Comment by stinkbeetle
13 hours ago
Those 2015-16 ones sure aged poorly, I'm reminded of this https://i.imgur.com/6Z9QQj3.jpeg
This is why people don't really buy the "but he had Trump at 30%, you just don't understand statistics" apologist line. Sure he hedged in the dying days of the campaign (a cynic might think to try to protect his credibility), but the tone overall was of a person who comprehensively failed to understand the mood of the country from beginning to end.
Which is a problem because these election predictions are not just pure "mathematical models" and "data driven" like 538 would have had you believe. What mathematical model should be used? What data should and should not be used? At some point those things are based on the modeller's understanding of reality.
I think Nate did a phenomenal job calling out pollsters in that time. Since 538 was predominately a poll aggregator that did tricky stats to rank the reliability of each poll. I remember specifically an interview with him griping about some of the unusual data he was seeing from pollsters that made it look like, and I quote, 'Someone has their finger on the scales'
Perhaps critiquing statistical methods used by polling was something he was good at. I have no real opinion of his work there, which I didn't pay attention to.
But predicting an election requires a lot more than polling datasets and statistics textbooks. That's the problem that he made himself out to be an election prediction wizard, but really that was off the back of his good prediction in quite a bland and conventional election.
When things got slightly more spicy and reality diverged from his vaunted "models", his "data science" predictably fell in a heap. The worst thing is almost not even that he got it wrong, it's that he seemed incapable of recognizing that present reality was quite significantly different from the past data he had used to build his models. Even after being wrong in so many of these predictions. He just kept churning out these pieces about how Trump was probably finished this time.
Okay, this is clearly an LLM response, but for the sake of being polite:
> But predicting an election requires a lot more than polling datasets and statistics textbooks. That's the problem that he made himself out to be an election prediction wizard, but really that was off the back of his good prediction in quite a bland and conventional election.
> When things got slightly more spicy and reality diverged from his vaunted "models", his "data science" predictably fell in a heap
The models were correct in two elections - arguably three because a 30% chance means that an outcome will occur in thirty times out of hundred. That is not zero.
To the person who is running this LLM, please find better things to do with yourself.
6 replies →
He didn’t hedge at the end. Nate always writes the models before election season then doesn’t touch them apart from actual bug fixes. The model actually organically predicted 30%.
I still think that’s about accurate. Maybe it should’ve been 40%.
Everyone forgets that it was a pretty close election. Clinton could’ve won without the Comey announcement.
I think he did hedge (or "strategically bug fix"). The prediction for Trump went from IIRC around 15 to 30 in the last week or so. It was a big swing, IIRC with a lot of waffle around why it happened but not a lot of verifiable fact.
> I still think that’s about accurate. Maybe it should’ve been 40%.
It wasn't accurate. This is something people misunderstand about these predictions. If the 2016 election was held 100 times, Trump would have won 100 times. It's not the same as rolling dice.
These election predictions don't say that. They say something like "the observations I have agree with scenarios that have Clinton winning, 70% of the time". Which is fine and correct as far as his data and model goes, but none of those scenarios were the reality he was trying to predict. They are all just figments of the model though. Getting down to the brass tacks, he predicted Clinton would win, and he was wrong.
Which is fine, we just can't know anything about his process from that failure. Certainly we can't conclude that it was "accurate", since it was not. If we had a good sample of elections where he used the same process and built up a good record then sure.
That's the beauty of this brand of pseudoscience. Statistical predictions of singular events like a particular election are totally unfalsifiable. You can just say "I guess we live in 30% world" or whatever, every time.
1 reply →
To give you a trivial example: The simplest way I can put this is that turn out varies based on the weather[1], and turn out is skewed by party. So if it rains on election day you are going to get a different result, and that result can flip the outcome of the election if the election is close. So it’s kind of a nonsense to say. “Trump would have won 100 times out of 100”. Are you saying Nate Silvers model should have had a perfect meteorological model to predict the weather? Or are you saying the election wasn’t close? In which case you’re just wrong on the facts.
The 70% figure is saying “we know most of the information needed to determine what the outcome of the election will be but we don’t know everything so can’t be certain”. There is no process where you can know every factor that determines the result in advance with absolutely accuracy and I don’t know why people expect there would be.
[1] https://www.sciencedirect.com/science/article/pii/S026137942...
3 replies →
That's where you're wrong, the election was very, very close. In fact, if roughly 40k voters (across three states) had switched from Trump to Hillary, she would have won, that's how close it was.
40k voters, that's really not very many. So it's hard to say whether Trump had a 30% chance of winning or 40% or whatever, but the election at most was a toss-up.
Many random events could have resulted in a different outcome.
3 replies →
[dead]