Jun Yan, University of Connecticut

Meredith Wills with the SportsMedia Technology data challenge finalists. From left: Cameron Grove, Durham University (UK); Jack Rogers, University of Minnesota; Kai Franke, University of Minnesota; Jackson Balch, University of Minnesota; Isaac Blumhoefer, University of Minnesota; Cale Williams, Georgia Institute of Technology; Ethan Rendon, New York University; Jack Weyer, University of Southern California; Billy Fryer, North Carolina State University; and Meredith J. Wills, SportsMedia Technology

The fourth UConn Sports Analytics Symposium was held October 8, 2022, in McHugh Hall at the University of Connecticut, Storrs. The event, hosted by the UConn Statistical Data Science Lab, attracted about 90 in-person and 170 virtual participants. A majority of the virtual participants registered through the symposium’s international partnership in China, and more than half of the registrants were graduate or undergraduate students.

There were two new features this year. First, a data challenge was sponsored by SportsMedia Technology, an industry leader in sports data collection and visualization. The goal of the challenge was to analyze an aspect of player movement (e.g., baserunning, movement while fielding, backing up a play) for minor league baseball players using in-game player and ball location data for multiple teams over multiple seasons. The data was provided by SportsMedia Technology, and the organization committee was chaired by Brian Macdonald of Yale University.

Two live coding sessions were held last summer, with 39 individuals/teams submitting entries. The judging committee, chaired by Meredith Wills of SportsMedia Technology, was an elite team consisting of sports analysts from both academia and industry, as well as players, umpires, and scouts. Six finalists presented their work in a poster session, with the following two students winning:

• Graduate: Cameron Grove, Durham University (UK), “To Shift or Not to Shift? Using Player and Ball Motion Data to Build a Highly Flexible Defensive Positioning Algorithm in Baseball”
• Undergraduate: Billy Fryer, North Carolina State University, “Should I Stay or Should I Go?”

The second new feature was that the symposium was supported by a three-year conference grant from the National Science Foundation. The proposal was titled “Conference: UConn Sports Analytics Symposium: Engaging Students into Data Science.” The principal investigator was Jun Yan, and the co-principal investigators were Laura Burton, Kun Chen, Robert Huggins, and Elizabeth Schifano. The grant supports travel expenses for student presenters and invited speakers. This year, 12 out-of-state students were given awards to travel to UConn to present their posters, including the six data challenge finalists.

The symposium started with a brief overview of the event by Yan. Welcoming remarks were delivered by UConn President Radenka Maric, NSF Program Director of the Division of Mathematical Sciences Edsel Peña, and UConn Field Hockey Head Coach Paul Caddy. The morning keynote, “Software Engineering for Sports Analytics,” was given by Tegan Ashby, senior full-stack developer of basketball systems at Brooklyn Nets. David Bergman, associate professor of operations and information management at UConn, presented the afternoon keynote, titled “Integration of Analytics Techniques for Algorithmic Sports Betting.”

A panel discussion about sports data science competitions followed the morning keynote. It was moderated by Alison Lukan, Seattle Kraken contributor and television analyst for Root Sports. Panelists included Michael Lopez, senior director of football data and analytics at the National Football League; Brendan Kumagai, data science intern at Zelus Analytics; Megan Risdal, lead product manager at Kaggle; and Asmae Toumi, director of analytics and research at PursueCare. Kumagai and Toumi were winners of the NFL Big Data Bowl in 2022 and 2021, respectively.

Greg Matthew congratulates poster award winner Shinpei Nakamura Sakai.

The poster session was held in person. Sixteen posters were on display and judged by a committee chaired by Brian Macdonald and consisting of Sean Ahmed of the Pittsburgh Pirates, Chris Harden of ESPN, Ethan Meyers of Yale University, Elizabeth Schifano of UConn, and Bob Wooster of Yale University. The Student Poster Award went to Shinpei Nakamura Sakai of Yale University for “Estimating Conditional Average Treatment Effects for Player Performance Over Time: Assessing Load-Management in Sports.”

Six training workshops in three 50-minute parallel sessions were well attended. The instructors included five undergraduate and one graduate student from UConn. The introductory-level workshops were “Introduction to R” by Fusheng Yang and “Introduction to Python” by Charitarth Chugh. The intermediate-level workshops were “Hockey Analytics” by Venkata Patchigolla and “Baseball Analytics” by Patrick Cummins. The advanced-level workshops were “Web Scrapping for Sports Data” by Hari Patchigolla and “TensorFlow in Sports Analytics” by Pranav Tavildar. View the archived training materials.

Learn more about the UConn Sports Analytics Symposium. Questions or ideas for future symposiums may be sent to Yan via email.