lean Republican and 5% independent. Georgia does not give a partisan breakdown, but in the four states that do that, and also give a gender breakdown, one would expect there to be about 338,477 votes from women. There have actually been 408,921 such votes. Roughly speaking, that means the women's vote is something like 20% higher than expected. Since normal is something like Democrats +5, it's reasonable to guess that Harris is on pace to win the women's vote by 6 or maybe 7 points, which would be a pretty big difference from 2016 and 2020. So, this bodes well for her, though keep in mind that the projection is a bit crude, and is based on a small and maybe unrepresentative sample. (Z)