Polls and Pundits

One of the interesting aspects of the presidential election is the gap between the reporting on the horse race and the way statistical nerds view the race. Today’s New York Times has an article about the narrowing presidential race, based on an analysis of a single national poll.  But at the same time on the Times’s 538 blog, Nate Silver shows what he’s been showing for some time — the odds of Obama winning are about three to one, and they’ve been growing steadily since a dip after the first debate. Sam Wang of the Princeton Election Consortium is even more certain that Obama is going to win, giving a probability of over 90%.  Here is another site run by a political science professor that comes up with similar results.

These guys aggregate polls and apply sophisticated mathematical models to the results.  And they have been successful. 538 correctly predicted the results in all but one state in 2008; the Princeton Election Consortium doesn’t predict individual states, but they came within one electoral vote of the actual result in 2008.  But with the success has come criticism:

[David] Brooks doubled down on this charge in a column last week: “I should treat polls as a fuzzy snapshot of a moment in time. I should not read them, and think I understand the future,” he wrote. “If there’s one thing we know, it’s that even experts with fancy computer models are terrible at predicting human behavior.”

On MSNBC’s “Morning Joe” today, Joe Scarborough took a more direct shot, effectively calling Silver an ideologue and “a joke.”

But it is self-evidently not true that experts with fancy computer models are terrible at predicting human behavior.  They do it all the time, as Silver and Wang have proven.  Statistics works — that’s why polling works.  Prediction is imperfect — that’s why none of these sites say with certainty that Obama is going to win; but data-driven prediction is surely better than Morning Joe talking to a bunch of other media types about the enthusiasm level at Romney rallies and the rising confidence level of Romney advisers.

I was watching TV at the gym today without the sound, and I saw a brief report on the presidential race on the local news.  There was one statistic displayed–Romney ahead by three in the latest national poll from the station’s network.  This was followed by brief clips of Obama acting presidential in the hurricane aftermath and Romney lifting a case of bottled water that had been collected for hurricane relief at his rally; then there was a 15-second comment from some Politico pundit.  And that was it.

There are many reasons why I don’t watch TV news anymore, and this is one of them: it just makes you stupider.  Silver, Wang, and others like them may end up being wrong, but ignoring them or mocking them just seems stupid.  About as stupid as thinking that Mitt Romney’s latest position about anything is what he really believes.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s