In case you don't recognize him, that's Nate Silver of FiveThirtyEight—he's the dude who successfully predicted most of the results of the 2012 elections. He was widely derided by conservapundits, but one of his biggest critics has already eaten crow.
I never saw how Nate's process could be seriously questioned unless one were to completely disregard the very concept of polling. All he did—brilliantly—was aggregate all the polls minus a very few whose methods were suspect. He is all about logic. Hence, he came out smelling like a rose.
But while I read on DailyKos and elsewhere that many conservatives were in denial about the polling and were expecting a big Romney landslide based on their gut feelings, I never believed they really believed that because I never traveled to Drudge or Red State or other right-wing sites myself.
Then, on Facebook and after the election, I saw a post from early on November 6 in which an absolutely gonzo right-wing friend of a Libertarian friend had posted that he was sure Romney would kick ass based on a fabulously positive Yahoo! poll...like, an online poll, not an actual poll of the field. You know, something that could be easily run up depending on who bothered to vote and who bothered to point to it. He was convinced by this.
And that amazes me.