I read somewhere years ago that the problem with defensive statistics was twofold: first that unless you have a system that quantifies every possible play--a Platonic system if you will--it was skewed by the player's ability; if a player couldn't make an attempt, it wouldn't have an effect. If a player made an attempt and failed, it would have a negative effect. Second that even with a real difference in ability, the number of plays in which that ability would be meaningful--the opportunity margin--was so small that it was impossible to quantify in a meaningful way. Which also raised the question of how truly meaningful it might be--not of course because it couldn't be quantified but because it had less effect on the outcome of games than other aspects of the game.
The ultimate test of individual defensive value, in my mind, was the Jeremy Giambi experiment in MoneyBall. Beane thought that no matter how bad Giambi was defensively, he could play left, and his offense would make up for it. If I remember the book right, it was an utter failure. Giambi may have stopped hitting, I can't remember, but he was also an utter defensive failure in left. So there's some point that individual defense is so bad that nothing makes up for it.
But for the most part it seems to me the difference in successful marginal opportunities for the best major league player and the average major league player is probably very small, and on an individual basis probably doesn't affect the game very often. If you told me that Adam Everett was making 5 plays more than Jeter per game, I'd be stunned. If you told me he was making 1 more play every 20 games I wouldn't be in the least surprised. What I'd also expect is that the result of those plays for the most part would be more singles allowed by Jeter than by Everett, and rarely Jeter would allow a run that Everett wouldn't allow.
And that's the sort of thing you could quantify, but what struck me last night was that even though those runs allowed could be quantified, it wasn't really what was important. Unlike the other parts of baseball, pitching or batting, defense is a team sport. If you were going to measure defense, whether by words or numbers, you had to think about the result of the team's collective ability to make plays. You don't build defense at shortstop, you build defense up the middle. Or throughout the diamond. Individual marginal plays aren't particularly important on an individual basis, but collective marginal plays are.
I've probably read that somewhere, probably here, but it had just never struck me as strongly. If I were the Yankees, would I trade Jeter for Everett? Are you nuts? But I'd surely want a better defensive team than I've got.
As an aside, the Astros are still telling us we're likely to see Burke at short a bit:
The Link