|
Abstract:
|
The purpose of this investigation is to assess the quality of pitches thrown in Major League Baseball (MLB). An initial reaction may be that this is a straightforward problem. For example, if a pitch results in a home run, then it was a bad pitch. Alternatively, if a pitch leads directly to an out, then it was a good pitch. However, sometimes home runs are hit on good pitches and sometimes outs occur on bad pitches. We therefore want to assess the quality of a pitch without referring to its direct outcome. The approach is based on characterizing pitches based on pitch count and a descriptor. The descriptor may include information on the type of pitch, the speed of the pitch, the location of the pitch, etc. We consider the estimation of probability distributions for the number of bases yielded by pitches of a particular characterization. This is a big data problem with roughly 750,000 pitches thrown in a single season of Major League Baseball. If we are able to assess the quality of pitches, then a host of practical questions may be addressed. For example, can we detect when a pitcher's performance begins to deteriorate prior to bad results?
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.