That's why it is liable to get very nasty in the Palmetto State. The narrative isn't set in stone yet, but the concrete is drying and there is precious little time to derail the Romney train.
The forecasts had Mr. Romney with 32 percent of the vote in South Carolina, Newt Gingrich with 24 percent and Rick Santorum with 20 percent. That didn't appear to be all that safe a lead for Mr. Romney, and the model gave him only a 55 percent chance of winning.
However, these results will probably be affected by what happened in New Hampshire.
The best empirical way to estimate the "bounce" a candidate will get out of a state is to look at how he performed relative to his polls.
Mr. Romney got 39 percent of the vote in New Hampshire, which almost exactly matched his polling-based forecasts. So he is unlikely to get much of a bounce for having exceeded expectations.
But bounces aren't entirely relative; absolute performance can matter, too, especially if candidates get favorable news coverage from their performance. Mr. Romney did get that favorable coverage Tuesday night. (This was well deserved in my view: he had a clear victory in New Hampshire, which was called by the news organizations almost as soon as the polls closed.)
So the worst-case scenario for Mr. Romney is probably that his numbers hold steady in South Carolina. More likely, he will get a modest bounce, and there is some chance he will get a fairly big bounce.
It is frustrating for other candidates and their supporters to deal with such intangibles as "momentum," and "electability." If they had their way, they would wish that each primary was an island unto itself, that what had happened previously in the race didn't matter.
This is impossible. In Romney's case, perception trumps all. And it is being perceived by party members across the country that Mitt is a winner and his campaign can't be stopped. It doesn't matter if it's true or not. The bandwagon has begun to roll and both ideologues and job seekers are hopping on board early.
As Silver points out,"The problem is that there may be no Republican candidate poised to benefit from a downtrend in Mr. Romney's favorability numbers." Even if attacks on Mitt work, they can only slow him down at this point, not stop him.