Two Cheers for Nate Silver & Co.: Polls, Predictions, & the Persistent Problem of Political Noise

A plea for people to subscribe to and read Nate Silver & Eli McKown-Dawson's "Silver Bulletin" as a vaccination against all the polling & pundit election bullshit this fall: Why Silver’s...

A plea for people to subscribe to and read Nate Silver & Eli McKown-Dawson's Silver Bulletin <https://www.natesilver.net/p/nate-silver-2024-president-election-polls-model> as a vaccination against all the polling & pundit election bullshit this fall: Why Silver’s election forecasting remains essential amidst the polling-noise election-misinformation tsunami…

Nate Silver’s election forecasts have always revealed an essential truth: the polls don’t change much day-to-day, & almost all breathless stories by reporters, pollsters, & pundits mislead readers into attributing substantive significance to what is overwhelmingly likely to be mere statistical noise. Before we can even start to try to shift away from horse-race political journalism to public-policy and real life-impact political journalism, we need to get many more people reading the Silver Bulletin…

---

I find Nate Silver these days to be annoyingly and gullibly tech-bro adjacent, hence misleading once he gets out of his wheelhouse.

But Nate Silver in his wheelhouse!

Pay attention, because this is important: I find him—I still find, as I have long found him—an indispensable guide to keeping a sense of proportion in this election fall. It is true that if you follow Nate Silver, you find yourself obsessing about election polls. But you will only obsess about election polls for about four minutes a day. That is what it will take you, reading Silver and McKown-Dawson’s daily “Silver Bulletin Presidential Election Forecast” <https://www.natesilver.net/p/nate-silver-2024-president-election-polls-model> to learn that there is, essentially, no news but only olds in the polls: .

Tracking back to when the Silver Bulletin started its model of Harris-Trump at the end of July, so far it has told us:

• 2024-09-28 to 2024-10-08: Toss-up

• 2024-09-27: Toss-up (but you would rather be playing KH's hand in this contest)

• 2024-09-17 to 2024-09-26: Toss-up

• 2024-09-15 to 2024-09-16: Toss-up (but you would rather be playing DT's hand in this contest)

• 2024-09-05 to 2024-09-14: 3-2 Trump

• 2024-09-04: Toss-up (but you would rather be playing DT's hand in this contest)

• 2024-07-31 to 2024-09-03: Toss-up

• 2024-07-29 to 2024-07-30: Toss-up (but you would rather be playing DT's hand in this contest)

On only seven of seventy-odd days do Silver and McKown-Dawson say that the most recent bunch of polls suggests a little movement in the race—and it is only a little movement. Telling people that almost all of the time the polls are essentially saying very close to the same thing they said yesterday is a valuable service. That “calm down” and “we are ignorant of where the polls are missing the mark” are the most important things for people to get from the polls this year. Silver and McKown-Dawson give them to us. And virtually nobody else does.

Indeed, there is only one thing Silver and Eli McKown-Dawson have said about his average of nation-level and state-level polls that I would take serious exception to.