Key excerpt (But it's worth reading the full thing):
But the real value-add of the model is not just in calculating who’s ahead in the polling average. Rather, it’s in understanding the uncertainties in the data: how accurate polls are in practice, and how these errors are correlated between the states. The final margins on Tuesday were actually quite close to the polling averages in the swing states, though less so in blue states, as I’ll discuss in a moment. But this was more or less a textbook illustration of the normal-sized polling error that we frequently wrote about [paid only; basically says that the polling errors could be correlated be correlated between states]. When polls miss low on Trump in one key state, they probably also will in most or all of the others.
In fact, because polling errors are highly correlated between states — and because Trump was ahead in 5 of the 7 swing states anyway — a Trump sweep of the swing states was actually our most common scenario, occurring in 20 percent of simulations. Following the same logic, the second most common outcome, happening 14 percent of the time, was a Harris swing state sweep.6
[Interactive table]
Relatedly, the final Electoral College tally will be 312 electoral votes for Trump and 226 for Harris. And Trump @ 312 was by far the most common outcome in our simulations, occurring 6 percent of the time. In fact, Trump 312/Harris 226 is the huge spike you see in our electoral vote distribution chart:
[Interactive graph]
The difference between 20 percent (the share of times Trump won all 7 swing states) and 6 percent (his getting exactly 312 electoral votes) is because sometimes, Trump winning all the swing states was part of a complete landslide where he penetrated further into blue territory. Conditional on winning all 7 swing states, for instance, Trump had a 22 percent chance of also winning New Mexico, a 21 percent chance at Minnesota, 19 percent in New Hampshire, 16 percent in Maine, 11 percent in Nebraska’s 2nd Congressional District, and 10 percent in Virginia. Trump won more than 312 electoral votes in 16 percent of our simulations.
But on Tuesday, there weren’t any upsets in the other states. So not only did Trump win with exactly 312 electoral votes, he also won with the exact map that occurred most often in our simulations, counting all 50 states, the District of Columbia and the congressional districts in Nebraska and Maine.
I don't know of an intuitive test for whether a forecast of a non-repeating event was well-reasoned (see, also, the lively debate over the performance of prediction markets), but this is Silver's initial defense of his 50-50 forecast. I'm unconvinced - if the modal outcome of the model was the actual result of the election, does that vindicate its internal correlations, indict its confidence in its output, both, neither... ? But I don't think it's irreconcilable that the model's modal outcome being real vindicates its internal correlations AND that its certainty was limited by the quality of the available data, so this hasn't lowered my opinion of Silver, either.
Jump in the discussion.
No email address required.
Notes -
I get so triggered by this logic because it’s so wrong. Elections are not a football game. They are not actually a random variable. On November 4th the result was already set in stone, unless one of the candidates died or something. You could replay November 5th 1000 times and Trump would win 1000 times. It wasn’t 50/50. It can never be 50/50. It is always 100/0.
Epistemic uncertain is a feature of the model and its inputs, not some inherent feature of the real world. There was enough data to conclude with relatively high certainty that Trump was on pace to win. Nate’s model didn’t pick up on this because it sucks. It has high epidemic uncertainty because it’s a bad model with bad inputs.
There have certainly been elections which were decided by tiny margins. They might well decided by the contrast in weather between the red and the blue part of the state. Now, you can say that Nate's model sucks because it does not sufficiently predict the weather months in advance.
We can score predictors against each other. A predictor who gives you a 50/50 on anything, like 'the sun will rise tomorrow' or 'Angela Merkel will be elected US president' will score rather poorly. ACX had a whole article on predictor scoring. If there is someone who outperforms Nate over sufficiently many elections, then we might listen to them instead. "I bet the farm on Trump, Biden, Trump and doubled my net worth each time" might be a good starting point, especially if their local election prediction results are as impressive.
Unfortunately, I have not encountered a lot of these people.
More options
Context Copy link
More options
Context Copy link