This person, mostly just a Twitter rando into this, articulates what I'm thinking pretty well...I'm just having a really hard time not reading the big improvement in early voting by Republicans as a positive. I know it doesn't have to be, but given it's the one solid data point we know, it's really hard to dismiss that given the closeness of these elections.
Polls: Polls are obviously mixed, though better for Trump than either of the prior presidential elections. They’re almost certainly somewhat “fixed” (or better) from the prior Trump under-sampling given that it was Trump’s numbers often showing up in the lower 40’s then that drove much of the disparity. They’re very tight now with Trump’s historical numbers. Could they have gone the other way and heavily overrepresented Trump on average? Sure, but the next topic makes me think that’s not happening to any material extent.
Early Vote: This is largely what’s driving my Trump EC pick. Registered republicans are pushing harder in early voting largely because of 2020/2022 and the Trump campaign + surrogates fully encouraging it now. So, the fact that R’s are a higher percentage of the early/mail vote in the swing states than they were in 2020 (or 2022) isn’t surprising. And mathematics dictates that the other side of that fraction (the Dems’ share of EV compared to the prior two elections) will be smaller if they didn’t make an *additional* push for EV compared to prior cycles.
It also means that the “cannibalization” conversation that folks are having is true to an extent. The R share of ED vote won’t be as high in many of the swing states where they’ve shifted more dramatically in the EV. The people that think PA will be some kind of mega Trump blowout because he was down 1.1M in 2020 (looking at [D – R] early votes) and is now only in the low 400Ks aren’t doing folks any favors. But the Dems have done well previously with the realization that a banked vote is a vote, while even a 99% chance of an ED vote isn’t yet a vote.
But here’s the reason I’m reading a bit more into the early vote tea leaves: the registered Republican share of no- and low-propensity voters is higher than the overall D/R early-vote split in every single swing state that has data for it. Take PA, for example. As of this morning’s data, The D/R split is D+22.9%. But the latest data for voters who haven’t voted in the last three general elections shows a split of just D+13.3%. So, R’s are moving a higher relative share of those voters to the early vote. The same holds true in AZ, NC, and NV.
So, we have a situation where R’s have come out more heavily in the EV in the swing states, and lower-propensity R voters have come out even more as a percentage than that. That feels like it could be *some* proxy for voter energy. Given how razor thin the margins were in the blue wall last time, this is what leads me to believe that PA & WI get picked off this time, while the Sun Belt also goes R for all four.
Other: Dobbs, J6, GOTV…all of these are certainly part of the mix and seem to tend toward Harris. And if I’m wrong by late tomorrow night or Wednesday, I’m guessing that Dobbs & day-of/day-before GOTV efforts were the reason. I just don’t know why those wouldn’t have manifested more fully in the polls or early vote throughout though. Many are talking about registered “Others” and even R’s breaking much more for Harris than did for D’s in prior cycles. Maybe so, but I would have expected to see that push in the zero-voter Dems in the swing states. Why are the Others and R’s motivated to come out and vote D in early voting, but not D’s to the same extent? I would think we’d see it in the D side first, dragging up the Others in that case (and even the crossover R’s). Anyway, what do I know.I don’t think I’d be surprised by any result in the seven swing states (other than any swing state being a 5%+ romp). But given all the data, I think it’s the map below.