We have to be careful with this. Silver goes to far when he says that if Newt continues with "a win in Florida, it would suggest that we had been weighing the evidence incorrectly all along." That's too strong. No political scientist argues that the party consensus candidate will win every single primary; indeed, I don't think there's really an argument out there that the party consensus candidate should win the nomination easily. And Mitt Romney, while certainly a solid leader by every measure of party actors that we have, isn't (as Seth points out) nearly as dominant by measures of party actor support as some previous frontrunners have been. So for Romney to have to struggle some is no big deal at all. Even if Florida turns out to be just like South Carolina (solid win for Newt, with Romney a solid second place) I would think that Romney will still be the overwhelming favorite to win the nomination. The next month is on much more favorable turf for the Mittster, giving him -- and party actors who strongly prefer him over Gingrich -- plenty of time to retool and kill Newt's campaign off again by the time we get to Super Tuesday in early March.
I'll also point you to a good Hans Noel comment to John's post. Hans notes:
The trouble with “this time it’s different” is that, even if the basic mechanisms are the same, sometimes it’s different. In a probabilistic world, one case is just one case, and should be treated as such...If [party leaders] lose (and I don’t think they will), that’s not proof that they don’t generally control their nomination. It’s evidence that even though they have a lot of influence, sometimes they lose. I think we’d need several contests to go to a Gingrich (or to a Carter) before we’d conclude that everything had changed.Hans is certainly right in general, but I think in this particular case I'd disagree. Newt Gingrich, from what we know, isn't really similar to a Carter 1976, Hart 1984, or Huckabee 2008. He's more like Hart 1988 or Giuliani 2008 -- someone who party actors, or at least one group of important party actors, strongly oppose. For Rick Santorum to beat Mitt Romney would be one thing; for Newt to do it would be a very strong signal that the way we understand these things -- certainly the way I understand these things -- is wrong.
Now, the question is whether our understanding would be wrong because it was always wrong -- party actors were never as important as we thought -- or because something has changed now about the process. I should add one more possibility: our understanding of the process is correct, but the party itself changed so much that standard measures of accounting for the influence of key party actors don't work any more.
For now, however, the best bet is still that Mitt Romney wins the nomination, and we'll look back and realize that he won it fairly easily -- and especially that we'll look back and agree that Newt Gingrich never had a realistic chance of being nominated. That's still my analysis for now, and probably will be almost regardless of what happens in Florida. I do want to be open to evidence that "this time is different," but so far at least I'm not really seeing any.