A Bad Defense is a Good Defense?

Over the first couple months of the season, you've probably listened to the radio or TV and heard that the Indians have the worst or one of the worst defenses in the American League. The question is, what is the criteria for determining the "worst" defense? Turns that today's standard is simply the number of errors a team has committed, or sometimes people use team fielding percentage.

Personally, I think this measure is inadequate. This is like determining the best offense in the league by using only team batting average; you don't get the whole picture. In the case of defense, making the routine play is one facet of a good defense, but making plays that the average fielder won't make is a mostly untapped measure. One of my favorite team defensive measures is "Team Defensive Efficiency," which you can view at Baseball Prospectus. It is simply the percentage of balls in play that become outs; there's no scorekeeper bias at work here. It includes both "errors" and "hits." And lo and behold, the Indians are 4th in the American League in this measure at .7105 (the league average is .6974).

What does this mean? First of all, it means that errors don't tell the entire story; while of course you should make the routine play, making the "unroutine" plays should also be part of the defensive picture. Secondly, range does matter; to illustrate this point, imagine two shortstops. Shortstop A has a limited range, but makes the routine plays, while shortstop B has a very good range while making some errors.

Shortstop A:
2 Errors
3 "Unroutine" plays

Shortstop B:
10 Errors
15 "Unroutine" plays

Which is the more valuable shortstop? If you just used fielding percentages, it would be Shortstop A. But you aren't looking at the entire picture. Now this example is a very simple one, but extrapolate this to an entire team, and you'll see why judging a team's defense based solely on the number of errors it makes is a mistake. In 2004, the team finished second to last in the AL in defensive efficiency at .6855. The difference between this year and last is that the team is turning 5 more balls per hundred in play into outs than last season. That's a sizable difference, especially considering the margin of error the team is dealing with in order to win games.

By the way, am I the only one who enjoys listening to TV broacasts sans John Sanders? Nothing against Sanders, but I think Rick Manning and Mike Hegan make for a much more interesting viewing experience.

One more thing: CC Sabathia = Calvin Pickering + 95 mph fastball.