Trading Defense for Offense

This past offseason, the Giants have made a couple of decidedly mediocre maneuvers to try and improve one of the worst offenses in the major leagues. Success came in 2009 on the backs of great pitching and a solid defense and despite ranking at or near the bottom in most offensive categories. The additions of Aubrey Huff over Travis Ishikawa, Mark DeRosa over Randy Winn, and Freddy Sanchez over Emmanuel Burriss are all, at the very least, lateral moves with upside. However, with the offensive upgrade, the Giants have sacrificed some of the key components of a defense that posted a team UZR of 51.2 in 2009, good for fourth in the major leagues. Combining this excellent defense with one of the deepest and best pitching staffs and a little luck led to a significant improvement over 2008, and one might wonder if the defensive downgrade will cost as much as the offensive upgrade returns.

The Opening Day lineup last year included Ishikawa and Winn, two players who contributed very little offensively and quite a bit defensively. In their stead are Huff and DeRosa, almost assuredly offensive upgrades and defensive question marks. While DeRosa’s outfield ratings are good, they do have small sample size issues attached; he’s only spent 412 innings in left field, presumably his primary position in 2010, and about 1600 innings total in the outfield (spread over nine seasons, which doesn’t help the credibility factor), about half of what’s recommended to be able to form an accurate assessment of a player’s defensive ability. He’s on more stable footing at second (2236 innings) and third (2408 innings), but is below average at both positions. Shortstop and first base suffer from an extremely limited sample size, but again DeRosa is below average at both positions. While possibly a valuable defender in the outfield, DeRosa is a liability in the infield, and he does stand to get multiple starts there with the glut of outfielders the Giants have and his positional versatility.

Aubrey Huff, meanwhile, will almost certainly not perform as well as Travis Ishikawa with the glove, considering he has been a DH for much of his career. But he’s a good bet to be at least as good for six times the price (copyright Brian Sabean) offensively. On paper, however, it looks like the runs Winn and Ishikawa stole on the field and gave back at the plate will now be the same runs DeRosa and Huff steal with the bat and give back with the glove. So where’s the upgrade?

Here’s where we get subjective, and ironically enough, subjectivity is the problem. I simply don’t think that any defensive metric is accurate enough to represent the true value of a fielder. UZR, fielding percentage, some third metric, none of them are devoid of observer bias. Certainly UZR is the best stat we have for measuring a player’s defensive skill, but even that has a five-run margin for error (and certainly the potential for more).

Perhaps it’s because I don’t understand UZR all that well, but there’s a big part of me that doesn’t trust it. The subjectivity of the metric (and subjectivity is why I’m not even going to talk about fielding percentage, because that’s just ridiculous) compared to stats like batting average, on-base percentage, and slugging percentage on the hitting side just make me skeptical. If I see a guy like Nick Johnson, I know he’s going to get on base 40% of the time. Ichiro is going to get a hit one-third of the time. These are rock-solid facts that I can count on (if they still hold water once luck is accounted for) because there is no room for interpretation: the hitter either got a hit or he didn’t. I don’t have to temper every stat citation with a disclaimer that says “Stat may be unreliable and thus relatively meaningless”. I can look at Freddy Sanchez’ ridiculous +19 2006 and say that he had an equally ridiculous .364 BABIP (40 points above his career average) and 27.5% LD% (3% above his career average) so I wouldn’t expect him to repeat that performance, and I can say the opposite about his -17.2 2008. But when he posts a +14.2 mark defensively (2005) in his first full season, I don’t really know where that came from, and I don’t know if I should expect it to continue. His subsequent seasons of 6.3, 9.2, -1.8, and 7.4 seem so wildly erratic, it’s hard to have faith in UZR; I understand why you might have an up or down year if your BABIP is fluctuating, but shouldn’t you be more or less the same defensively year-to-year? Certainly a general decline is expected, but looking at Sanchez’ numbers, I’m not sure it’s possible to predict a player defensively unless they’ve posted multiple consecutive seasons in the same general territory.

And that right there is why I think the Giants’ decision to trade some defense for some offense is a good one: offense can be predictable, while defense really can’t, and offense can be evaluated with a fair degree of certainty, and I don’t think the same can be said of defense. Maybe Huff is a +10 defender this year, Renteria goes +5, and Sanchez and DeRosa are both in negative territory. Combined with the strikeout nature of the staff (tops in the majors in K’s last year), I’ll trade unreliable defense for reliable offense any day. Of course, it’s arguable whether Huff, etc. provide reliable offense, but the overall point seems valid.