Smell the Glove
When you hit like Manny, you don't have to worry about defense. But some other free agent sluggers didn't even sniff the cash they wanted -- because a new wave of fielding analysis is changing the way players are valued.
You can almost picture Bobby Abreu on the phone, leaving message after message -- like Mikey in Swingers -- longing for a little love on the other end of the line.
I just wanted to leave my number. Didn't want you to think I was desperate or anything ...
The 35-year-old outfielder entered free agency with an impressive résumé: career .300 hitter with a .903 OPS in 11 full seasons; 100 or more RBIs seven of the past eight years; 20 or more stolen bases every season since 1999. Rumor had it that Abreu, who made $16 million with the Yankees last season, was hoping for $48 million over three years.
Nobody picked up.
Finally, on Feb. 12, just days before pitchers and catchers reported to spring training, Abreu signed a one-year deal with the Angels for $5 million. A day earlier, Adam Dunn, who had hit 40-plus home runs in five straight seasons and made $13 million with the Reds and Diamondbacks in 2008, settled for a two-year, $20 million offer from the Nationals. It was the same deal he'd previously ignored because he was certain something bigger and better would come along.
So what happened? Are we looking at further proof of the global economic crisis? Jackpot-seeking big league sluggers take a hit in their fat wallets. Story at 11. Nah, the real story is that those gaudy offensive numbers are only half the story.
Used to be we would talk about a player's being good or bad with the glove, but we wouldn't get too specific. These days, thanks to baseball-loving number crunchers who've been trying to crack the riddles of defense for the better part of 20 years, we have some very specific formulas that determine each player's defensive contribution. Or, in the case of Abreu and Dunn, their defensive damage.
By the calculations of John Dewan, author of The Fielding Bible Volume II, Abreu is one of the worst rightfielders in all of baseball. (Forget that Gold Glove he won in 2005. Those awards, voted on by managers and coaches, are often influenced by a player's defensive reputation and even his offensive performance.) Over the past three seasons, Abreu has made plays on 28 fewer balls than a league-average rightfielder would be expected to make, costing his teams (the Phillies and Yanks) 19 runs. Toronto's Alex Ríos, the highest ranked at his position, got to 26 more balls than the average rightfielder and saved his squad 49 runs (or nearly five wins) during that same span. Dunn, meanwhile, reached 31 fewer balls than an average leftfielder would and cost his teams (Cincinnati and, briefly, Arizona) 39 runs. The Cubs' Alfonso Soriano, surprisingly enough, has the highest rating among leftfielders over the past three years. He got to just three more balls than the average guy, but because of his strong arm, he saved his team 42 runs. The Rays' Carl Crawford, who got to 22 more balls than the average leftfielder, was second with 22 runs saved.
So when defense is factored in, Abreu and Dunn weren't elite free agents after all. Their offensive output is so significantly undercut by their defensive deficiencies that the market judged them as one-dimensional major leaguers.
Traditional scouting can tell you if a player has a good or a bad glove and generally whether one guy is better than another at fielding his position. "The difference with the advanced metrics is that you get a firm measure of how many runs better he is," says one AL team executive. "Or you get a real sense of what he's costing you on defense relative to what he does with his bat."
The Baseball Prospectus metric Wins Above Replacement Player, or WARP, takes into account both defense and offense. Last year Dunn had a WARP of 3.7, meaning he was worth that many wins over the rough equivalent of an average Triple-A player at his position. Compare that with the Royals' David DeJesus, a standout in leftfield but no slugger, who had a WARP of 5.2. "We're starting to see teams take a more complete picture of a player's value," says Mitchel Lichtman, a former statistical analyst for the Cardinals who created Ultimate Zone Rating (UZR), another measure of how many runs a player saves or costs his team relative to the positional average.
How widespread is the thinking? Well, not everyone is on board. Witness Philadelphia's old-school splurge on Raúl Ibañez (three years and $31.5 million), whose UZR score was –11.9 last year and –21.2 in 2007. But there's evidence that defensive metrics are influencing the thinking of many clubs, and it's not just the financial hits taken by Abreu and Dunn. This winter, the Mariners insisted on getting slick outfielder Franklin Gutiérrez from the Indians in the three-team trade involving J.J. Putz and the Mets. Seattle also acquired Endy Chávez -- you might remember his circus catch in Game 7 of the 2006 NLCS -- in the same deal. The Tigers, one of the worst defensive teams in baseball last year, took a flier on shortstop Adam Everett, who, before missing most of the past two seasons due to injury, was considered a god among the folks who build fielding metrics (34 runs saved in 2006). "The genie is out of the bottle," says one AL general manager. "There are quite a few teams seriously looking at defensive analysis, and more and more each year."
Although measures like UZR are available to seamheads everywhere, front office wonks are wary of going on record about their specific approaches and conclusions. No one wants to give up his team's formula, for reasons that go beyond efficient allocation of resources and smart negotiation of contracts. In a poststeroids era, when power is on the wane, clubs are grinding for every tactical edge. "Your focus is on the so-called little things," says one AL player development assistant. "Which, of course, aren't little things at all."
Everyone wants to be the next Rays, whose 2008 Cinderella season was dramatic testimony to what defense can do. Last spring, Baseball Prospectus projected that the Rays would win 88 to 90 games, based largely on their improved glove work. In 2007, Tampa Bay had one of the worst defenses in modern history; according to BP's calculations, the fielders cost the pitchers 117 runs compared with a league-average defense. But a few smart moves turned it all around: switching B.J. Upton from second base to centerfield; trading Delmon Young, who had been playing some center; and establishing Jason Bartlett, acquired in the Young deal, as the starting shortstop. Then, two weeks into the season, Evan Longoria came up from the minors and solidified third base. The collective result was one of the best defensive teams in baseball, saving 59 runs in the field. The net turnaround, from –117 to +59, accounted for almost 20 wins and was a huge factor behind the team's success. "It's all about run differential," says a National League assistant GM. "There's a competitive advantage to be gained if your analysis is right."
Offensive outcomes are relatively easy to measure. You know if a guy hits, and you know what that hit is worth. Defensive analysis is more complex. You need to measure not only what happened but what might have happened and what should have happened. You're dealing in probabilities as well as outcomes. Defensive assessments are more nuanced now, often representing a blend of traditional and innovative methods. Derek Jeter's defensive value to the Yankees, for example, should be determined by more than his fielding percentage, fifth best among all shortstops in 2008, and his flare for dramatic plays. You also have to consider the number of balls he doesn't reach compared with the average major league shortstop. By that measure, according to The Fielding Bible, Jeter ranks dead last at his position over the past three seasons.
The rise in fielding metrics corresponds to a technology-driven explosion of information about batted balls in play. Companies like Baseball Info Solutions and STATS, Inc., track where a ball is hit, how hard it's hit, who fields it and how (or if) he converts it into an out. Stat guys like Lichtman and Dewan keep running tallies for every player at every position and compare each fielder against the mean. "The information has been collected for years, but the databases have become more granular, and they're much more widely available than before," Lichtman says. "It's a pretty big leap. We're in a period where the curve, in terms of the difference that advanced analysis can make, is getting steeper."
When Michael Lewis wrote Moneyball, six years ago, the recasting of the accepted offensive analysis set off a baseball holy war. But this time around, expect no such uproar over a new defensive analysis (aside from an anguished cry raised by the Bobby Abreu Society for Aging Outfielders). For starters, neither scouts nor statheads were satisfied with the old numbers. "There's not much you can learn from fielding percentage, errors and putouts," says an AL front office executive. "The guy with the most errors is not the worst fielder, and the guy with the highest fielding percentage is not the best." Also, fielding metrics tend to overlap with traditional scouting evaluations. Ask a scout and a stat analyst to identify baseball's best defensive third basemen over the past few years, and both will have Adrián Beltré and Pedro Feliz at the top of their lists. As Dave Studenmund, co-owner of The Hardball Times, puts it, "Unlike the argument for OBP, which really challenged conventional wisdom, these defensive ratings are often reinforcing what we thought was true."
There's still a lot to learn. Minor league defensive data are wildly inconsistent. Even major league data, depending on the formula used, yield varying results. Analysts would prefer to have a technologically precise record of a batted ball's hang time, speed and landing point. Statisticians are just now starting to evaluate the effect of misplays that aren't ruled errors and just beginning to track a fielder's route to the ball. One big league official fantasizes about embedding GPS chips in caps, and an Internet stat analyst imagines eye-in-the-sky camera views, similar to scouting video in college and pro football. "The overall attention to defense isn't new at all," says one NL exec. "But now we're very much in the process of building systems."
The game is once again evolving. Fielding analytics are becoming an integral part of how contracts are structured and how teams are built. Defense was long considered the undiscovered country of sabermetrics. As these measures become part of baseball's conventional wisdom, defense enters another province.
"In my mind," says one NL general manager, "it's the new frontier."
This story appears in the April 6 issue of ESPN The Magazine.