When it comes to evaluating individual defense in baseball, perhaps no player has caused more controversy than Derek Jeter. The analytics – the data driven methods that attempt to measure how many balls in play a defender turns into outs – paint Jeter in a notoriously poor light. But the close-to-the-field baseball community – Jeter’s peers, the award-voting media, coaches and so on – have credited Jeter as a great defender over the years, as evidenced by his five gold gloves. Everyone agrees that the various performance-based approaches to defense, including Ultimate Zone Rating (UZR), Fielding Runs Above Average (FRAA), Total Zone and so on are flawed. They each miss important pieces of the defense puzzle. That’s why most proponents of those methods feel confident saying a player has a good or bad glove only if the systems tend to point in the same direction.
In Jeter’s case, they emphatically do. Many Yankee fans know instinctively that Jeter does not field his position well. Yes, his jump throw in the hole is an impressive display of athleticism. The “flip” in the 2001 playoffs against the A’s was one for the ages. And he plays the game with a cool and swagger that just makes him look good doing what he does. But “past-the-diving-Jeter” has long been one of the most well-worn phrases in the Yankee announcer lexicon. And it reflects the evidence suggesting that, whether due to a lack of proper positioning, late reaction time, or poor lateral movement, Jeter does an especially poor job of keeping balls on the infield that other shortstops get to.
Ben Lindbergh has an engrossing piece on Grantland about The Capn’s defense (including lots of neato GIFs). He combines an analytical approach with a fine-grained visual analysis of Jeter’s best and worst defensive plays to assess his approach to the position in comparison with other shortstops (including the legendary-in-sabermetric-circles Adam Everett).
Lindbergh sums up the evidence:
“Jeter gets outs on an above-average percentage of the balls he gets to, which helps obscure the fact that he gets to so few. It’s telling that errors are recorded only when a fielder has mishandled the ball, even though not even getting close to it might be the greater crime. We’re more likely to remember a fielder’s sins of commission than we are his sins of omission.”
At his Hardball Talk blog, Craig Calceterra notes that the approach Lindbergh describes approach is too labor intensive to be broadly useful, but is nevertheless illuminating: “None of this is particularly helpful from a data analysis perspective — it would take an awful long time to write up these sorts of defense stories for everyone and you still don’t have data sets you can compare — but it is pretty spiffy and probably tells us more than anything else on the matter can.”
Lindbergh concludes with an interesting story – that after the 2007 season, Yankees’ GM Brian Cashman went to dinner with Jeter and politely told him that his defense was really hurting the team. Jeter actually went to work that off-season improving his lateral movement and changing his approach to positioning. In 2008 and 2009, he rated as an above-average defender by the same metrics that had been killing his performance for years. But by 2010 (Jeter turned 36 that June), he was getting very old, especially by the standards of the position. And his defense has been regressing since. In other words, Jeter learned how to really play the position too late in his career to cash in on what he’d learned (defensively speaking, obviously).
In the great debate between the “eye test” approach to evaluating athletes and the data-driven approaches, one of the funny inversions in sports discourse, at least to my mind, is how the latter are usually viewed as arrogant, pointy-headed know-it-alls who have the gall to think their newfangled methods are superior to the experienced judgments of those who’ve played and been around the game forever. This was the tension at the heart of Moneyball. What’s funny about this is that those who rely to a significant degree on data to make judgments are saying that they don’t trust their own eyes. Why? Because they know that humans, themselves included, have an endless capacity for bias and self-deception. Analytic approaches don’t solve that problem. But at least they acknowledge it, providing a check on our natural inclination – as the ancient adage goes – not to “see things as they are, but as we are.”
Yes, the system Lindbergh describes is a visually based one, but it’s subject to the rigors of statistical analysis, boiling down plays to component parts that make it harder for the viewer to merely read into the evidence what he or she wants. In that regard, it avoids the certitude of too many defenders of Jeter’s defense who just know – because their own senses are all they need to consult – that contrary evidence is simply wrong.