Defensive statistics

[wombat_socho]

If I had brains, I'd be home on sick leave, but since I'm feverish and my brain is simmering slowly in its own juices, I'm here at work mulling over this post by Chris Needham about the sucky defense of our Nats. It's an annoying post, because Chris rips on defensive stats for not providing clear proof of how inept Royce Clayton and Jose Vidro are, when right there in front of him the range factors and zone ratings do exactly that...sigh. Stats are what they are. They count the things that can be counted and are valuable to the extent that you understand the context.

First of all, bitching about the fact that range factors for outfielders playing behind flyball pitchers are going to be higher (on average) than those playing behind groundball pitchers or flamethrowers like Clemens or Johnson or Liriano is pointless, stupid, and throughly played. Range factors are designed to compare fielders on a league basis over a whole season, which is why they aren't really valid for defensive replacements and why the flyball/groundball/flamethrower caveat is a load of crap. Most teams don't have pitching staffs made up exclusively of one type or another, for one thing, and while there is a certain amount of park effect involved, over the course of a season it's just not that significant. Especially for infielders. Bill James explained all this until his typing fingers were sore twenty years ago, and it's appalling that people still don't get it.

Secondly, zone rating is not "range factor on steroids". As a former STATS reporter, that made my brain hurt. Range factors are purely statistical: you sit down at the end of the season, add up the putouts and assists for each player at each position, multiply the result by 9, and divide by the number of innings played at that position. Zone rating is based on actual observation of where the ball is caught and who catches it. It's actually more like defensive average, which measures the percentage of balls caught by a fielder within their zone of responsibility.

The bottom line is that you don't need one superstat to tell you that Clayton and Vidro stink on ice defensively. Needham serves up the stats, compares them to the stats of other fielders in the league, and shows pretty conclusively that C&V are as bad as he thinks they are...which explains why our pitchers, who nobody will confuse with Clemens, Santana, Johnson or Liriano, get smacked around on a regular basis. It's pretty hard to rely on your defense when it's that awful.

Comments

[(Anonymous)]

I geeked out on your stat geeking out. It sure is great to read someone geek out on basball who REALLY KNOWS WHAT HE'S TALKING ABOUT. You could put all the sportswriters in Syracuse to shame.

[wombat-socho.livejournal.com]

Thanks. This is what I used to do with the Rebel Baseball Review e-zine that covered the independent minor leagues - even after Baseball America started covering more than just the Northern League, I did a lot of sabermetric analysis that BA never bothered with. Their stock in trade has always been prospects and toolsy players, and by definition there aren't many of either in the independent leagues.