23 October 2020 — 0839 mdt
Last night’s debate, Trump’s Montana numbers,
Bullock v. Daines, Cooney v. Gianforte
My reaction to the debate. Last night’s debate didn’t leave me feeling unclean for having watched it. Trump maintained a reasonable tone of voice much of the time, but could not resist slinging shovels of mud at Biden. Biden kept his cool, authoritatively counterpunching with clean facts. Who won? According to a CNN snap poll, Biden. But “Who won?” is the wrong question. The right question is “how many voters changed their minds?” I suspect Omar Khayyam provided the answer in the twelfth century:
Myself when young did eagerly frequent doctor and saint,
and heard great argument about it and about:
but evermore came out by the same door as in I went.
Expecting I would post this roundup before the debate, I yesterday wrote “I’ll be watching and live Tweeting the debate. Will Trump do better than he did in the first debate? Probably: it would be difficult to do worse. Will he do well enough to wipe out Biden’s mean 9.8-point national lead, or reduce it enough to give him a reasonable chance of winning the electoral vote? Not likely. As long as Biden keeps his cool — if his family is attacked, he should show a flash of controlled anger — he will maintain a substantial lead. Will Trump keep his cool? Probably not even with five milligrams of lorazapam.” Trump sort of kept his cool some of the time, which was out of character for our sewer brawler in chief. Did he discipline himself on his own, or did he employ a pharmaceutical assist?
Trump still leads in Montana. FiveThirtyEight’s weighted average has him up by 8.2 points, 52.1 to 43.8 percent. The most important number is Trump’s absolute share of the vote, not the Trump-Biden difference. Civiqs reports Trump’s approval number in Montana has declined, but it’s still close to 50 percent:
Caveat. Civiqs Trump tracking operation, FiveThirtyEight notes, is more akin to modeling than to polling:
Polls like Civiqs or Microsoft News tracking polls that are produced using MRP (short for “multilevel regression with poststratification”), a modeling technique that produces estimates based on survey interviews that are then used to calculate probabilities; these probabilities are then projected onto the composition of the population in question. While this is a valid technique for understanding public opinion data, we exclude them because we consider them more like models than individual polls.
Trump did not self-immolate last night. Therefore, I don’t think he lost ground in Montana. His staunchist partisans stick with him in this white working class state because as much as they may dislike his crudity and cruelty, they dislike urban Democrats, whom they consider snooty intellectuals with unAmerican skin colors, even more — and with reason: many members of today’s national Democratic Party consider the white working class unrepentant, irredeemable, racists who should be excluded from the party. That hurts Democrats in Montana.
U.S. Senate, Bullock versus Daines. This is the second most important statewide election in Montana this year. FiveThirtyEight’s weighted average has Daines up by three points, 51.5 to 48.5, but the raw average of the last three polls gives Daines a one point lead. Daines is hugging Trump so tightly it’s a wonder the president’s ribs aren’t broken. FiveThirtyEight gives Bullock a 33 percent chance of winning.
If Trump self-destructs to the extent his popularity in Montana falls below 50 percent, Daines may fall with him. Bullock’s biggest problem now is the increasing number of new Covid-19 cases in the state. The number is rising in large part because Trump worshiping citizens are flouting public health measures such as masking-up, although politically Bullock cannot say so. He can, however, act decisively to cool off Covid-19 hot spots, as he did in Flathead County yesterday. That may win him some votes.
Although Bullock may be a point or so behind, he still has a legitimate chance of winning.
Montana governor, Cooney versus Gianforte. Cooney’s in huge trouble, trailing Gianforte by an average of approximately six points, give or take a point or two, with 10 days left in the campaign and many votes already cast. Gianforte has poured at least $7.5 million of his own money into his campaign, a powerful example of the danger of a system that allows the superrich to treat elections as auctions.
FiveThirtyEight maintains a list of polls on this election, but does not calculate an average or conduct a probability analysis. Averaging polls simulates the effect of a larger sample size, which reduces the sampling margin of error. I calculated a average for nine polls from late June through mid-October, weighting for sample size and age (giving newer polls greater weight). Take the error bars with a kilo of salt. The result is the last set of markers on the plot below.
According to the high quality MSU Bozeman poll, women prefer Gianforte by two points, men by seven. This is not just odd for a Democrat: it’s usually a kiss of death at the ballot box.
Cooney is ≈ 10 points ahead among voters under 40, and ≈ 10 points behind with voters over 40. There are many more voters over 40 than under.
Democrats should not give up on Cooney — long shots sometimes hit the bullseye — but they should begin steeling themselves for Governor Gianforte and the absence of a Democratic veto to shield Montanans from the excesses of a legislature controlled by radical right Republicans hellbent on rendering government impotent to help those most in need.