FiveThirtyEight on Horse-Race Journalism
Nate Silver, the writer/statistician behind the very fine FiveThirtyEight blog, has found himself in a number of crosshairs of late. His politically motivated attackers are risible zealots unworthy of discussion, but things are more interesting when you consider the attacks and the passive-aggressive grumbles from political journalists. (Oh, and the New York Times’ public editor called him on the carpet for a Twitter bet, though as a fellow ombudsman I’m not touching that one.)
The argument within journalism is almost an exact replay of one that’s exceedingly familiar to Silver: the “Moneyball” scouts vs. stats debate within baseball.
If you’ve followed baseball over the last decade or so, you’ll instantly see it’s the exact same debate.
For the old baseball scouts who trust their eyes and their guts, sub “the savvy,” the journalists who assemble narratives from interviews, observations and their own experience — their ears and their guts, if you will. For the baseball stats guys, sub folks like Silver who wade into polls and try to weigh bias, calculate probabilities and make predictions.
Here’s the thing, though: This debate actually ended a long time ago in baseball. Every front office has people who mine advanced stats and try to value players objectively. Some front offices give it more weight than others, but all of them understand the value of the Moneyball approach and take at least some heed of its lessons. The debate only continues among hack columnists and announcers, intellectual refusenik fans and people who know better but can’t resist fighting with them.
So what about journalism? Well, I suppose it’s progress that FiveThirtyEight now appears under the umbrella of the New York Times. But some elbows are getting thrown beneath its shelter. Here’s Silver today, turning a merciless eye on the latest batch of battleground state polls and explaining what has to be wrong with their methodology for Mitt Romney to be elected on Tuesday night. And here are Jeff Zeleny and Jim Rutenberg, opening with a breathless fusillade of words about “a stubborn landscape of competitive states that right to the end are producing equal shares of hope and fear amid conflicting signals about the outcome.”
The intellectual disconnect is startling, to say the least. The mission of both articles is to inform readers of one of the world’s marquee newspapers about the state of the race. But their conclusions are diametrically opposed. And the fact that they’re sitting side by side tells me that journalism has a long way to go if it wants to catch up to baseball in terms of how to measure what matters.
The thing is, measuring what matters is much harder in baseball. Even a bright child understands that the popular vote has nothing to do with the Electoral College. To be fair, Zeleny and Rutenberg don’t mention the popular vote — they focus on the battleground states. But their story is almost entirely the stuff of the savvy: a narrative about the race so far, details of travel schedules and snippets from speeches, and windows into the hopes and fears of well-placed campaign insiders. It’s a compelling narrative, but one built almost entirely of qualification with precious little quantification — and in the end, Election Day is nothing but quantification.
Perhaps Silver has been more annoyed by his critics than he’s let on, because he closes today’s post with a devastating critique of the kind of journalism his colleagues are practicing: “If the state polls are right, then Mr. Obama will win the Electoral College. If you can’t acknowledge that after a day when Mr. Obama leads 19 out of 20 swing-state polls, then you should abandon the pretense that your goal is to inform rather than entertain the public.”