Is Nate Silver destroying our ability to argue?
When I was a kid, my parents would have what I would call “encyclopedia” arguments. Say, did FDR die on April 12 or April 13? They would marshal personal history (“My cousin’s birthday was April 12”), dubious logic (“If I remember our anniversary, how would I forget a date like Roosevelt’s yahrtzeit?”), and ad hominem attacks (“Oh, you can never admit when you’re wrong….”).
This could go on for an entire evening, while I was thinking, “Just look it up in the World Book, for God’s sake.”
Of course, the point of some arguments is not being right, but making your case no matter how ludicrous or counter-factual. A lot of climate change skeptics and gun rights folks have to know that they are defending the indefensible, and yet the challenge is to create a convincing argument out of gossamer — what Colbert calls “truthiness.” I don’t think my parents cared about nailing down the date of Roosevelt’s death, but enjoyed the argument for its own sake.
And of course, we were all too lazy to actually walk to the shelf, pull out volume “Q-R,” and check the date.
Smartphones and tablets have changed that. Who played Officer Levitt on Barney Miller? The iPhone tells me it’s Ron Carey. Where did Carmelo Anthony go to college? A Google search on Galaxy coughs up Syracuse.
I love this speedy fact-checking, although I used to enjoy getting there without Siri’s help. “You remember, he played the chauffeur in Mel Brooks’ Hitchcock movie. What was that called? Right, High Anxiety. He was the ‘Got it, got it, I don’t got it’ guy. Tod something, or Ron. It was Ron. Ron…Carey!!”
So in that sense, modern technology is only a time saver, and a sort of joyless one at that.
But where Google and Wikipedia change our relationship with facts, Silver transforms our relationship with opinion. Silver started out crunching baseball statistics, and then rose to prominence by correctly predicting, on his FiveThirtyEight.com blog, the outcomes of the last two presidential elections. But he didn’t just pick the winner; day by day he analyzed dozens of individual polls, separating the useful data from the noise. Along the way, he subverted the role of media pundits. Columnists who relied on their gut instincts and experience now found themselves confounded by hard data.
As a result, pundits and anyone who likes a good argument began to resent Silver the way liberal arts majors resent engineers. Punditry and barroom banter is about the poetry of the campaign trail, the impressionism of the locker room, the word on Main Street. By comparison, “data journalism” seemed bloodless, unsentimental, and deterministic.
Once a reporter could cover a rally and note that Candidate X “sounds like a winner” or that Candidate Y is “losing momentum.” Silver’s blog, meanwhile, would weigh the statistical evidence to show that neither conclusion was correct. You saw the Nate Silver Effect on election night, in the now famous exchange between Karl Rove and Fox colleague Megyn Kelly. When Rove refuses to accept the results being reported by Fox’s own team of Silverian number-crunchers, Kelly asks him, “Is this just math you do as a Republican to make yourself feel better?”
Silver has gone on to apply his methodology to subjects as disparate as choosing a pope, maximizing your returns at the salad bar, and especially picking winning sports teams.
Old-school reporters liked Florida Gulf Coast University’s chances in the NCAA tournament because the team had “moxie” and “grit” and because everyone loves a good story, in this case the one about Cinderella.
Silver, meanwhile, analyzed all 512 tournament games since 2003, using a model he describes in a 3,000-word blog post. It factors in multiple computer and human rankings, injuries, geography, and “power ratings” — and calculated FGCU’s chances of winning it all at 0.019 percent, or 1 in 5,180.
Lacking Silver’s mathematical training and statistical creativity, I’ve found myself suddenly going quiet in the middle of an argument or talking through a column. I have opinions — on the most efficient way to board an airplane, on what kinds of messages are effective in advocating for Israel or increasing synagogue attendance. I like to defend meteorologists against those who turn on them when they get it wrong. But then I remember Silver and his fellow data heads, and suspect that they could create a model that would trump my hunches with cold, hard facts.
As it turns out, Silver is also a fan of meteorologists — and not just because weather forecasts are “much better than they were 10 or 20 years ago.” Silver worries that many experts rely too heavily on poorly designed computer models or, conversely, value their own subjective judgment much too highly. Modern meteorologists, he blogs, strike a “healthy balance between computer modeling and human judgment.”
I have some of the latter, and can do none of the former. Luckily the data journalists are leaving a trail, easily accessible by smartphone or tablet, of good, useful analysis that even an English major could understand.