As football season approaches, students and recent alumni from the analytics master's program at the USC Viterbi School of Engineering are building a comprehensive analytics database for USC Football , organizing play-level data from thousands of Football Bowl Subdivision games into a searchable platform that will help coaches and analysts identify trends, evaluate strategy and prepare for opponents.
Every football game generates thousands of data points, from tackles and touchdowns to formations and penalties. Turning that information into actionable insights, however, remains a challenge.
Working under Bruce Wilcox , associate professor of industrial and systems engineering practice at USC Viterbi, the team is organizing years of college football data into a searchable data warehouse that will allow USC analysts to quickly test ideas, compare trends across seasons and opponents and eventually study individual player performance alongside play-by-play results.
“I’m just astounded at the amount of data that’s collected,” Wilcox said. “We hope, by the end of the summer, to have that first data warehouse operating.”
Building a better playbook
Wilcox assembled the team at the beginning of the spring semester and asked Saloni Deepak Prabhu, who earned her master's in chemical engineering from USC in 2025, to lead the project's day-to-day technical work. After taking Wilcox's data-focused coursework, Prabhu joined his research group and later the industrial and systems engineering department, where she now helps lead the project.
Prabhu said the research team's goal is to eliminate the time analysts spend preparing data so they can focus on analysis.
“We are building a data warehouse so the analysts can get direct access to well-modeled structured data, so that they can spend time on their analysis and the time spent on preparing the data is eliminated altogether,” Prabhu said.
This data project was commissioned by Conor McQuiston , the director of football analytics for the USC Trojans. McQuiston explained that organizing this data will give his department more room to test out ideas for the Trojans to understand what might or might not work on game day.
“So many different teams have tried so many different things that just having access to all this data lets us run natural experiments,” McQuiston said. “What has and hasn’t worked, and what can we learn from that?”