Battleground State Early Voting Analysis as We Approach Two Weeks Out

October 20, 2024

It’s Saturday night. Instead of watching my Bruins lose an embarrassment to something called the Utah Hockey Club, I dove into L2’s early voting data for several key battleground states. I did not include Nevada because only 11,000 people have voted early so far and any data we have from that can only be misleading at this point. Likewise, I did not include Wisconsin - although I did consider it. The problem with Wisconsin is that there is registration by party there. It also appears they do not have basic demographic data as well since the share of “unknown” across all categories was too high for my comfort level. If you are interested you can look at the data yourself here. By the time you get to it, the numbers below may have changed since L2 updates daily (perhaps even more frequently than that).

Michigan: 928,500 voters received (total electorate: 7,923,907 registered voters)

Party characteristics

• Dem: 58%

• Rep: 37%

• Other: 5%

Key demographic characteristics

• Women: 56%

• Black voters: 9.5% (overall voter share = 11.0%)

• Under 40 voters: 10.6% (overall voter share = 35%)

Brief analysis: There has been a strong showing of early voting in Michigan. Nearly one in eight voters have already voted. About 5.5 million people voted in 2020, which means that perhaps as many as 20% of those who will actually vote this year have already voted. So far, the numbers look good for Harris. The turnout is overwhelmingly Democratic with a large share of women having voted as well. Black voters are almost at par with their overall share of the electorate. Young voters (under 40) are lagging, but there is an expectation that many young voters are going to turn out on Election Day this year.

Pennsylvania: 767,800 votes received (total electorate: 8,432,982 registered voters) 

Party characteristics

• Dem: 64%

• Rep: 27%

• Other: 9%

Key demographic characteristics

• Women: 56%

• Black voters: 8.0% (overall voter share = 8.5%)

• Under 40 voters: 14.6% (overall voter share = 32%)

Analysis: Another strong showing for early voting in Pennsylvania, and it looks good for Harris. While the overall turnout is lower than Michigan, there is an even stronger Democratic turnout so far . Women are voting in very high numbers. Black voters are voting very close to their share of the electorate, as in Michigan. There is a stronger turnout among young voters. Michigan and Pennsylvania are nearly a must-win states for Harris to win and so far she has reason to feel optimistic.

North Carolina: 418,000 voters received (total electorate: 7,067,166 registered voters)

Party characteristics

• Dem: 36%

• Rep: 33%

• Other: 30%

Key demographic characteristics

• Women: 54%

• Black voters: 20% (overall voter share = 21.0%)

• Under 40 voters: 14.1% (overall voter share = 36%)

Analysis: North Carolina lags behind Michigan and Pennsylvania, but it also started early voting later. The news here is better for Harris than it may appear. Democratic turnout is only 36% but that is three points higher than Republican turnout. Independents (and others), who make up a whopping 38% of registered voters in North Carolina, might be expected to break about evenly. In the last several election cycles, the top office on the ballot has gone for the winner by roughly 1.5%. This usually swings from Democrat (governor) to Republican (president), but not always (Obama won in 2008). If Democrats can outpace Republicans by three points, there is a good chance that Harris can win. Women are turning out in high numbers and once again the Black vote is close to its overall share of the electorate.

Georgia: 878,400 votes received (total electorate: 7,923,907 registered voters)

Party characteristics

• Dem: 45%

• Rep: 45%

• Other: 10%

Key demographic characteristics

• Women: 55%

• Black voters: 30% (overall voter share = 34%)

• Under 40 voters: 14.3% (overall voter share = 38%)

Analysis: After a monster opening day of in-person early voting, Georgia still has a very strong early vote turnout. Over 10% of registered voters have already voted. You might be concerned that Democrats are equal with Republicans at 45%. Independents or others represent 29% of registered voters, so the 10% turnout is an underperformance so far. It’s not as bad as it looks. Remember that Biden won the state in 2020 by less than 12,000 votes. This is a 50-50 state. Looking at the key demographics here, it looks promising for Harris. Women are voting at 55% and while Black voters are falling short of their overall share of the electorate, fully 30% have already voted. If I had to guess who was winning so far, I would guess Harris.

Arizona: 266,300 votes received (total electorate: 7,923,907 registered voters)

Party characteristics

• Dem: 36%

• Rep: 44%

• Other: 20%

Key demographic characteristics

• Women: 50%

• Black voters: 0.9% (overall voter share = 1.4%)

• Hispanic voters: 11% (overall voter share = 22%)

• Under 40 voters: 8.2% (overall voter share = 34%)

Analysis: Okay, here is a state that is not looking good for Harris so far. Republicans are far outpacing Democrats, although there is a strong independent turnout (but still not matching their overall 29% of registered voters). Both major parties are voting about 5 points higher than their relative shares of registered voters. There is no advantage among women here, and there is not a significant number of Black voters in the state. There are, however, a lot of Hispanic voters - 22%, to be exact. Their share of the early vote is half that so far. Unless a lot of Democratic-leaning independents are turning out, Harris is probably losing Arizona right now.

There are still a lot of votes to be cast. We have good reason to believe that we will see more Democrats voting on Election Day, and perhaps more people overall voting on November 5th. A lot of Democrats seem to like voting on Election Day; they take some civic pride from it. I think Republicans used to like the convenience of absentee voting, but then came 2020 and everyone lost their minds. There has been some effort this year to convince Republican voters to vote early and that will probably pay off. So, we may see a lower share of Democratic early vote and a larger share of them on Election Day compared to 2020. If that is true, then Harris is in even better shape than it would appear in all of these states except Arizona.

We may very well see results this year that look odd to us. That happens a lot in politics. And in such a closely-divided environment as the electorate appears to be this year, it would not be surprising to see some odd results. For instance, if Harris does win Georgia and North Carolina, she can lose Wisconsin, Arizona, and Nevada and still win with 291 electoral votes. In looking at these early vote numbers, that alignment would not surprise me at all - even though I think she will win Nevada in any case.

I’ll update this in a week, but if you follow me on Notes you may see some quick daily updates on one or two states each day.