Wednesday, March 24, 2010
Great. I have to get my calculator out of the trash now?Posted by Mike Fast
Morgan Ensberg has been blogging for a few weeks, and he's written some interesting posts. His latest post was his highly-anticipated treatise on sabermetrics from a player's point of view.
It was an entertaining and thoughful post. Ensberg gave several examples of how sabermetric input from family and coaches has influenced him. He also noted that there are many questions that players face for which there has been little or no sabermetric study. He used the example of bunting to illustrate the point.
I need to know the probability of success when sacrifice bunting a ball in a 5 foot by 5 foot box that stops 20 feet from home plate and no more that 5 feet from the foul line (this can be either foul line).I'm not aware of a study of bunting that addresses any of these questions, although I have pondered doing such a study in the past. The MLBAM Gameday fielding location data would tell us the probability of a successful sacrifice bunt at any location on the infield. The PITCHf/x data would tell us about the pitch type and location of what pitchers throw in various bunting situations. So a lot of this kind of data exists, and there are studies that have been done with player-centric views on topics other than bunting.
I need to know the probability of success when sacrifice bunting in that spot when a batter shows the bunt as the pitcher comes set (early) vs. showing the bunt as the pitcher begins his delivery (late).
I need to know the probability of success when showing the bunt early and pulling the bat back to slash. A slash is where the player pulls the bat from a bunting position back to a hitting position and tries to hit the ball to the side of the charging infielder.
But what I really need to know is the opposing pitcher’s pitch and location when the batter shows the bunt early. This will allow me to tell my batter what to expect when he is up there trying to get this bunt or slash down.
I need to know the probability of my opponent throwing a fastball away so that I can anticipate which middle infielder will cover second if I put the runner in motion for a hit and run in this sacrifice situation.
However, the data about what the batter did as far as showing the bunt is not something currently available, and I'm not sure what I'd propose to collect it. Again, this is not a question restricted to bunting. We currently know very little about the intentions and behaviors of many of the players on the field, although the upcoming FIELDf/x system will presumably tell us a lot about what fielders and baserunners do.
Ensberg is correct that most of the sabermetric research is oriented toward questions in the general manager or field manager purview. I agree with him in that my primary interest in sabermetrics lies with the player-centric questions, particularly for me, with the intricacies of the batter-pitcher matchup. I think the fans in general will also benefit from seeing the player viewpoint through the sabermetric lens.
Greg Rybarczyk is a leading thinker and evangelist for this viewpoint, promoting an immersive experience for fans through a FIELDf/x and GAMEf/x presentation, with its ancillary benefits for sabermetric research. Marv White was also a big proponent of the value of this approach in his leadership at Sportvision in developing PITCHf/x, HITf/x, and FIELDf/x.
I'm excited to be part of the current revolution in sabermetric thinking. It's a lot of fun, and it may well transform the game, even for the players, while we're at it.
Mike Fast is a Royals fan who enjoys investigating baseball questions using data of many sorts. He is a member of Complete Game Consulting. He welcomes comments via e-mail.