Sam Wang did not eat a bug for breakfast.
Before Tuesday’s election, Wang, a Princeton neuroscientist and part-time election forecaster, promised to consume an insect if either Pennsylvania or Minnesota ended up going for Mitt Romney. If Ohio turned red, he pledged to eat “a really big bug.” (If you missed it, here’s Tuesday’s story about the rise of the poll quants.)
But he was right about those states and everything else. Wang was 50 for 50. (As I write this, Florida’s results are still being tallied. Wang had it in Obama’s column, but he also called the state a coin toss, correctly predicting that it would be the tightest race.) His prediction of the percentage of the popular vote going to each candidate was dead on: Obama 51.1, Romney 48.9. Oh, and he was also 10 for 10 in U.S. Senate races.
The poster boy for election forecasting, Nate Silver, who had been ridiculed by some conservatives, was also vindicated and deserves plaudits. But he wasn’t the only one to crunch the numbers correctly. Along with Wang, Drew Linzer of Votamatic was also correct across the board. Linzer, an assistant professor of political science at Emory, predicted way back in June that Obama would get 326 electoral votes, and his projections remained remarkably stable throughout the campaign.
Simon Jackman nailed it, too. In the earlier article, I noted that Jackman, a professor of political science at Stanford, had Obama down for 277 electoral votes—but those were the states in his model that leaned strongly toward Obama. In the end, he predicted, like Wang, Linzer, and Silver, that Obama would get 332, with the caveat that Florida was an extremely close call.
So the quants and their statistical models were right, while the pundits and their guts were wrong. This was also true in 2008: Silver and Wang were both nearly perfect. But in 2012 the victory of mathematics over bloviation was even more resounding somehow, perhaps because the battle lines were more clearly drawn.
Before the election, Peggy Noonan of The Wall Street Journal asserted that “nobody knows anything” about the potential outcome. David Brooks of The New York Times said forecasters lived in “sillyland.” Jennifer Rubin of The Washington Post tweeted that “averaging polls is junk.”
Can anyone seriously argue that now?
I checked in with the triumphant number-crunchers the morning after. Jackman writes by e-mail that the takeaway isn’t that Noonan and company were wrong and that he and the others were right, but rather that examining the data in a systematic, rigorous fashion is valuable: “It’s about the method, the approach, not the person doing it—think of that as another nail in the coffin of consulting the great oracle.”
That’s how Wang saw it too: “A major problem with the older generation of pundits is that they base their statements on gut feeling and political convictions, including their own biases. Noonan is a perfect example of someone who was unable to process the quantitative data right in front of her.”
Does this mean we should stop listening to pundits like George Will and Dick Morris, who both predicted a landslide for Romney in spite of the evidence? Wang puts it this way: “I believe that a little self-examination on their part would be in order.”
Linzer says he was heartened by the attention given to forecasting this time around. “We know that polls aren’t perfect, but they’re really good and we have a lot of them,” he says. “Why wouldn’t you want real information?”
For the record, Wang was confident enough not to shop around for an edible bug, though a rabbi friend offered the helpful tip that locusts are kosher.