Skip to content, or skip to search.

Skip to content, or skip to search.

Databases Loaded


Runs saved isn’t the only defensive metric out there: Ultimate Zone Rating, put together by a sabermetrician and former Cardinals consultant named Mitchel Lichtman, is more widely used than Runs Saved because it’s publicly available on the FanGraphs site without a subscription. (You must pay a monthly fee to Bill James’s website to access Dewan’s numbers.) There are several more systems, but they’re all based on the same concept, of dividing the field into zones and comparing each fielder to his peers; they only occasionally disagree on a player (Dewan is a slightly bigger fan of Mets outfielder Jeff Francoeur than Lichtman is). If you want, you can go to the message board of Lichtman’s website to see him and Dewan arguing over, for example, the ways their final rankings make sure not to punish outfielders who played in particularly large outfields. In the end, as Baseball Prospectus co-founder Clay Davenport says, “everyone’s pursuing variations of the same idea,” and the competition between metrics leads to greater accuracy for all.

That’s a good thing, since the field is still young and subject to volatility. Consider the example of Derek Jeter. For the first part of his career, Jeter was regarded as one of the best fielders in the game, but the reputation was mostly based on the anecdotal observations of sportswriters, broadcasters, and managers impressed by his jump throws from the hole at shortstop and his famous “flip play” against the A’s in the 2001 American League Division Series. (He won three Gold Gloves in a row, starting in 2004.) But defensive metrics like Dewan’s showed him to be not only an overrated defender but in fact a wretched one. His Runs Saved number was -28 in 2005, -18 in 2006, and -23 in 2007, making him one of the worst shortstops in baseball all three years. (As far back as 2001, Baseball Prospectus famously described Jeter’s defense as “yecch.”) But in the last two years, he has improved dramatically in Dewan’s estimation, a highly unusual development for a player in his mid-thirties. His Runs Saved number was -9 in 2008 and actually +2 last season.

Did Jeter get better, or did the numbers become more accurate? Probably both, but it’s impossible to be too precise; there are fielding-statistics innovations yet to be made. (For example, none of the leading fielding metrics measures the defensive abilities of catchers, who are the most important defensive players on the field.) The reason this data is so rare and so valued is that it’s extremely labor-intensive: Not everyone—not even most teams—has a staff of people available to pore over 2,430 baseball games. And even if they do, there are still a lot of human judgments—see Dewan’s “slow” line drives—in between what happens on the field and what comes out in the numbers. The “zones” concept is more advanced than any kind of fielding evaluation that had been done before, but it’s still a relatively crude way to describe what happens on a two-acre baseball field to balls hit with infinitely varying levels of speed and spin and trajectory.

To that end, analysts and observers are eagerly awaiting the data from MLB Advanced Media’s Hitf/x system, which tracks the actual trajectory of the ball when it comes off the bat, in a similar way that its Pitchf/x, introduced a few years ago as part of its MLB Gameday online function, tracks the curve, break, and velocity of pitches. The two systems require special cameras installed at every ballpark, and Hitf/x is not online yet. “Once that information is out there,” Davenport says, “there will be a rush to get the best analysis, with lots of claims and counterclaims that will ultimately be sorted out. How people extract the data will be an ongoing thing, but we know it will bring us closer to making defense as objectively quantifiable as offense. Someday, the notion of watching these plays on video may seem archaic.” Fielding geeks will no longer have to manually figure out how many balls were hit on vector 187 and make a judgment about the left-fielder’s play; they’ll be able to query a database of balls hit at a 30-degree angle to the third-base line, with between 40 and 42 degrees trajectory upward and 90 and 100 miles per hour speed off the bat, and see exactly how many were turned into outs by which players.

By the time that happens, other teams will have caught up with the Yanks’ and Sox’ analysis—Dewan estimates that only ten or fewer teams do serious research into advanced defensive metrics—and Ultimate Zone Rating and Runs Saved will go the same way OBP and OPS went, to the mainstream, where everyone knows and accepts them. And then there will inevitably be something else to study: Dave Cameron of FanGraphs suggests that older players like Damon and Jim Edmonds are undervalued (advancements in kinesiology and physical therapy have changed the way players age in ways that haven’t been quantified yet), and some think that speed and base-stealing have gone too far out of style (the last decade of mammoth home-run totals made one extra base seem almost piddly; with home-run totals dropping—perhaps because of steroid testing—baserunning becomes relatively more important again). “As much as it might seem, sabermetrics is not rocket science,” Lichtman says. “In the world of science and applied math, it is frighteningly simple, comparatively. Defensive metrics are a little more complicated, but not much more. They’ll get it figured out, and they’ll move on to the next thing.”


Current Issue
Subscribe to New York

Give a Gift