Blog Credo

The whole aim of practical politics is to keep the populace alarmed (and hence clamorous to be led to safety) by menacing it with an endless series of hobgoblins, all of them imaginary.

H.L. Mencken

Sunday, November 3, 2024

That Selzer Poll

 Ann Selzer only polls Iowa. That's her whole thing. When she says Harris has staked a 3 point lead in Iowa, that's important. Here's why.

First, Selzer is insanely accurate at doing her "one thing."  Here are some other polls and actual results.

2022 Senate: R+12 (R+12) 2020 President: R+7 (R+8) 2020 Senate: R+4 (R+7) 2018 Governor: D+2 (R+3) 2016 President: R+7 (R+9) 2014 Senate: R+7 (R+8) 2012 President: D+5 (D+6)

So she missed in 2018, otherwise she's been spot on or least within the margin of error. In particular, her poll was the canary in the coal mine in 2016, when she showed Trump winning a state that Obama had carried comfortably. What she measured was the movement towards Trump in the Midwest more broadly.

So, there's still very much a chance that Trump wins Iowa, because that would be within the margin of error. However, Trump winning Iowa by a razor thin margin is likely to signal the sort of movement away from Trump that dooms him in the Blue Wall.

What's more, the polls shows a massive defection from Trump amongst older white women. They are simply sick of his shit.

Finally, it matters because of one big methodological quirk to Selzer's polling. She measures the data in front of her. You would think every pollster does this, but - as I wrote the other day - pollsters herd together. Even the Red Wave pollsters are starting to move slightly towards Harris in order to preserve their viability in future elections. 

Pollsters are afraid of a 2020 level miss, so they herd together to show a tight race. If either candidate could win than either result would "validate" their polls. 

Selzer is basically "fuck all that" and simply reports what she sees. That's why she was right in 2016. Everyone knew Trump would lose, but she saw what she saw and what she saw was accurate.

All this means less about Iowa in particular and more about the shape of the race in general. It also suggests that my theory of the election might be right after all, and Harris is headed to comfortable win on Tuesday.

No comments: