Monday, June 14, 2004
Being Defensive …
As I’ve said … oh, probably about thirty times … I blew my mind reading Michael Lewis’ description of Bill James 1977 Baseball Abstract and its attack on fielding percentage as a statistic on the grounds that it was subjective and failed to measure a player’s actual defensive ability in Moneyball (see, p. 66 of the paperback version). The idea that fielding percentage is both an inaccurate gauge of a player’s ability and is subjective is vital to understanding Moneyball and sabremetrics.
I was intrigued by the idea of quantifying a player’s defensive ability and expressed frustration at how subjective defense really is. An at-bat has an objective outcome: you get on base or you make an out. A pitch has an objective outcome: it is a strike, a ball or it is put in play positively or negatively. Defense is subjective: if something is an error or a successful play depends upon one’s point of view: how many times have you gone to baseball game and seen a close call where the third baseman grabs the ball, bobbles it for a split-second, and then throws to first and narrowly misses getting the runner. Error? Was he playing too deep? Was the ball too well-hit to play anyway?
We know good defense when we see it: Willie Mays catch in the 1954 World Series is the best defensive play I’ve seen. We know bad defense when we see it: Chuck Knobloch beaning people in the stands throwing to first base while playing for the Yankees comes to mind … sitting in my bar exam lecture Friday afternoon during the seventh and final excruciating hour of lecture I began thinking about the inability to quantify defensive stats when the lecturer, who was talking about U.S. Constitutional Law, began discussing obscenity (i.e. pornography). The lecturer, illustrating the gray areas of the law, quoted Justice Potter Stewart’s famous statement that he’d know pornography when he saw it. Lacking in clear guidelines as to what constitutes pornography, the lecturer intoned, the legal standard of obscenity is very subjective: Maxim could be considered obscene in Salt Lake City, but not in New York City. (Tellingly, Wal-Mart refuses to stock Maxim but they carry Tim LaHaye’s Left Behind series. My local Barnes and Noble here in the city of Pittsburgh carries Maxim, Stuff, FHM and all of those “laddie” magazine knock-offs, but jams their copies of “Left Behind” in the back of the store.) A good defensive player reminds me of … well … obscenity. I’ll know good defense when I see it. (Read the difference between Johnny Damon and Terrence Long playing centerfield on pages 135-136 on Moneyball.)
The best defensive stat is not fielding percentage (read Moneyball’s Chapter Five to find out how inaccurate and anachronistic a statistic fielding percentage really is), but ZR. (I’ll explain.) A recent look at the Phils ZR stats revealed an interesting bit of information to me.
Consider the differences between Phillies second basemen Placido Polanco and Chase Utley: the consensus I have read is that Polanco is the superior defensive second baseman and the better on-base man, whereas Utley hits for more power and needs to work on his defense and selectivity at the plate. Here is a startling thought: Utley is the better defensive second baseman.
Fielding Percentage: (as of June 10, 2004)
Polanco: .987
Utley: .968
Polanco has just two errors, while Utley has four. Polanco has turned more double plays (21 to 12), made more put-outs (67 to 54), and has more assists than Utley (81 to 68). But here is the kicker: Utley’s ZR is higher:
Zone Rating (ZR): (as of June 10, 2004)
Polanco: .795
Utley: .871
What in the heck is ZR? ZR is “Zone Rating”, a statistic developed by STATS, INC to attempt to objectively quantify defensive ability by measuring how much a player dealt with balls hit into his defensive “zone”. STATS, INC. divides the field into quadrants and then measures the balls entering the quadrant and how successful that player is in playing that ball. (I hasten now to confess to not being entirely certain what I just said is accurate: Steve from Mariners Wheelhouse recently told me that ZR’s formula makes it a counting stat. I am working off ESPN’s definition of ZR: “The percentage of balls fielded by a player in his typical defensive ‘zone,’ as measured by STATS, Inc.” If I have utterly misunderstood ZR I apologize for leading everyone astray.) Thus, it is strongly suggested that Utley is the better defensive second baseman.
Other applications of ZR:
-Tomas Perez has logged about 160 innings of work filling in around the infield and he has done well: his second base ZR is .889, etc.
-Bobby Abreu looks to be the Phils top outfielder:
Abreu: .884
Burrell: .849
Byrd: .832
Glanville: .809 (again: why?)
-Phils are ninth in ZR right now of 16 teams with .847 … the Fish are fourth in ZR. Unsurprisingly, Colorado is fifteenth (.834), but much to my surprise the Pittsburgh Pirates are dead last in ZR (.817), despite being a pretty decent defensive team.
-A good reason why the Padres are going to be tough in the post-season despite their lack of offensive firepower: they are third in ZR (.860) …
-The Phils were second in ZR in ’03. (.859 to the Giants .860 …)
Something to think about …
(Hey, if I screwed up the understanding of what exactly ZR is, someone tell me and don’t spare my feelings.)
I was intrigued by the idea of quantifying a player’s defensive ability and expressed frustration at how subjective defense really is. An at-bat has an objective outcome: you get on base or you make an out. A pitch has an objective outcome: it is a strike, a ball or it is put in play positively or negatively. Defense is subjective: if something is an error or a successful play depends upon one’s point of view: how many times have you gone to baseball game and seen a close call where the third baseman grabs the ball, bobbles it for a split-second, and then throws to first and narrowly misses getting the runner. Error? Was he playing too deep? Was the ball too well-hit to play anyway?
We know good defense when we see it: Willie Mays catch in the 1954 World Series is the best defensive play I’ve seen. We know bad defense when we see it: Chuck Knobloch beaning people in the stands throwing to first base while playing for the Yankees comes to mind … sitting in my bar exam lecture Friday afternoon during the seventh and final excruciating hour of lecture I began thinking about the inability to quantify defensive stats when the lecturer, who was talking about U.S. Constitutional Law, began discussing obscenity (i.e. pornography). The lecturer, illustrating the gray areas of the law, quoted Justice Potter Stewart’s famous statement that he’d know pornography when he saw it. Lacking in clear guidelines as to what constitutes pornography, the lecturer intoned, the legal standard of obscenity is very subjective: Maxim could be considered obscene in Salt Lake City, but not in New York City. (Tellingly, Wal-Mart refuses to stock Maxim but they carry Tim LaHaye’s Left Behind series. My local Barnes and Noble here in the city of Pittsburgh carries Maxim, Stuff, FHM and all of those “laddie” magazine knock-offs, but jams their copies of “Left Behind” in the back of the store.) A good defensive player reminds me of … well … obscenity. I’ll know good defense when I see it. (Read the difference between Johnny Damon and Terrence Long playing centerfield on pages 135-136 on Moneyball.)
The best defensive stat is not fielding percentage (read Moneyball’s Chapter Five to find out how inaccurate and anachronistic a statistic fielding percentage really is), but ZR. (I’ll explain.) A recent look at the Phils ZR stats revealed an interesting bit of information to me.
Consider the differences between Phillies second basemen Placido Polanco and Chase Utley: the consensus I have read is that Polanco is the superior defensive second baseman and the better on-base man, whereas Utley hits for more power and needs to work on his defense and selectivity at the plate. Here is a startling thought: Utley is the better defensive second baseman.
Fielding Percentage: (as of June 10, 2004)
Polanco: .987
Utley: .968
Polanco has just two errors, while Utley has four. Polanco has turned more double plays (21 to 12), made more put-outs (67 to 54), and has more assists than Utley (81 to 68). But here is the kicker: Utley’s ZR is higher:
Zone Rating (ZR): (as of June 10, 2004)
Polanco: .795
Utley: .871
What in the heck is ZR? ZR is “Zone Rating”, a statistic developed by STATS, INC to attempt to objectively quantify defensive ability by measuring how much a player dealt with balls hit into his defensive “zone”. STATS, INC. divides the field into quadrants and then measures the balls entering the quadrant and how successful that player is in playing that ball. (I hasten now to confess to not being entirely certain what I just said is accurate: Steve from Mariners Wheelhouse recently told me that ZR’s formula makes it a counting stat. I am working off ESPN’s definition of ZR: “The percentage of balls fielded by a player in his typical defensive ‘zone,’ as measured by STATS, Inc.” If I have utterly misunderstood ZR I apologize for leading everyone astray.) Thus, it is strongly suggested that Utley is the better defensive second baseman.
Other applications of ZR:
-Tomas Perez has logged about 160 innings of work filling in around the infield and he has done well: his second base ZR is .889, etc.
-Bobby Abreu looks to be the Phils top outfielder:
Abreu: .884
Burrell: .849
Byrd: .832
Glanville: .809 (again: why?)
-Phils are ninth in ZR right now of 16 teams with .847 … the Fish are fourth in ZR. Unsurprisingly, Colorado is fifteenth (.834), but much to my surprise the Pittsburgh Pirates are dead last in ZR (.817), despite being a pretty decent defensive team.
-A good reason why the Padres are going to be tough in the post-season despite their lack of offensive firepower: they are third in ZR (.860) …
-The Phils were second in ZR in ’03. (.859 to the Giants .860 …)
Something to think about …
(Hey, if I screwed up the understanding of what exactly ZR is, someone tell me and don’t spare my feelings.)
Comments:
Post a Comment
