Skip to main content
ABC News
The Florida State of Play

We’re exactly one week away from the Florida primary, which looks as though it could be pivotal in the Republican nomination campaign.

Let me give you a sense of what I’m seeing in the polls. The evidence isn’t necessarily all that straightforward. Five polls have been released in the state since Saturday’s South Carolina primary, and they have reasonably diverse results, including a nine-percentage-point lead for Newt Gingrich and a two-percentage-point lead for Mitt Romney:

You can nitpick at some of these polls if you like. The firm InsiderAdvantage has generally had favorable results for Mr. Gingrich, as it does in this instance. If you just average these polls together, however, they show a four-point lead for Mr. Gingrich.

FiveThirtyEight’s forecast model gives Mr. Gingrich a slightly larger lead, seven points. It assumes that there is still some favorable momentum for Mr. Gingrich that may not be fully accounted for by the surveys.

Is that a good assumption? One thing that makes this tricky to analyze is that nobody was polling in Florida last week, during which time Mr. Gingrich had begun to make significant gains on Mr. Romney nationally. Without that data, it is hard to separate out the momentum that Mr. Gingrich had before South Carolina from the additional momentum that he might have picked up from his big win in that state.

The clearest case that Mr. Gingrich still has momentum is not in the Florida data but instead from national surveys, particularly the Gallup national tracking poll. Mr. Gingrich now leads Mr. Romney in that poll by four points, 31 to 27. Moreover, Mr. Gingrich has continued to gain considerable ground on Mr. Romney every day — but only two-fifths of Gallup’s interviews in the tracking poll postdate his win in South Carolina, which means that it may still be a lagging indicator. (Another national poll, from Rasmussen Reports, which consisted entirely of interviews conducted since Mr. Gingrich’s South Carolina victory, instead gives Mr. Gingrich a seven-point lead.)

The national polls are worth looking at because the margin separating Mr. Gingrich and Mr. Romney in Florida has tracked pretty closely to the margin separating them in national polls. That makes sense, since the state is diverse enough to do a better job than the first few primary states of representing the G.O.P. electorate in microcosm. Mr. Romney does have some advantages in Florida — his campaign has long focused on Florida, a state that benefits a resource-rich candidate. But Mr. Gingrich does, too, since he is from neighboring Georgia and has strength among older voters.

The diversity in Florida may also help to explain why the polls don’t seem to be in strong agreement with one another. The Republican electorates in Iowa and New Hampshire are relatively homogeneous, demographically and geographically. In Florida, that is less true, so a pollster that has a poor response rate or does not properly balance its demographics may get unreliable results.

But whether Mr. Gingrich’s lead is four points or seven points or some other number is probably not worth worrying about too much, since since there’s still a week to go in the campaign. There are several dynamics in play that could affect the outcome:

Early voting. We discussed early voting at some length on Monday; it is probably an advantage for Mr. Romney, but not a great one.

What seems to be reasonably clear is that Mr. Romney banked some votes in Florida a week or two ago, when he was ahead in the polls there. But not necessarily all that many — about 225,000 Republicans had voted as of Sunday, representing only about 12 percent of a turnout that should eventually reach about 2 million.

Moreover, early voting is a continual process. Potentially, for instance, Mr. Gingrich is banking votes in Florida right now, since he seems to have the lead in the polls there now. So just as Mr. Romney was helped by early voting when he was ahead in the polls, he may be harmed by it now.

There’s a separate question about whether a candidate who benefits from early voting will tend to be underrated by the polls. In theory, this shouldn’t be the case — the pollsters can have an “already voted” category and can account for early voting when they weight their surveys. In practice, pollsters may not be that careful, or they may fail to distinguish a “likely voter” from one who has already cast his ballot. So if the polls show an exact tie, I would tend to break that tie for the candidate who had the edge in early voting. But I suspect that it does not represent more than, say, a one-point advantage for Mr. Romney relative to what the polls say.

Advertising. So far, Mr. Romney has had a big advantage in advertising in Florida. Mr. Romney’s campaign, and a “super PAC” that supports him, have collectively committed about $14 million to advertising in the state, according to NBC News.

But there are three important mitigating factors. First, Florida is a large state: spending $14 million on advertising there is the rough equivalent of spending about $2.5 million in Iowa. Second, much of this money has already been spent and therefore should already be priced into the polling.

Third and most important, Mr. Gingrich has now evened the score somewhat after a $6 million commitment from a super PAC that supports him, Winning our Future. If Mr. Gingrich had been running five points ahead of Mr. Romney in Florida in an environment in which the advertising was running heavily against him, it is plausible that he could do better than that once the distribution of ads becomes more even.

Debates. My personal view was that neither Mr. Gingrich nor Mr. Romney was at his best at Monday night’s debate in Tampa, Fla., but that Mr. Romney performed somewhat more strongly on balance. A survey  of Tampa-area debate watchers, meanwhile, showed no clear advantage: Mr. Gingrich was deemed to have won the debate by 30 percent of respondents, and Mr. Romney by 26 percent.

Either way, that debate does not seem likely to be a game changer, especially given the many competing threads in the news cycle, like the State of the Union address. Thursday’s debate in Jacksonville, Fla., of course, could be critical.

Dirty laundry. In the debate, Mr. Romney did a good job of focusing on two particular liabilities of Mr. Gingrich: his ties to Freddie Mac and his resignation as speaker of the House.

Mr. Romney was not making any new charges against Mr. Gingrich. But I don’t think it should be assumed that voters had already accounted for these issues in their assessments of the candidates. Polls regularly reveal that voters know much less about the biographies of the candidates than is sometimes assumed by the news media.

However, Mr. Romney also has his own liabilities. His tax returns are still dominating the news cycle, and the strategy that his campaign adopted — release 2010 and provisional 2011 returns but not earlier years — may not satiate voters and the news media’s curiosity about them. The super PAC ads are going to hit Mr. Romney on his health care bill, a potentially underexploited vulnerability. And Mr. Romney gave a somewhat awkward answer on immigration, an issue that is highly pertinent in Florida, at the debate Monday night.

Summing up. Our forecast model currently gives Mr. Gingrich a 75 percent chance of winning the state. That seems like too confident a prediction, frankly. On the other hand, Mr. Gingrich does appear to have a modest lead in the polls — and the nonpolling factors do not clearly favor one or the other candidate, in my view. So I would call Mr. Gingrich the favorite for now — but not a clear favorite.

Correction: A previous version of this post mistakenly said that a Florida poll conducted by Cherry Communications surveyed registered voters. The poll’s sample was of Republican voters deemed likely to vote in the state’s Jan. 31 primary.

Nate Silver is the founder and editor in chief of FiveThirtyEight.