On The Defensive

It seems we've had an awful lot of defensive talk of late. Yesterday we looked at the relative scooping skills of Jason Giambi and Mark Teixeira. Derek Jeter's defensive renaissance this year has been well-documented to the point that even if he were to go in the tank with leather for the rest of the season the overwhelming public perception would be that he had a great year with the glove.

Given all the defensive talk, I suppose it's appropriate that when I took a little mental break at work yesterday at hit the "random page" link on b-r.com I wound up on the page summarizing the 1935 American League defensive statistics. The Yankees gave up the fewest runs per game that year (4.24) in part due to finishing third in fielding percentage and first in defensive efficiency. It wasn't enough to overcome Detroit's herculean offense, and the Yanks finished 3 games back.

As I've stated, I'm not entirely sold on defensive metrics yet for a number of reasons. They can be a bit misleading as well - witness the 2009 Seattle Mariners: last in the AL in fielding percentage, but first in defensive efficiency. What's going on here? Is Safeco surpressing HRs that much? Do they have that many fly ball pitchers on staff that a disproportionately large percentage of batted balls are being turned into outs? I'm not sure.

Back to 1935 though. Continuing my little mental break, I sorted the list of fielders by range factor per game (A + PO /Innings * 9). Not surprisingly, first basemen and catchers topped the list. What did surprise me is that the next most prominent position on the list was not the hallowed shortstop position, but second base, as eight of the top eleven middle infielders played second base as their primary position.

This seems very odd to me. Shortstop has long been considered the most important defensive position in the infield. So why then would the seemingly inferior second base position appear to have greater range?

Wondering if this was an anomaly specific to that year, or that era, later on I looked at 2009 statistics. This year, AL second basemen have a range factor of 4.4555 while AL shortstops are at 4.3928, a difference of more than 9% (calculated with the raw PO and A values). In a single game, the average second baseman would be involved in 0.07 additional fielding plays, but through play Monday, second basemen had participated in 591 more fielding plays over the course of the season.

So why is this? I thought it might be a result of a double play bias, with perhaps the second baseman being the pivot on a disproportionate quantity of double plays, thereby doubling up on his range factor. This may have something to do with it, as second basemen have a 587 to 415 edge in times as a doubleplay relay man. Without considering 6-6-3 and the rare 4-4-3 double plays, this would appear to account for only 172 of the 591 plays.

What of the rest? Well given the predominance of right handed hitters, it could be that second basemen are covering the bag more on stealing attempts, padding their PO numbers. American League catchers have thrown out 307 runners at second base this year. So even if second basemen recorded every single one of those putouts, that and the double plays would still only cover 479 of the 591 plays.

What about outfield relays? AL second basemen have 70 relay assists to the shortstops' 58. That still doesn't cover it.

I'm really perplexed on this one. Granted, range factor isn't as evolved as other defensive metrics. But why does it appear here that second basemen are more involved defensively than their keystone counterparts? Is is the shorter throw to first leading to fewer errors? Is it that shortstops make more errors (184 to 109 through Monday) thereby reducing their range factor? Any ideas Fackers?