I have absolutely no idea what’s going to happen in Nevada tomorrow, and any pollster or pundit who says they do is fooling themselves. You simply can’t predict turnout with enough accuracy for polls to be worthwhile, or organizational efficacy with enough accuracy to predict who’ll run the best GOTV, in a race this brand-new. This is the first time Nevada’s had caucuses that are worth a damn, and so we have no precedent on which to base predictions. So I won’t make any.
I will, however, predict that Mike Huckabee will beat John McCain in the Republican primary in South Carolina. Part of this is wishful thinking; a Huckabee win would render the race overall a three-way tie between him, McCain, and Romney, and increase the likelihood of a brokered convention, which would be all kinds of awesome. More than that, however, it would serve as an upset against McCain, who’s much more electable than the rest of the field while not being substantially better on policy, and who thus must be denied the nomination, even if it means a crazy person like Huckabee or a freakishly transparent liar like Romney gets the nod instead.
But this isn’t just projection; I think Huckabee has late momentum going into tomorrow. Check the polls at Real Clear Politics. A five-poll average right after New Hampshire has McCain up 6.4%; an average of the last five polls shows him up only 3.4%. Also, I think Huckabee’s supporters are better organized, being regular churchgoers, while McCain’s are largely independents who may opt at the last minute to vote in the Democratic primary next week instead.
My main reason for thinking Huckabee will pull this off, however, is that this primary season has thus far been, in both parties, a huge validation of the Feiler Faster Thesis. According to Bruce Feiler, as quoted by Mickey Kaus, the increased speed, adaptiveness, and quantity of media has made political momentum a much more fickle phenomenon, such that it can swap candidates within the course of a few days. This was first used to describe Bush’s come-from-behind win against McCain in (heh) South Carolina in 2000, only a couple weeks after New Hampshire. By 2004, it seemed like the thesis was defunct as Kerry parlayed his Iowa win into victories in all but a few primaries. And I, like many others, thought it would be wrong this year too. But Feiler Faster came back with a vengeance. New Hampshire, rather than repeating Iowa’s choice of Huckabee and Obama, chose McCain and Clinton instead, with only five days between the contests. Michigan, rather than keeping the McCain bandwagon rolling, chose Romney, only a week after New Hampshire. Feiler was right, and I was totally wrong. And I see no reason why he won’t keep being right. And Feiler Faster predicts a Huckabee victory. After New Hampshire and Michigan, so do I.