While most of the political world’s attention has been focused on South Carolina, Mitt Romney has moved into a clear lead in a set of national polls conducted since his victory in the New Hampshire primary.
On average, Mr. Romney has 35 percent of the vote in an average of four national polls completed since New Hampshire, giving him a 19-point lead over Newt Gingrich and Rick Santorum.
Just how safe is a 19-point lead at this point in the campaign? Based on historical precedent, it is enough to all but assure that Mr. Romney will be the Republican nominee.
I went through our database of past primary polls for the 16 competitive nomination races from 1972 (when the current primary system was adopted) to 2008. For each cycle, I took a simple average of all national polls conducted in the two weeks after the New Hampshire primary was held. (The lone exception was the 1976 Democratic race, when there were no polls conducted within this window, so I expanded the range to three weeks after New Hampshire instead.)
In 13 of the 16 cases, the candidate leading in national polls after New Hampshire won his party’s nomination. In another case, the 1984 Democratic race, two candidates — Walter Mondale and Gary Hart — were tied in national surveys at this point in the race, and Mr. Mondale emerged victorious.
The only cases where a candidate came from behind to win were in 1972, when Edmund S. Muskie had a narrow 2-point lead in the polls over Hubert H. Humphrey, but George McGovern, far back in the polls, went on to win the race, and 2008, when Hillary Clinton had a 9-point lead over Barack Obama after New Hampshire but lost the nomination to him.
Note, however, that the leads held by Ms. Clinton and Mr. Muskie were considerably narrower than the 19-point advantage that Mr. Romney now has. Candidates who had a lead of the magnitude of Mr. Romney’s — like George H.W. Bush in 1988 — generally won their nominations with some ease.
One way to evaluate the data is through logistic regression analysis, which gives an estimate of how likely a candidate was to win the nomination based on the size of his national polling lead after New Hampshire. The analysis suggests that a 19-point lead in national polls at this stage of the race translates into roughly a 98 percent chance of winning the nomination.
To this point, I have avoided using the word “prediction” in the context of Mr. Romney’s lead or his chances of winning the nomination. The sample size of past nomination races is not very large, which means that the model’s estimates are not very robust and that it isn’t all that hard for a precedent to be broken.
If we had run this analysis at this point in 2008, for instance, the model would have noted that no candidate with larger than a 2-point national polling lead had lost his or her nomination up to that point in time, and the formula would have given Ms. Clinton almost a 99 percent chance of winning the Democratic nomination in 2008 — an estimate which would clearly have seemed dubious both at the time and after the fact.
Nevertheless, national polls are not quite the abstraction that they are sometimes taken to be, and they translate into significant leads for Mr. Romney in a number of important states.
In addition to having an 86 percent chance of winning South Carolina, for instance, Mr. Romney has roughly a 20-point lead in Florida and has in excess of a 90 percent chance of winning there. There have been no recent polls conducted of Nevada, but Mr. Romney’s position has generally been stronger there than his overall national polling.
The same holds for Michigan, which will vote in late February, and where Mr. Romney has always held a large lead. Arizona votes on the same day as Michigan; Mr. Romney holds a 27-point lead in a recent poll there. And if the contest makes it to Super Tuesday on March 6, Mr. Romney’s large resource advantages should give him a big edge.
Meanwhile, in a Gallup poll conducted just before New Hampshire, Mr. Romney was the only candidate deemed to be acceptable by a majority of Republican voters, although Mr. Santorum and Mr. Gingrich were close to the 50 percent mark.
Results like these, and recent polling that shows Mr. Romney leading each of his Republican opponents in hypothetical one-on-one match-ups even in a conservative state like South Carolina, suggest that Mr. Romney would remain the clear favorite to win the nomination even if the field consolidates and there are just one or two conservatives left to run against him. Meanwhile, Mr. Romney has the overwhelming support of the party establishment, which means that close calls are liable to be resolved in his favor.
Thus, although Mr. Romney would help himself to lock up the nomination with a win South Carolina, it is not clear how vulnerable he would be even with a loss there. Polls normally become considerably less volatile after New Hampshire as voter preferences become firmer, which means that Mr. Romney’s Republican rivals have already missed their best window to upend him.
I don’t know that I’d go so far as to say that Mr. Romney has a 98 percent chance of winning the nomination. But the odds attributed to him by the betting market Intrade, which now gives him a 90 percent chance, may if anything be too conservative.