In the never-ending battle between advanced defensive metrics and the eye test, there's at least one point everyone can agree upon: Tampa Bay's Kevin Kiermaier is indisputably an elite defender, whether you go by his record-setting DRS total or the multiple defensive awards he won as a Minor Leaguer. By the old
In the never-ending battle between advanced defensive metrics and the eye test, there's at least one point everyone can agree upon: Tampa Bay's Kevin Kiermaier is indisputably an elite defender, whether you go by his record-setting DRS total or the multiple defensive awards he won as a Minor Leaguer. By the old methods or the new, everyone agrees that Kiermaier is a stud on defense.
On the other hand, Detroit's Anthony Gose didn't fare nearly as well with the numbers in 2015. Among 25 center fielders who played at least 1,000 innings last year, Gose was 24th in UZR and tied for 22nd in DRS, and predicably, he wasn't very happy about that, making some noise earlier this spring when he called analytics "a scam," going on to say that he considers himself among the better fielders in baseball.
Whether or not you agree with Gose -- and surely we're long beyond the point of thinking that analytics aren't heavily used in every front office -- he brought up something extremely interesting about his positioning in his comments:
"I play shallow. I know that," he said. "I don't know if that has anything to do with it, but I feel like I'm one of the better ones."
Comerica Park's center field is 420 feet deep, the second deepest in baseball. If it's true that Gose really did play shallow, then in theory, he'd have a ton of ground to cover behind him, which may leave him more susceptible to allowing damaging extra-base hits to fall in, which in turn would impact his defensive ranking.
For years, "in theory" is all we had to go on other than a player's self-evaluations when it came to positioning. The current advanced metrics, while miles better than anything that came before, don't do a great job of accounting for positioning. We've never really been able to measure outfield positioning in any reliable way, other than tossed-off quotes from players or coaches thinking they play deep or shallow.
Until now, that is. As we dig deeper into the data that Statcast™ provided in its first season, we're now able to to collect the start points of each fielder on every ball in play and combine them to find out where they started. By adding in controls for the varying outfield distances at each park, we can show who really is "shallow" or "deep."
Let's compare Gose's positioning to that of Kiermaier, who plays a notoriously deep center:
Gif: Kiermaier-Gose OF Positioning
That's a stunning difference. Kiermaier's average start position was 327 feet from the plate, but even though the center-field wall is 16 feet further away in Detroit, Gose stood 24 feet closer to home, an average of 303 feet away. That meant that he's got 117 feet of grass behind him, while Kiermaier had just 77 feet behind him. Even if they were equally talented, there's simply less ground for extra-base hits to get over Kiermaier's head.
We chose Kiermaier because of his reputation and Gose because of his statements, but when ranking all center fielders in terms of positioning, they're very nearly the deepest and shallowest, anyway.
Let's explain: We measured 36 center fielders who were on the field for at least 1,000 batted balls, and we compiled their average start positions, shown as distance in feet from home plate. (To keep things simple, we excluded players who called multiple parks home last year; we also didn't account for the various road park sizes in this very initial version, though that should be a relatively equal effect across players.)
Then, to ensure we accounted for fielders who play in very deep or shallow home parks, we compared that average distance to the center-field wall at their home stadiums in order to get a percentage that shows where they tend to stand on the line from home plate to the wall:
Gose played shallower in a home field-adjusted sense than any other center fielder other than Houston's Jake Marisnick. Since Rajai Davis was Gose's Detroit teammate last year, it seems like it was a team mandate, and indeed, manager Brad Ausmus has said this offseason that he would like his outfielders to play deeper, hoping to limit the extra-base hits -- after all, only four teams suffered more non-homer extra-base hits on balls hit more than 320 feet. It's tough to defend that big outfield playing shallow.
Positioning can't be used to explain all of Gose's issues, of course, nor is it the only reason that Kiermaier graded out so well. In Gose's case, the fact that center field is populated by a ton of great defenders like Kiermaier, Lorenzo Cain, Mike Trout and Kevin Pillar makes it more difficult to stand out, and even beyond that, there's evidence that Gose had merely a decent year, not a great one. But if he wasn't put in a good position to do anything about that, he was always starting from behind, perhaps not by his own doing.
Still, we'd long known that positioning was a missing piece from our defensive evaluations, there just wasn't much to be done about it. Now it seems that there is, and it's a good first step towards doing a better job in applying the new data towards real-world performance.
Mike Petriello is an analyst for MLB.com and host of the Statcast podcast. He has previously written for ESPN Insider and FanGraphs.