First off, I was wrong. I predicted that Kamala Harris would win and she lost. This "hobby" is only fun if you get very specific and I was, very specifically wrong.
But I'm not beating myself up about it. The polls were very close and the election could have gone either way. In my post, I said there were four possible outcomes and one of them was that for a third straight election, the polls had underestimated Trump's ability to turn out his voters. That turned out to be the right answer.
But I'm not letting myself off that easy. Here are the gory details.
1. I was Two For Seven in the Swing States.
There were seven competitive states in this election. Nate Silver said the most likely outcome was for Harris to win all seven and the second most likely scenario was for Trump to win all seven.
I went five for Harris, two for Trump. I was wrong by an average of 3.0 points. My error in every swing state was in favor of Harris.
2. I was off by Even More in the "Stretch States."
Like many, I was fooled by the Iowa poll that showed Harris winning. I was off by an average of 6.7 points in these states. My error in five of these states was in favor of Harris. The only exception was Virginia, which I had one point closer than Harris' final margin.
3. The Big States Were Bad Too.
I had Harris winning NY by 20.4. She won by just 13.6.
I had Harris winning California by 27.9 she won by 20.1.
I had Harris winning Illinois by 17.4. She won by 10.9.
I had Trump winning Ohio by 6.7. He won by 11.2.
So I was off in favor of Harris for all 4 of these. The average error was 6.15%.
4. I missed the House, But It Was Close.
My prediction was Dems 223, Republicans 212. The actual result was GOP 220, Dem 212.
So the Dems won eight fewer seats than I predicted.
5. I Missed Two Senate Races.
I had the GOP holding the Senate but they won a bigger margin than I expected. I had the Dems winning Ohio and Pennsylvania. They lost both.
I predicted margins in 11 Senate races. I was off to the left in all 11 of them. The average error was 4.3 points. So I did slightly better here than in the White House.
Bottom Line and Grade:
I sucked. I made 29 predictions and I predicted the right winner in 20 of them, which is not impressive. Worst of all I was off in the same direction in 28 of the 29 predictions.
Giving myself a D for the night. The only think I can take any pride in is that I correctly predicted that Florida would vote to the left of Texas. Trump won Texas by 13.7 and Florida by 13.1. Yipee!
What Can We Learn?
As always, bias is a bear. I really wanted Trump to lose and I put too much faith in the voters to make that happen. He really is a malignant, terrible person and nothing about the first seven months of his term have changed my belief in that.
In the week or so after the election, some Harris campaign pros admitted that they more or less knew she was going to lose. It seems that the Democratic internal polling was a lot more accurate than the public polling. I really wish the party would release all or most of that data. It would help inform the 2028 primaries.
I think the most significant lesson is that Trumpism is not just mainstream but the dominant political force in American politics. Trump's policies and complete contempt for not just norms, but for the rule of law has been rewarded. And there hasn't been a lot of pushback from Republicans. There is no reason to think that the 2028 nominee will do anything to distance himself from the incumbent. I personally hope that Trumpism without Trump fails. But that will also tell us something ugly about our electorate.
We chose the guy. Twice. What can be more disappointing than that?
No comments:
Post a Comment