Various election things

(1) The Selzer Iowa poll is a big deal. Selter’s 2016 and 2020 Iowa polls correctly called one-sided wins for Trump when the national polls were saying it was much closer. Her data show older white women breaking hard for Harris, and if that’s accurate there’s every reason to think there are systematic polling errors underestimating this shift across the Blue Wall states at least.

(2) FWIW I think there are some real signs that the last week of the campaign has been a disaster for Trump with late deciders, starting with Nazipalooza at MSG on Sunday. Tim Alberta has an interesting super-gossipy piece in the Atlantic about all the infighting in the Trump campaign between staffers trying to keep him on message, meaning inflation and immigration, and those wanting to let his freak flag fly, with the latter clearly winning out down the stretch run of the campaign. The juiciest tidbit is that apparently the same little Nazi in training who convinced Vance and then Trump to go with the Haitians are eating your pets moral panic is also the genius behind putting the Puerto Rico is garbage comic in the leadoff spot at the MSG rally.

(3) Speaking of freak flags, have you ever seen a commie drink a glass of water Mandrake?

(4) The most encouraging data I’ve seen to this point are that, of the 75 million votes cast already as of last night, women were outvoting men by a margin of 11 to 9. Meanwhile IPSOS has women picking Harris by 14 points, and the final Siena poll has Harris ahead by 17 among women voters. The latter number explains why in the same poll Harris leads by eight among the 40% of respondents who had already voted.

Given these numbers, national and state polls are as close as they are only because they’re modeling a massive gender gap on Election Day itself, with so many more men than women voting on Tuesday that at least half of the 55% to 45% women over men gap in the early voting is eliminated in the final vote totals. Thus the other big unknown in this regard is how much of the final vote tally will already be banked prior to Tuesday. I suspect that it’s going to be very close to a 50/50 split between early votes and Election Day votes, with about 85 million in each group (it was 47/53 in the midterms, but early vote percentage totals in midterm elections are consistently a little lower relative to the percentage of Election Day votes in presidential elections; for obvious reasons you can’t really use 2020 as a baseline for predicting this year). And while higher overall turnout is better for Dems all other things being equal, I think we want as high a percentage of the vote to be early voters this year as possible, given the big gender edge for Democrats in the early vote.

All in all I think things are looking quite a bit better for Harris and the Democrats now than they did a week ago, mainly because Trump just couldn’t help screaming so much racism and misogyny into the ears of the Ariana Grande vote at the very last minute.

40 hours.