Getting it right

In 1633, the Roman Inquisition formally charged the astronomer Galileo Galilei with heresy for holding the belief that the Earth revolved around the sun. The Church’s qualms, of course, were not with the accuracy of Galileo’s evidence or the methodology used to gather and analyze it; rather, Papal authorities saw the heliocentric model of the universe as an ideological affront. In a letter to his contemporary Kepler, Galileo complained that many of the Jesuit astronomers who objected to his theories refused even to look into his telescopes, despite frequent invitations to do so.

Though it took until 1992, the Church has reversed its condemnation of Galileo. However, the collection and presentation of objective data are still being confused with the ideological advocacy of the data’s implications. Nowhere in recent memory has this conflation been more apparent than in the United States presidential election.

Former baseball statistician Nate Silver made a name for himself by correctly predicting the winner in 49 of 50 states in the 2008 election – as well as the winner of every senate race – on his blog FiveThirtyEight. The principle difference between his method and those utilized by nearly every other pundit and talking head in the political media is that instead of relying on “gut feeling” and other subjective cues, Silver relies solely on numbers.

Amazingly, the stunning accuracy of Silver’s 2008 predictions did not cause other pundits to rethink their methodologies and phase out forecasting based on subjectivity and superstition. In 2012, the reality was quite the opposite. When reliant on subjective cues, a forecaster can use a prediction to reaffirm his or her ideology, and that’s exactly what happened. Republicans pundits predicted a Republican win, not because they had access to different data, but instead because that’s what they wanted to happen.

Former Reagan speechwriter Peggy Noonan made the following case for a Romney landslide in a Wall Street Journal blog post: “All the vibrations are right.” She asks, “Is it possible this whole thing is playing out before our eyes and we’re not really noticing because we’re too busy looking at data on paper instead of what’s in front of us?” Rather than data on paper, she cites the number of Romney yard signs she saw on a trip to Florida, the energy of the Republican’s supporters at rallies, and Obama’s distracted look at a charity dinner.

Related  Forget the Foetus, that's enough royal baby for now

Countless other conservative media figures made similar predictions. “The average pollster is either biased or has terrible gut instincts” wrote Wayne Allyn Root in the Washington Times. “I have a history of predicting political winners and losers without ever taking a poll. I just take the pulse of the thousands of people I know… What I see and hear is a coming Mitt Romney landslide.”

Patent dismissals of polling data allowed these pundits to make whatever prognostications best served their belief systems. Many high-profile members of the conservative media ignorantly assumed Silver was doing the same. Like Rome to Galileo, they likened his objective data to an ideological manifesto based on the data’s implications. Because Silver had long predicted an Obama win, conservatives attacked Silver as a partisan demagogue, attempting to advance liberal agenda through his predictions. Referring directly to Silver, MSNBC’s Joe Scarborough remarked that “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.” Scarborough cites his “gut” in a Politico column as a source for the tossup analysis. On top of disputing the possibility that Silver would repeat his 2008 success, UnSkewedPolls.com’s Dean Chambers noted that “Nate Silver is a man of very small stature, a thin and effeminate man with a soft-sounding voice that sounds almost exactly like the ‘Mr. New Castrati’ voice used by Rush Limbaugh on his program.” Chambers predicted a comfortable Romney win. David Brooks of the New York Times charged, “If you tell me you think you can quantify an event that is about to happen… I think you think you are a wizard.”

Legitimate complaints about Silver’s methodology were vanishingly rare. The reactions to Silver’s predictions betrayed a fundamental misunderstanding of data and mathematics. Comments indicating poor comprehension of statistics were common, and those that conflated data with ideological advocacy were ubiquitous.

Related  Grad student: Oxford fails to accommodate perspectives it invites

Brooks got it wrong. Silver was not the self-assessed wizard. If a pundit could make an accurate prediction based on “vibrations” or “gut feeling,” a letter from Hogwarts might be in order. No, Silver’s predictions were not the stuff of magic. He simply aggregated polling data and converted the figures into probable electoral votes, with no room to factor in either his personal politics or his indigestion.

And yet it moves. The Earth revolves around the sun, and Silver correctly predicted the winner of all 50 states. This was a landslide in the contest of mathematics versus superstition. Gut feelings most likely will not graciously concede defeat anytime soon, but there are glimmers of progress. Sales of Silver’s book The Signal and the Noise: Why So Many Predictions Fail – But Some Don’t have skyrocketed in recent days. Quantitative election analysts Drew Linzer and Sam Wang, whose predictions proved almost as accurate as Silver’s, have also benefitted from increased visibility. Perhaps a heightened interest in data science will afford the public a better understanding of predictions. Most importantly, Obama’s affordable healthcare laws should allow more Americans to cure those curious premonitory feelings of the gut.