Log in

No account? Create an account

Previous Entry | Next Entry

Lost in the weeds

After reading several "standard texts" on sabermetrics over the last few weeks, I have a gut feeling that one of the reasons a lot of professional baseball people don't have a lot of time for sabermetricians is that they waste a lot of time mangling statistics in various inventive ways to answer stupid questions like "Was Honus Wagner a better baseball player than Cal Ripken?" "If Roger Maris played for the Yankees today, would he hit sixty home runs a year?" and "Was Dom DiMaggio really better than his famous brother Joe?" The answer to all of these questions, as far as I'm concerned, is NOBODY CARES. The kind of people who usually get into arguments like this are the kind of folks that don't care about Equivalent Averages, normalized OBP+SLG, and VORP, because those are all damned statistics, as in "Lies, damned lies and statistics." You can find those people in any sports bar or the comments sections of most popular sports blogs, arguing with the guys that insist that defensive stats prove that Adam Dunn is a horrible first baseman and not worth what he's being paid for knocking in 120 guys a year with his ~40 dingers. *facepalm*

I submit that to the extent sabermetrics tries to do more than compare apples to apples by straining out park effects and league effects, it's a waste of time. General managers want to know MLEs so they can get a handle on whether Slugging Joe Smith, your favorite team's hot prospect at West Bumfuck, is going to keep slugging when he's brought up to replace Harry Hasbeen at first base, so they can let Joe and his multi-million dollar salary go gracefully into the good night, or to Kansas City if they're dumb enough to trade for him. There's also some useful tools for helping you make an informed guess as to whether, ceteris paribus, a player is going to get better or worse next season, what will happen if we move a player from a bandbox like Wrigley to a pitcher's park like Kauffman Stadium, and should we keep a pitcher who won twenty games but only struck out two men per game. These are also the questions that informed fans, whether or not they play fantasy/Rotisserie baseball, want answered.

Now in my case, I've been vexed for years over the matter of the independent leagues. How do they compare to the farm systems? Was the Northern League and its imitators really operating at an AA/AAA level when I covered them from 1995-2000? Was the Frontier League comparable to the Midwest League, or was it more like the NY-Penn or Sally Leagues? Trick question; the FL, being a short-season league mostly staffed with undrafted college players, can't really be compared to the Midwest League or SA, which played full seasons with rosters full of guys who would have been considered LS-2 men, which is to say limited service players with two years or less in pro ball. The reason sabermetrics may not help me figure this question out is because unlike the farm teams, you don't have large numbers of players moving systematically between leagues. The farm teams move hundreds of players up and down the pipeline every year from short-season A to triple-A, and that gives you a big enough sample size to see if your predictions work out when a player moves between leagues. You don't have that with the independent leagues. They exist outside the far team pipeline by definition, and player movement from the independents to the farm system is a newsworthy event because it doesn't happen all that often. So you're starting with a very small sample size, and that makes whatever statistical comparisons you do kind of iffy. Going to see what I can do, though.