This week's U.S. election had former baseball analyst Nate Silver and his stunningly-good election predictions at the centre, suggesting that we may be on the verge of a new era of advanced statistics in politics. Silver correctly predicted the outcome in what looks to be a perfect 50 for 50 states on Election Day in 2012 after going 49 for 50 in 2008.
It's not just the analysts using them, either, as Dan Levy makes a solid case that acquiring and using advanced statistical information on how to reach voters in specific areas and demographics was crucial to Barack Obama's re-election. What's remarkable, though, is that while the anti-stats backlash in politics proved reminiscent of the wars from years ago in baseball, Silver and the others using advanced statistical models for politics have already made more progress on one key front than the sports statistics community.
What was really unique here was that Silver shattered the TV ceiling, appearing on The Daily Show and The Colbert Report and having his statistical predictions discussed to some degree on just about all channels' election coverage. While advanced statistics people in sports have made huge progress over the last decade, gaining substantial presences on the web, in print and even as radio guests, it's still extremely rare to see them (or even their stats!) featured on sports television coverage. Instead, we're stuck with the same old "personalities" yelling at each other on First Take and the bizarrely uniform lineups of ex-players and ex-coaches on pre-game and studio shows. It's a world where few *new* voices are heard and featuring as mainstream an internet writer as Bill Simmons is progress.
Furthermore, a statistical model that predicts outcomes to perfection or near perfection takes away from the "embrace debate" mentality that has gripped sports. It's much less dramatic to have one person tell you what's going to happen based on numbers and theories than it is for two people to argue and debate a certain topic. That's the drama of the horserace and the state of punditry today, manufactured or not. Silver famously quipped to Stephen Colbert, "“I’m not very pro-pundit, I have to say. “If pundits were on the ballot against, like, I don’t know, Ebola, I might vote Ebola, or third party.”
Politics followed sports in the rise of advanced statistics, but will sports follow politics in giving television prominence to analysts and their statistics?
If you go by ESPN's moves over the last few years, the answer is "No."
One of the most fascinating parts of Blogs With Balls 5 this year was the statistics panel's discussion of "Numbers Never Lie." At one point, that was intended as an advanced stats show (fantasy-oriented, but still willing to use advanced stats). Football Outsiders' Aaron Schatz, a well-respected advanced stats guru in the football realm, was one of the original people involved. As current Numbers Never Lie host Michael Smith said at Blogs With Balls, “It now is a debate show, like most other shows on ESPN. … I hate to say it’s not about analytics, but it’s not about analytics.” That's clear from some of the segments they've done on Alex Smith and Tim Tebow, using carefully-selected numbers to draw provocative conclusions that will let them embrace debate. What was most interesting was how Smith said the move wasn't thanks to ESPN higher-ups' own desires, but rather because research suggested their viewers didn't want to see stats that had to be explained. "For that reason, Numbers Never Lie got away from the hard analytics," he said. Based on that, there wouldn't be a lot of incentive for sports television broadcasters to feature advanced stats people more prominently. After all, the Worldwide Leader tried it, and it didn't work.
Tastes change, though, and sometimes coverage isn't just providing what people have already determined they want to see, but what you think they might like to see. It's highly doubtful that a poll of people planning to watch Election Night coverage would have demanded discussion of advanced statistical approaches beforehand, but the statistics proved to be such a story (thanks to their direct opposition of the many pundits who either said Mitt Romney would win or figured the election would be a toss-up) that networks discussed them anyways. Did that alienate viewers? Sure, some probably weren't in favour of it, but Silver's approach has obviously attracted a lot of positive interest as well; as of Friday, his recently-released book, The Signal and the Noise, was the No. 2 best-seller on Amazon.com. It would seem a lot of people were intrigued by Silver's ideas and the discussion of them, even if they hadn't necessarily been that interested in statistics beforehand.
That may present a way forwards for advanced statistics in sports television. What if sports television took a cue from political television, though, and rather than making advanced stats its own corner, integrated it into the regular coverage? What if you gave Aaron Schatz a seat (or even a two-minute segment! gasp!) on one of the broadcast networks' NFL pre-game or halftime shows, or gave John Hollinger or Tom Haberstroh a larger presence during an ESPN NBA telecast, or featured someone like Fangraphs' managing editor Dave Cameron on a baseball pre-game show? Why not offer a new voice that wouldn't have to necessarily destroy sports punditry, but add a new element?
Advanced political statistics of the kind Silver does were well back of advanced sports stats in recognition before this election, but now they've gone mainstream. The TV breakthrough was a huge part of that, bringing stats to an audience that wasn't necessarily demanding them. By contrast, advanced sports statistics have become dominant in much of the web and print-based baseball coverage, and they're used heavily in everything from basketball to hockey to football, but they haven't yet really broken through that TV ceiling that is the next step forward.
Some revolutions need to be televised.