Add new comment

Yeah I know, I end up putting another website on the spot, but good night....enough about the defense is important schtick. We all get it.
I wish I could remember where I read that it took roughly 3 years worth of data to come up with reliable offensive sabermetrics and that it would take roughly 3 times as long to come with a reliable count of defensive metrics to come up with a statistically reliable sample.
We still don't even use useful data such as trajectory and speed of a batted ball, positioning, and the relative values of the guy next to us. Sorry, but saying that because Wlad has a "-5" rating means Chavez should have been in there sounds like junk science to me. When a player's UZR jumps all over the place from year to year, I'm more liable to watch the guy than trust the numbers.
If it's going to take 9 years to get a reliable sample, I'll trust my eyes that Wlad has reasonable speed and a reasonable arm, and that the leverage of his error was a function of its timing and context in the game. Who's to say that however many fewer times Chavez makes an error in his career that it would not happen at that time? You can't eliminate chance from the game.
At any rate, even if the 3-yr, 3x thing isn't accurate, UZR and its associated dogma still leaves a lot to be desired IMHO.