Mitt Romney was so sure he'd win the presidency that he didn't listen to national polls or write a concession speech, a sure sign that even an acclaimed data-loving technocrat found it wasn't able to resist the ongoing science-denial among some conservatives that we saw in the last months of the campaign. Romney's campaign has pushed back against that idea — and some supporters' speculation they were misled by the campaign — by insisting its internal polls showed a much closer race. Now comes word that the campaign has revealed some of those poll numbers to The New Republic's Noam Schieber, but they're not entirely exculpatory. 

The Romney campaign gave TNR the November 3 - 4 internal polls for Wisconsin, Pennsylvania, Minnesota, Iowa, Colorado, and New Hampshire. They show Romney and Obama tied in Iowa, and Romney ahead in Colorado and New Hampshire. (Obama won all of those states.) They did not make public the internal polls in Florida or Virginia, though the Romney campaign was reportedly confident they would win both. (He lost both.) Scheiber writes:

Together, New Hampshire, Colorado, and Iowa go most of the way toward explaining why the Romney campaign believed it was so well-positioned. When combined with North Carolina, Florida, and Virginia—the trio of states the Romney campaign assumed were largely in the bag—Romney would bank 267 electoral votes, only three shy of the magic number. 

But only three votes shy of the magic number is still three votes shy of the magic number. Romney pollster Neil Newhouse told TNR that its internals showed Romney behind by just 2 points in Ohio. But only 2 points is still 2 points. Romney's team felt they had enthusiasm and momentum on their side, and they measured this by looking at the people who were most enthusiastic about the election, people the campaign found were more likely to back Romney. When asked to explain why his polls were off, Newhouse said "we didn’t have much of a house effect," and that its incorrect results "may be a function of Sunday polling."