New Author, Park Factors, and BABIP

I'm SalB918 (you can call me Sal) and I am a new author at Beyond the Boxscore. I have half the brains of Marc, Dan, and Richard, but I have twice the ego. I have a mathematical and scientific bent, which is how I hope to present my analyses. You'll find me posting regularly at AthleticsNation (a member of the SportsBlogs Network) from my laptop in the Northeast and I hope to post semi-regular columns.

We're well aware of park factors and how they suppress or support offensive output. Many of you also know that park factors can be broken into components. For example, Dodger Stadium is more or less neutral when comes to home runs, tends to suppress doubles. Lately, I've been a little obsessed with BABIP, or batting average on balls in play. Voros McCracken posutlated (and I am oversimplifying) that a pitcher has no ability to control the batting average of opposing hitters once they hit the ball in play. The same is not true of hitters; they have some degree of control over their BABIP (line-drive hitters tend to have high BABIPs).

Do individual parks have different BABIPs? That is, if I put a ball in play in Cleveland, is it more or less likely to fall for a hit than if I put a ball in play in Tampa Bay (based on my performance in intramural softball, I won't be putting any balls in play anywhere). I collected the cumulative hitting statistics for all American League batters who accumulated more than 50 PA from 2002-2005, sorted by ballpark. Enough players come and go that I will assume that a three-year average smooths the data sufficiently. I ignored the NL because pitcher's batting statistics are so far off the charts (in a bad way) that they screw up BABIP calculations.

*Park factors are three-year batters park factor averages from Baseball-Reference

Some interesting notes:

1. Fenway Park has a BABIP more than two standard deviations over the average. Perhaps in Boston, a strikeout is worse than other types of outs - Mark Bellhorn, this is your life!
2. Another hitter's park, US Cellular, actually has a low BABIP. It might make sense for Chi-town hitters to swing for the fences and PUT IT ON THE BOOOOOARD....YES! Ozzie Guillen's collection of slap hitters and scrappy hustlers should take notice.
3. Both of the turf fields in the American League have BABIPs well over the league average. Balls skipping through the infield or bouncing over fielder's heads could be responsible.
4. All that foul ground in Oakland is a prime suspect for having the low BABIP at the Coliseum. Someone should do a detailed study and see if there are significantly more foulouts at the Coliseum and whether this accounts for the low BABIP.
5. In general, BABIP and park factor are not well correlated.

This information needs to be taken with a grain of salt. For example, there have been whispers around Oakland that Billy Beane has figured out defense, and that might account for the low BABIP at the Coliseum. On the other hand, the current incarnation of the Red Sox team isn't exactly known for its world-beating defense.

Still, there is value in this analysis. As a second-order effect on Mr. McCracken's DIPS theory, park factor BABIPs can have an effect on roster construction. If you have a home park that suppresses BABIP, for example, you may be able save some money by going after pitch-to-contact bullpen artists instead of strikeout types. With the money saved, you could get some more hitting or finally give your bullpen catcher that raise he so richly deserves.