George,

It's dizzying to watch and listen to the various kinds of spin about the Democratic results in the New Hampshire primary. People will say anything at all that crosses their mind! I'm thinking especially of the bloviations of the CNN "best political team on television." Given the low quality of commercial television, this may be accurate.

That said, Kevin Drum, in his blog at washingtonmonthly.com, looked at the percentages totted up in the various opinion polls taken during the days before the vote, and suggested a small, sensible hypothesis. I like it. In general, opinion polls, as far as I'm concerned, fall under the rubric of lies, damned lies, and statistics. Here is Kevin Drum's post.

HOW DID HILLARY WIN?....So what happened last night? How could the polls have been so far off, predicting an Obama landslide only to have Hillary Clinton pull off a narrow victory?

The first answer is: primary polling is historically difficult. For a variety of reasons, especially in an open-primary state like New Hampshire, it's really hard to get reliable results. Being off by double digits isn't exactly common, but it's not all that rare either.

That said, last night's results really were at the high end of unusual, and none of the obvious possibilities seem to explain it. For what it's worth, Time's Jay Carney, via "a social scientist friend of a colleague" who did some comparisons of polls vs. actual turnout, seems to have the most plausible explanation:

What he found...is that a certain percentage of Democratic voters in the last days of polling presumed Biden (especially) and (to a lesser degree) Dodd hadn't dropped out. By and large, come election day, those Biden and Dodd supporters ended up casting ballots for Hillary. Also, of the 5 percent or so who were still undecideds in the last polls, almost all broke for Hillary.

This makes sense to me. None of the "big" explanations seem to pan out, so it's most likely a collection of little explanations: a few points from Biden supporters, a few points from Dodd supporters, a few points from undecideds, a little bit better turnout from women, and perhaps a bit of polling error in the post-Iowa polls. Add it all up and you get a 10-12 point swing. It's not a sexy explanation, but it seems like it's probably the right one.

Always nice to have a reasonable alternative to conspiracy theories. Especially given the constant, tiresome misogyny Sen. Clinton is subjected to. But that takes us back to the pundits' bloviations, doesn't it, which is not my real subject.

Katherine McNamara (Electronic mail, January 10, 2008)