The Comey Effect
This series of tweets from Will Jordan is fascinating. There was a very clear, major shift towards Trump in battleground states after October 28th, and the more polling that was done in this period, the more likely it was to be picked up. So this helps to explain the polling error in the Midwest. And it also makes it highly implausible that Comey’s letter and the 24/7 thigh-rubbing about Clinton’s EMAILS it catalyzed didn’t put Trump in the White House. I’m not saying Comey explains all 5 points, but how likely is it that he explains none of it? I dunno, maybe putting a former Ken Starr goon in charge of the FBI wasn’t a great idea? It also would have been nice if Anthony Weiner could have confined his sexting to adult women, although to be Scrupulously Fair it’s entirely possible that Comey could have found some other snipe hunt as an excuse to send a letter with no content about a pseudo-scandal with no content that would ultimately blow up the world.
In other findings, there’s no evidence that the “shy voter” effect contributed to the polling error.