If the Oakland Athletics are the MLB team most often recognized for using analytics, the Philadelphia Phillies have been the club most ruthlessly mocked for ignoring them. When the Phillies won, they scouted well. When they lost, stats weren’t the escape.
By the time they added a lead analyst in 2013, they were years behind clubs like the A’s and Rays. Even then, they weren’t actually paying their stats guy, who was employed by MLB and wasn’t exactly an inner-circle decision maker in Philadelphia.
Two seasons later, a lot has changed. The Phillies have a young, open-minded GM in Matt Klentak, whose recent hires include a 30-year-old former Google employee. They now seriously value analytics, using them to inform player moves. These are not token gestures.
The Phillies’ rapid conversion stands out, but their recent hires mirror the industry as a whole. Five or 10 years ago, teams might have employed one analyst to keep up with and apply the latest sabermetric research. Now? Teams are expanding departments to the extent that they have the potential to outpace public research. If this keeps up, we might not even know the next big stat until years after MLB teams do. An MLB source says there may now be three or four times as many data-driven jobs as there were two years ago. Clubs have built out their analytics departments rapidly, growing from one or two people to six or eight. The days when Bill James readers were years ahead of clubs are gone.
Download it now: iOS | Android | Windows | Special Offer
Teams haven’t left us all behind just yet, though. Even with legions of new hires now populating MLB front offices, there still isn’t a huge gap between freely available research and proprietary findings. The public sphere has long enjoyed advantages over teams: peer review and sheer volume, to name a couple. Two of the biggest advances of the past 15 years—defence-independent pitching stats and catcher framing metrics—were developed online. As long as front offices and their analytics departments remained small, teams were at a disadvantage. Only recently have clubs been able to gain much ground.
“My feeling is that maybe 50 or 60 percent of what teams are doing privately gets reported,” Atlanta Braves GM John Coppolella says. “It’s usually a function of the attrition that exists within front offices, but for the most part, teams like to keep things proprietary, and for good reason.”
One statistically savvy exec says public metrics still compare favourably with most teams’ internal tools. Internal projection systems may be slightly more accurate than what’s out there on sites like Baseball Prospectus and FanGraphs, but the difference would rarely be enough to affect a decision.
But that could change within a few years. Many expect the gap to grow as teams continue expanding and hiring aggressively. While funding analytics departments was once a challenge for Billy Beane and like-minded GMs, owners are now buying in like never before, hiring young GMs like the Phillies’ Klentak, the Brewers’ David Stearns and the Dodgers’ Farhan Zaidi. Just as importantly, they’re giving these execs resources their predecessors couldn’t access.
“They’ve decided to throw money at nerds,” says Dave Cameron the managing editor of FanGraphs, a leading baseball-stats site.
He knows better than most, having lost numerous writers to teams—including Kiley McDaniel, who now scouts for Coppolella’s Braves. “The guys who bought in five or 10 years ago have been given the pocketbooks,” Cameron adds.
FanGraphs can’t hire from Google. The Phillies can—and did.
“They can say, ‘Hey, let’s steal this guy and pay him six figures,'” Cameron says. Eventually, those hires will make a real difference, especially as the job description becomes more complex.
Precise technical backgrounds will often be required of those poring over data from Statcast, the league’s new tracking technology. As Cameron notes, “This isn’t people messing around in Excel anymore.”
There’s more data than ever. That stacks the odds in favour of teams, not only because they have potentially exclusive access to it, but because they can afford top data scientists—even when that means plucking them from places like FanGraphs. There’s bound to be a brain drain.
That said, information flows both ways. MLB executives are extremely protective of their research–it’s baseball’s equivalent of a new iPhone, after all-—but eventually gossip trickles out. Plus, some say it’s increasingly difficult to make meaningful advances.
Of course, that’s what everyone assumes before the next big thing emerges. If teams could use Statcast data to improve player development, the possibilities for impact would stretch far. For example, identifying and acquiring a batter who’s a simple mechanical tweak away from boosting his line-drive rate would pay huge dividends. Eventually, the public would find out, but it now seems possible to envision years of lag time.
And who knows, maybe the Phillies will be the ones setting the pace.