DiPrima Lecture Characterizes Skill in Real and Fantasy Sports Competitions

Anette (Peko) Hosoi visited the mathematical sciences department on September 9 to present the 2024 Richard C. DiPrima lecture, which honors and remembers a former faculty member who through his research, teaching, and service as department chair, was instrumental in the transformation of our department in the 1950s-1980s to a prominent center for the application of mathematics to science and engineering concerns.  Dr. Hosoi’s highly interactive colloquium:  “A Few Short Stories About Mathematics and Sports”  began with describing a machine-learning methodology based on player-tracking data to characterize the scoring potential of player configurations in basketball, and the skill with which particular players capitalize on momentarily favorable scoring opportunities.   Her group’s analysis provides a means to automatically label the parts of a game where scoring opportunities were missed, potentially assisting the review of games by coaching staff.  The second story concerned research pursued to address a legal concern of companies like FanDuel about whether performance in fantasy sports is largely a measure of luck or of skill.  Dr. Hosoi’s team developed a quantitative metric to characterize the luck vs. skill of both real and fantasy sports based on a geometric representation of the persistence of success.  The results of the analysis indicate that both real and fantasy team sports are largely, but not wholly, skill based, with the fantasy sports having a similar luck vs. skill balance as the real sports competitions!

 

Peko Hosoi is the Pappalardo Professor of Mechanical Engineering at MIT, where she co-founded the MIT Sports Lab which connects the MIT community with pro-teams and industry partners to address data and engineering challenges that lie within the sports domain.

Back to top