Today, Nate Silver confessed that he screwed up.
Well, sort of.
The title is perfect. Perhaps Silver should have stopped there. He goes on to say …
I also wonder if there’s been too much #datajournalist self-flagellation. Trump is one of the most astonishing stories in American political history. If you really expected the Republican front-runner to be bragging about the size of his anatomy in a debate, or to be spending his first week as the presumptive nominee feuding with the Republican speaker of the House and embroiled in a controversy over a tweet about a taco salad, then more power to you.
The FiveThirtyEight “polls-only” model has correctly predicted the winner in 52 of 57 (91 percent) primaries and caucuses so far in 2016, and our related “polls-plus” model has gone 51-for-57 (89 percent). Furthermore, the forecasts have been well-calibrated, meaning that upsets have occurred about as often as they’re supposed to but not more often.
Please stop the self-congratulations. Nate, please tell us how the hell you put Indiana in the “win” column.
Silver was wrong on Indiana every step up the way with poll after poll after poll after poll projecting Cruz would win.
On April 30, “Silver’s Secret Sauce” projected Cruz had a 65% chance of winning Indiana. On May 1, I reported Silver Flushes Secret Sauce Down Toilet, Now Projects Trump has 69% Chance.
Silver puts that monstrous last-minute flip-flop in the win column. I call that pathetic.
There’s a lot more to the story, so I’m going to proceed in five sections:
- Our early forecasts of Trump’s nomination chances weren’t based on a statistical model, which may have been most of the problem.
- Trump’s nomination is just one event, and that makes it hard to judge the accuracy of a probabilistic forecast.
- The historical evidence clearly suggested that Trump was an underdog, but the sample size probably wasn’t large enough to assign him quite so low a probability of winning.
- Trump’s nomination is potentially a point in favor of “polls-only” as opposed to “fundamentals” models.
- There’s a danger in hindsight bias, and in overcorrecting after an unexpected event such as Trump’s nomination.
Silver’s admission on point 1 is quote revealing. In effect he admitted blatant bias, overruling his own model.
In points 2, 3, and 5 Silver attempted to sweep his bias under the rug.
Point 4 is also an interesting admission. That is something I hammered Silver on for months on end.
Silver’s “Polls Plus” forecasts have been little more than biased bullsh*t.
The entire rest of his lengthy post was noting more than an attempt to sweep the mess under the rug.
Here’s the question I ask: Did Silver learn a damn thing?
Here’s a few things I know.
- Changing social attitudes matter.
- It was clear in December that voters did not care what silly things Trump said.
- Silver favored historical precedents over massive attitude changes. That was a big mistake, and one I warned Silver on in December.
Reflections on Being Wrong
I am not purposely attempting to brag. I may have been wrong.
I have been wrong on many things, most notably the prolonged effect QE had on the market.
I was totally wrong.
I now think Trump will win the election. I may be wrong about that. And I may change my mind thirty days from now.
But I won’t change my mind one day before the election with a proclamation that I got it right.
Mike “Mish” Shedlock