Quote:
Originally Posted by Tyrone Slothrop
I don't understand the value of a model that tells the probability that a candidate will win a presidential election (e.g., Silver gives Trump a 64% chance of winning the Electoral College, per that article). This is not poker, where you play a lot of hands in a short period of time, and you can affect the stakes. A model that gives you that kind of prediction for a contest that only happens once every four years, with fixed stakes, has no value that is apparent to me other than in generating content to get people to click on stuff.
|
I don't understand it either. If Harris is up roughly 1-2% in every battleground state, I can't figure out how Trump can have a 64% chance of winning.
When he beat Hillary, where she was up huge #s everywhere, he had something like a 15% chance of winning. If you adjust those numbers to the present, it appears he's got somewhere between a 45-50% chance of winning.
Threaded Mode