This title is highly misleading. It says that since the difference between Harris and Trump being only 3.2 million means that only 3.2 million Democrat voters stayed home, which is incorrect on multiple levels. That's just not the case.
Joe Biden won 81.2 million votes in 2020.
Kamala Harris won 72.4 million votes in 2024.
That is just shy of a 9 million vote drop.
Where did they go? Even if you assume that a portion of former Biden voters voted for Trump, trump only gained 1.2 million votes. Which means there's about 7.8 million votes unaccounted for at a minimum. Where are they? They stayed home.
Saying that Harris lost 9 million votes compared to Biden and saying that only 3.2 million of them stayed home is not only factually and mathematically incorrect, it gives off the impression that the rest of the voters switched to Trump (which I'm sure Trump will claim if and when given the opportunity.)
That actually makes it worse, since covid hit Repubicans in general harder because they skew older and less likely to follow protocols. When you actually consider that, Trump actually picked up even more voters.
Regarding voter registrations, most areas use provisional ballots or some equivalent when someone's eligibility to vote is in question, and those ballots are typically only touched if counting them would actually impact the election.
Had there been enough people challenging their voting eligibility in even a single state, Harris and her campaign team would be all over that like flies on shit. The fact that this wasn't brought up a single time is a strong indication that even if there was some voter suppression going on, it wasn't enough to impact the outcome.
One extremely important thing to note about 2024 turnout—it was only down in uncompetitive states. Turnout was, on average, higher in swing states compared to 2020. Democrats did not 'stay home' in the states where it mattered and you should ignore anyone trying to explain Harris's loss that way.
That said, it feels kind of squicky that it's WaPo and the University of Florida presenting this. Given their exciting editorial turnabout on the endorsement and Flordia being Florida - eh.
It looks like this 3.2M is the difference of the popular vote Trump received over Harris. A rough search looks like about 56.5% voter turnout amongst voting age population.