The gossipy, horse race-obsessed outlet Politico ran a story on October 29 about the credibility of polling expert Nate Silver, whose 538 blog at the New York Times is a must-read for people interested in election forecasting.
What Silver does isn't, on one level, all that tricky–his model combines national and state polls and generates probabilities about election outcomes. This model finds it highly likely that Barack Obama will win the election. It's probability, not a crystal ball or a bet.
Politico's Dylan Byers notes that Silver's model says this "even as the polls have [Romney] almost neck-and-neck with the incumbent." By which he seems to mean the national polls, which are not entirely relevant to the election outcome, given the existence of the Electoral College.
The real issue here is that Silver's approach seems to bother some pundits and Beltway journalists. This is ironic, since they obsess over the very same horse race, minus the statistical rigor. To them, the horse race is about unquantifiable things like "momentum" and "gaffes" and the body language of the candidates. Silver's model suggests that the election has been a lot more stable than the one many campaign reporters have been reporting.
So if this piece is about Silver's credibility–and it is, since it is concerned with the question of whether or not if Silver is a "one-term celebrity"–who are the critics questioning his approach?
What do they have to say? Not much. Byers points to a Brooks column that diminished polling itself:
If you tell me you think you can quantify an event that is about to happen that you don't expect, like the 47 percent comment or a debate performance, I think you think you are a wizard. That's not possible.
Whatever that might mean is anyone's guess, but I have never been under the impression that Nate Silver thinks he "quantifies events that are about to happen."
Scarborough's critique is much more direct:
And anybody that thinks that this race is anything but a tossup right now is such an ideologue, they should be kept away from typewriters, computers, laptops and microphones for the next 10 days, because they're jokes.
So we've got a critique of Silver based on one columnist's nonsensical dismissal of polling, and a TV pundit's feeling that the race is actually very close–based mostly on what the campaigns are telling him. This is, it should be noted, the same Joe Scarborough who tore into critics of the Iraq War and demanded that some of them apologize for being wrong, because we won the war in April 2003. Yes, let's hear more about what he thinks about how faulty other people's predictions can be.
It's hard for one story to capture so much of what is wrong with corporate media, but Politico managed to do it–and probably not on purpose. We have a questionable premise: "Critics Question Nate Silver's Methods." Those critics do not offer legitimate analysis; they have vague criticisms that seem to misunderstand probability and statistics. And these people happen to have high-profile positions in the media, where they can continue to churn out election coverage obsessed with phantom concepts like a candidate's momentum–things that, hey look at that, it turns out they get to define.
It's enough to make you glad the campaign season is almost over. Unfortunately, corporate media are likely to start their 2016 election coverage sometime next week.