4350
Comments (1069)
sorted by:
You're viewing a single comment thread. View all comments, or full comment thread.
12
usausailoveusa 12 points ago +12 / -0

So I have a feeling Nate Cardboard will come out tonight and say "Just because Trump won, doesn't mean our models were wrong, we gave him a 10% chance of winning and he won, it's just probability.

Your correct response to that is, well Nate cardboard gave Trump a 30% chance of winning in 2016 and a 10% chance in 2020. If those two events are independent, then the probability that he got both calls wrong while his models were correct is 3%.

So do we think there is a 3% chance that his models were right but got the election wrong twice...or do we think there's higher than a 3% chance his models are garbage?

2
deleted 2 points ago +2 / -0
2
Examiner717 2 points ago +2 / -0

Which he's also been doing poorly. He only tracked them when they looked bad but when Covid declined he suddenly stopped tracking the numbers.