I hear what you’re saying and I think what you are alluding to is a reasonable consideration.
However, what most people do is go with a zero level analysis of “Nate said 98% and it didn’t happen. Nate is a idiot.” which completely ignores what 98% actually means. One needs to do a bit more than that to actually find fault with his projection.
On the other hand, if he projects 70% and it happens, he does take a victory lap. So I don’t feel bad for him. Bad logic just makes me sad.
Hmm, CA03 is gonna be a long shot. CA-41 definitely within reach. CA13 possible. CA22 is another long shot, might lose that one.
I’m super bummed about CA27. I got to meet Christy Smith at an event this year, and she’s pretty great. I also got to meet Katie Hill at that same event and was admittedly a little bit starstruck.
If Jay Chen can come back to win CA45, that means we can come back to win our race…the percentages are exactly the same (lol not happening)
If mail in are breaking more heavily blue than expected I still have 03 on the board. You gotta remember the bay area transplants to Placer since the pandemic are one of the higher migration numbers in the state
A little too much realism being injected in this thread.
I hate to defend Nate, but we should remember that 538 is a formula (which requires some subjective weightings). They live with their model, and claim to NOT put their thumb on the scale. When underlying fundamentals are different than their historical averages then the result is going to be off. Individual polls that do shitty in 2022 will be weighted less in 2024….etc.
Oh and with about 500 total races (house, senate, governors), we are going to see some outliers even if the model was 100% accurate. Shit happens outside of 2SD about 5% of the time.