At this late stage of an election cycle, it's natural to pay attention to the most dynamic contests: those that are close, and where one candidate or another has made recent major gains in the polls. Indeed, you often hear about such candidates having "momentum going into Election Day."
But as Nate Silver demonstrated rather convincingly yesterday, "momentum" is a term that is misused as often in politics as in sports.
When people say a particular candidate has momentum, what they are implying is that present trends are likely to perpetuate themselves into the future. Say, for instance, that a candidate trailed by 10 points in a poll three weeks ago -- and now a new poll comes out showing the candidate down by just 5 points. It will frequently be said that this candidate "has the momentum", "is gaining ground," "is closing his deficit," or something similar.
Each of these phrases are in the present tense. They create the impression that -- if the candidate has gone from being 10 points down to 5 points down, then by next week, he'll have closed his deficit further: perhaps he'll even be ahead!
There's just one problem with this. It has no particular tendency toward being true.
Nate then goes on to examine polling of every Senate general election race since 1998, and establishes there's no clear relationship between gains made in one period of a contest and gains or losses experienced in the next.
As Nate acknowleges, there are sometimes factors which help a candidate improve his or her poll standing one week that could continue to operate later on. Most obviously, a candidate who hoards money until late in the campaign and then spends it is likely to make late gains. A candidate with low name ID who cuts a first TV ad is likely to make gains with a second TV ad as well. And many contests "tighten up" towards the partisan default position as voters begin to pay attention. But that's not the same as saying that this or that candidate "has momentum" as though polling trends have their own inexorable logic and power.
One scenario where "momentum" does make sense is when gains by a candidate appear to make the contest more or less competitive than was earlier the case, driving financial decisions by party committees or donors. And obviously, there can be external events--whether it's a change in the political climate or a candidate gaffe--that can drive the numbers in the same direction for an extended period of time. But these factors don't appear to come into play often enough to make a major mark on the empirical data.
Primary contests (which were outside the scope of Nate's study) may be different, if only because signs of candidate strength may be interpreted by voters as an indication of "electability," which some voters value. But the bottom line is that there is no particular reason to believe that candidates who are making gains in the polls released today are going to keep making them between now and Election Day. When a pundit talks about candidate "momentum," you should probably pay it no more mind than a sportscaster's assertion that a football team scoring a touchdown has "seized the momentum." You really can't know that until the horn sounds.