Ultimate Analytics: A study of elite teams' offenses

by David Zhang

Institution: Harvard University
Degree: AB
Year: 2015
Keywords: Statistics; Mathematics
Record ID: 2058380
Full text PDF: http://nrs.harvard.edu/urn-3:HUL.InstRepos:14398550


Many traditional, powerhouse sports are currently undergoing an analytics revolution. While ultimate is a relatively young sport, it is certainly not immune to this revolution. Most ultimate data presently track basic summary statistics, but some more advanced work in the field on spatial analytics has been done. We stand on the brink of an explosion in advanced ultimate analytics. This paper attempts to progress that field, analyzing data from elite men's club teams, namely Denver Johnny Bravo, Boston Ironside, and San Francisco Revolver. We analyze how variables beyond just spatial location affect the probability of scoring. Our results demonstrate that elite teams should attack downfield to gain yards while prioritizing the creation of ``entropy" – throwing more passes rather than fewer and holding the disc as briefly as possible; once in the red zone, teams should modify their offense to maximize space while changing their points of attack by swinging the disc laterally. We propose two new end zone sets for offenses to run. Our method, when combined with conventional ultimate wisdom, provides a strong attempt at streamlining offenses to be more efficient at scoring.