Foxy Nate Silver and why old-media hedgehogs could soon be old news
Nate Silver – statistician, analyst, blogger, author, and Xanax for liberals – is a one-man startup, and his brand is on the line. Aside from the candidates themselves, perhaps no man stands to lose more from this election. Silver, the 34-year-old-wunderkind whose weighted statistical model for forecasting the election result offers (as of today) a 80.9 percent probability that Barack Obama will win the Presidency on Tuesday, has effectively staked his reputation on the outcome.
Not only that, but Silver’s disruptive approach is also being put to the test against the conventional pollsters and traditional media pundits who have for decades dominated political discourse. It’s kind of like the "Moneyball" whiz kid meeting the aging baseball scout who favors his gut over data. Silver may represent the beginning of a new era in which punditry could be undermined, and perhaps overthrown, by a data-driven approach that eschews popular media narratives. But that all depends on Obama winning Tuesday, despite some polls (namely Gallup, which Silver brushes off as a Republican-leaning “outlier”) that put Romney ahead in the popular vote.
Silver has become a national figure thanks to his FiveThirtyEight election forecasts blog, which is published by the New York Times; Twitter, on which he has more than 275,000 followers; and his appearances on talking head shows on MSNBC, the Daily Show, and the like. This has helped him get around, and inside, the old-media fortress that has for decades protected pundits like Dick “the World’s Worst Political Pundit” Morris and “Morning Joe” Scarborough, who aside from an aptitude for television seem prone to basic errors of math.
On his morning news show yesterday, Scarborough took a shot at Silver, although he didn’t name him. “Anybody that thinks that this race is anything but a toss-up 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,” Scarborough said.
Silver couldn’t help himself. “If you think it’s a toss-up, let’s bet,” Silver Tweeted at Scarborough. “If Obama wins, you donate $1,000 to the American Red Cross. If Romney wins, I do. Deal?”
It sounded like trash talk – the New York Times’ public editor certainly thought so – but a more dispassionate assessment of the Tweet underscores the data-versus-gut tensions that underlie the whole Silver debate. If you think the odds are really 50/50, Silver was saying, let’s see you put your money on Romney. I, meanwhile, like Obama’s odds. A lot. Silver was subtly shifting the debate back to probability, and away from the emotive but ill-informed “too close to call” narrative peddled by Scarborough and his ilk. It's a problem of "understanding" versus "hedging," as Natalia Cecire pointed out in a blog post last night. "But understanding has nothing to do with it; there are simulations, and they indicate the probability of potential outcomes. It's not understanding; it's pointing."
You see, what it really comes down to is that Scarborough is a hedgehog, and Silver is a fox. Let me explain.
In his book “The Signal and the Noise,” Silver highlights a study by psychologist Philip Tetlock that showed that the more an “expert” appeared in the press, the worse his predictions tended to be. Tetlock ascribes that failing of the expert to a trap in the way such people think about the world. In a book titled “Expert Political Judgment,” Tetlock describes two types of thinkers: “Hedgehogs” and “Foxes.” (Update: Readers have pointed out that Isaiah Berlin actually came up with the categories; Tetlock, like Silver, was merely citing them.)
Hedgehogs are “type A personalities who believe in Big Ideas – in governing principles about the world that behave as though they were physical laws and undergird virtually every interaction in society.” Foxes, on the other hand, are “scrappy creatures who believe in a plethora of little ideas and in taking a multitude of approaches toward a problem. They tend to be more tolerant of nuance, uncertainty, complexity, and dissenting opinion.” Hedgehogs are more easily seduced by clear narratives. Foxes are more data-driven, less willing to stake out strong positions.
One of these groups is more likely to appear on television. Guess which.
Tetlock’s research suggests, however, that foxes are considerably better at forecasting than hedgehogs.
Yet media pundits, because they’re hedgehogs, are more easily distracted by compelling narratives such as Romney’s momentum or Obama’s debate problems at the expense of focusing on what the data actually says. These hedgehogs might also suffer from what the psychologist Daniel Kahneman calls “confirmation bias.” Says Kahneman in his book, “Thinking Fast and Slow”: “Contrary to the rules of philosophers of science, who advise testing hypotheses by trying to refute them, people (and scientists, quite often) seek data that are likely to be compatible with the beliefs they currently hold.”
That may explain why we’ve seen a sustained attack on Silver’s credibility this week. On Monday, for instance, Politico’s Dylan Byers published a piece suggesting Silver might be a “one-term celebrity,” quoting New York Times columnist David Brooks and MSNBC’s Scarborough casting doubt on Silver’s forecasts. Because they’re hedgehogs, people who are not only seduced by compelling stories but also tell them, they fail to separate Silver the Man from Silver the Analyst.
So caught up are they in his celebrity and the media narratives that surround the election, they neglect to consider the story of Silver’s data. And actually, Silver isn’t alone in his forecast of an Obama victory. As he pointed out in a Tweet yesterday, “Every bookmaker from Las Vegas to London” strongly favors Obama. So do predictions markets like Intrade.
They fixate on Silver because they mistake him for a pundit whose opinions are developed as a marketing tool instead of the product of math, and it’s easier to rail against an individual – especially one affiliated with the liberal paper of record – than it would be to take on faceless oddsmakers or masses of people betting on the election outcome.
Silver also poses a problem to the pundit class. If Silver’s method proves more accurate than their soapbox prognostications, perhaps the public will decide it’s no longer interested in what they have to say. The big polling companies – Gallup, Rasmussen, Pharos – would feel threatened, too, because Silver’s model reveals and contextualizes the shortcomings of their individual efforts, even as he relies on them for data aggregation.
These are well-founded fears. Silver is a disrupter. His methodology and communicative approach challenge the established ways of the pollsters and the pundits. His scientific approach and technological tools pay little heed to the prevailing media narratives of the day. He threatens to set in motion a movement in which hacks and hucksters should no longer expect to be given a platform based on their perfect hair or ability to speak to camera. When it comes to polling and punditry, the gut is out and science is in.
Unless, of course, Romney is voted into office. Then Silver’s reputation would be in tatters and the hedgehogs will have won. But according to Silver there’s only a 19.1 percent chance of that happening – at least today.
Tomorrow, who knows?
[Photo by Randy Stewart]