Quote:
Originally Posted by Trolly McTrollson
Most of us were lol'ing because we thought Nate was grossly overestimating Trump's chances.
Any poll aggregation model is going to be garbage in / garbage out. Given that the polls were all systematically off, there's no way any model that worked for previous elections would get it right. *Why* the polling data sucked is an interesting and useful question, but lol'ing at Silver because he couldn't foresee a black swan event seems dumb.
The polls were not really systematically off. The national polls were pretty damned close, while the state ones were not as great, and, as Nate wrote in the lead up to the election, Trump was just a normal polling error away from winning.