Crunching Numbers

For the past three years, I’ve scouted our Florida opposition by looking at the previous season’s stats for team batting average; runs scored; home runs; stolen bases; errors and earned run average.  It’s also good to know who the best returning players are.  I then forward the info to my daughter.  We like to be prepared.

That said, I’m not sold on sabermetrics.  Algorithms and baseball (it doesn’t seem to have infected softball yet) aren’t necessarily the perfect tools Bill James would have you believe.  Consider the case of Diamondbacks outfielder Gerardo Parra.

According to baseballreference.com, last season Parra had the greatest defensive year ever by a right fielder.  Parra’s significant stats were 274 chances with 15 outfield assists (runners thrown out trying to advance) and 3 errors for a .989 fielding average.  Plug these and some other numbers into the magic equation, and Parra’s defense won the Diamondbacks four more games than his replacement was likely to have done.  Well, we wouldn’t want Roberto Clemente out there in right, now would we?

Or Jeff Francoeur, for that matter.  In 2007 when he played right field with the Braves, Francoeur handled 351 chances—77 more than Parra—with 19 outfield assists and 5 errors for a .986 fielding average.  So, the player with more chances and more assists (and just two more errors) generated a measly 1.3 defensive wins above replacement.  Huh?

Or consider OPS, on-base plus slugging percentage.  On-base percentage—basically, hits plus walks as a new kind of batting average—has its uses.  Players who don’t walk a lot tend to make poor leadoff batters, and strikeouts are the bane of sabermetrically inclined G.M.s like Billy Beane and Theo Epstein.  The danger is too great a belief in obp.  Maury Wills had a career on-base percentage of .330 and his teammate Willie Davis totaled just .300.  But Davis still managed to play for eighteen years because he had other skills equally important.

Slugging percentage is a mathematical way of separating slap hitters from sluggers.  Maury Wills’ career slugging percentage was .331.  The determining formula assigns a value of .250 for a single, .500 for a double, .750 for a triple and 1.00 for a home run, so it’s pretty easy to see that Wills mostly hit singles over the course of  his fourteen-year career.  In comparison, Hank Aaron’s career slugging percentage is a more robust .555 and Babe Ruth’s a fairly incredible .690.

But all slugging percentage really does is tell you Maury Wills shouldn’t be batting cleanup or Babe Ruth at the top of the order.  Now take that measure of limited value to make yet another, OPS, and you have the essence of sabermetrics.

Or so says Johnny Old-timer.