We are beginning to see more national surveys now, including this week’s New York Times/CBS News poll, which show Mitt Romney with a slight lead over President Obama in the general election matchup. To be sure, there are also a number of polls that put Mr. Obama slightly ahead. But his lead does seem to have narrowed — from about three or four points in an average of national polls a month or two ago to more like a point or so in surveys today.
Has something fundamental changed in the race?
Probably not. Instead, I suspect we are seeing some reversion to the mean. It could be that Mr. Obama’s larger lead from before was somewhat ephemeral, although there are a couple of factors that may be working in Mr. Romney’s favor at the margin.
Although we are getting to the point where these national polls are at least worth a passing glance, it is still also worth paying attention to Mr. Obama’s approval rating. These have a history of predicting electoral outcomes at least as closely as head-to-head polls in the early stages of the race, especially for incumbent presidents.
Mr. Obama’s approval ratings have not moved all that much. For the last month or two, they have been essentially even. Right now, in the RealClearPolitics average, 48.3 percent of Americans approve of the job that Mr. Obama is doing, and 48.6 percent disapprove.
A president can get re-elected with numbers like those. Obviously, he can also lose. But the fact that Mr. Obama’s approval ratings are close to even means that it should not be surprising that the numbers in his matchup against Mr. Romney are getting closer to even, too.
I am not a purist who says that candidates and campaigns make no difference. That said, the most reliable benchmark in the past of when presidential results deviate from those predicted by approval ratings is when one of the candidates has a relatively “extreme” ideology, like Barry M. Goldwater or George S. McGovern. Mr. Romney does not qualify as an extremist by the various measures we can look at that attempt to quantify this objectively — neither does he qualify as a moderate. Instead, he is a “generic Republican,” who might run fairly close to the outcomes predicted by Mr. Obama’s approval ratings.
Mr. Romney also went through a period where his favorability ratings were quite poor. However, they have since improved to about even, possibly because his job has been less complicated since the effective end of the Republican primary campaign. It is not uncommon for favorability ratings to shift over the course of a campaign, particularly once the primaries end.
I am working on our general election model, which can begin to look at some of this data in a more formal way. No model is perfect, but some things probably benefit from applying a more formal and rigorous structure. For instance, Mr. Obama’s state-by-state polls look a pinch better on balance than his national numbers. It is hard to know if there is something real there; the timing of the polls and the relative ordering of the states in the Electoral College need to be carefully considered.
Likewise, most of the polls we are seeing now are of registered voters or adults rather than likely voters. That is something that needs to be accounted for. Usually a shift from registered voters to likely voters will improve the Republican candidate’s standing by a couple of points because Republican voters are older, wealthier and more likely to turn out in the average election. On the other hand, I do not think it is a good strategy to look solely at likely voter polls while ignoring the others. Some survey firms that use likely voter models have a history of a partisan skew in their polls that has nothing to do with the likely voter model itself.
The last thing to remember is that when an election is quite close, it does not take very much to shift the race from one candidate being a 60/40 favorite to it being about even.
At the betting market Intrade, Mr. Obama’s odds of re-election have consistently been around 60 percent. While, on the one hand, it is good not to overreact to new data at this early stage of the race, it is also worth remembering that even a one-point shift in a president’s approval ratings, or a modest change in the economic forecasts, can move a president’s re-election odds at the margin.
The marginal factors in the past four or six weeks have looked slightly better for Mr. Romney than Mr. Obama. The recuperation in Mr. Romney’s favorability numbers, for instance — if it was somewhat to be expected — also reduces the risk that his personal qualities might cause him to lose an election that he otherwise would have won, as during another economic downturn.
Meanwhile, the tumultuous situation in Greece may increase the chance of an economic downside case for Mr. Obama. And data from the domestic economy has not been as strong lately. (Although it might also be mentioned that the situation in the Middle East is thought to be improving, and oil pries have receded somewhat.)
Put another way, if you are being very detail oriented, there is a case to be made that Mr. Romney’s odds of being elected have improved somewhat over the past six weeks. I suppose we can be even more detail oriented once we get the general election model up and running. If you just want the 30,000-foot view … well, it looks pretty close, as it did before.
What I am less convinced by is the idea that anything in the campaign — the day-to-day stories that the news media covers — has mattered very much so far. One of the reasons that campaign stories have been so trivial lately is because if one of the campaigns has an especially strong line of attack on their opponent, or a great piece of opposition research, it does not make a lot of sense to drop it now when most voters are not paying attention yet. It is still extremely early for a general election campaign. If the period after Labor Day qualifies as the pennant race, and the summer of the general election year the regular season, we are still playing preseason baseball now.
A version of this column that appeared in print is available here.