this is waaaay better than even I expected. Wow!
somehow Nate Silver's math isn't adding up to me - all of these polls should improve Clinton's chances.
Are you critiquing Silver or me?
I mean I guess the big swinger is the Missouri poll showing Trump up 11, but really why should a Missouri poll swing it that much when we basically know Trump is going to win Missouri?
It wasn't meant as critique of this particular post, but about Nate being a clickbait.
I mean it is more or less the same model that was used in 2012 etc. Plus Nate is probably extra cautious bc of ~Brexit'ish phenomena.
I think, Pew was the mostly responsible for it. It has pretty big sample/weight.
State correlation?