What the 2024 presidential election looked like in terms of all eligible voters

Somebody wrote a SubStack that I can’t find at the moment pointing out that the 2024 election results look significantly different if you’re looking at all eligible voters, as opposed to merely looking at everyone who actually voted. It’s a good point, and I broke down the numbers myself, using the Florida Election Lab’s estimates of the total voter eligible population (about 5% of American adults are ineligible to vote, mostly because they aren’t US citizens, although there’s a non-trivial percentage of disenfranchised felons as well).

Here are the results:

If the entire VOTER ELIGIBLE POPULATION is reduced to 1,000 people, the 2020 presidential results looked like this:

338 of those people voted for Biden

309 of those people voted for Trump

12 of those people voted for minor presidential candidates

Five of those people voted but didn’t vote for president

336 of those people didn’t vote at all.

Now looking at the same breakdown in 2024:

316 voted for Trump — a gain of just seven people per 1000 eligible voters. In other words, once we account for increasing population, Trump got almost exactly the same percentage of Americans to vote for him in 2024 as in 2020. His percentage of the actual vote increased by three percentage points (that would, by contrast, be 30 people in our hypothetical sample) almost solely because of DECREASED OVERALL TURNOUT of the voting eligible population.

307 voted for Harris. By contrast, Harris lost 31 voters of the VEP relative to Biden. That was the whole ballgame right there. Trump didn’t gain voters: Harris lost them.

The proportions of third party votes and people who voted but not for president were the same in both elections (12 and five respectively)

Meanwhile 360 people in our hypothetical room didn’t vote at all — a 24-person decline, i.e., very similar to the decline in Harris’s proportion of the VEP relative to Biden’s.

So there was a significant increase in people who didn’t vote at all, and that group is just about the same size as the size of the group of people who voted for Biden but didn’t vote for Harris. Obviously they weren’t exactly the same people, but I bet the overlap between those two groups is extremely high. Meanwhile, Trump gained basically no support, on a population-adjusted basis.

That’s the victory that’s been described as “decisive” a few tens of thousands of times in the last two and a half months. All of this, I acknowledge, was already more or less known, but I thought it would be useful to lay out in easy to digest numbers.