Trench Chess: CMU Sophomore Uses NFL Data To Examine Pre-Snap Play

Six-year-old Abhishek Varadarajan watched with his family as Eli Manning and the Giants defeated Tom Brady’s Patriots in Super Bowl XLVI, a game that kicked off his lifelong love of football. 

Now a sophomore at Carnegie Mellon University, Varadarajan’s gaining yards toward his own goal line: a career in the NFL. His project “Trench Chess(opens in new window),” a semifinalist in NFL’s 2025 Big Data Bowl(opens in new window), analyzes different ways a defense can create confusion at the line of scrimmage.

Raised in New Jersey, Varadarajan could sense his community’s excitement leading up to the 2012 championship, as even the students in his elementary school buzzed about how the Giants might win the Super Bowl. He religiously followed the team and the NFL, and in his senior year of high school, he wanted to learn more about what was happening on the field. 

“I’ve been watching football for as long as I can remember. I started digging into the schemes and the player techniques in high school. I was watching coaching clinics and game film on my own time just to learn as much as I possibly could about the game,” Varadarajan said. 

At Carnegie Mellon University, Varadarajan found his place studying statistics and machine learning in the Dietrich College of Humanities and Social Sciences(opens in new window). 

Varadarajan set out to explore his interest in the world of sports analytics, and during Orientation Week, learned about statistician Ron Yurko(opens in new window) and the Carnegie Mellon Sports Analytics Center (CMSAC). 

Yurko,  CMSAC’s director and an assistant teaching professor in the Department of Statistics & Data Science(opens in new window), has been a driving force behind CMU’s continued success at the annual NFL Big Data Bowl. He’s guided multiple projects for the contest since its inception, including two from statistics doctoral student Quang Nguyen that were named finalists in 2023 and 2024. Nguyen and Yurko were also semifinalists for their submission this year. 

“Abhi is a football fanatic,” Yurko said. “His project looked to objectively quantify these concepts of ‘stunts’ and ‘disguises’ using modern tracking data. And he found some interesting relationships there.”

Using positional data from sensors worn by each player on the field during the first half of the 2022 NFL season, Varadarajan tracked how successful teams were at disguising defensive pressure and using stunts (when defensive linemen switch roles).

“Teams tend to either disguise a lot, or have one player who’s great at winning in isolation,” Varadarajan said. “If teams don’t disguise their pressures, they aren’t going to get pressure without a key rusher. If you don’t have that talent, you are forced to allocate more resources to causing confusion in order to get pressure.”

A football fanatic

In his pursuit of football knowledge, Varadarajan didn’t stop once he found the Sports Analytics Center. This fall, he joined the CMU football team(opens in new window) as a manager to get closer to the game. 

In the role, Varadarajan took on an equivalent time commitment to that of a student-athlete. He filmed games for use in post-game review by the coaching staff, helped to set up the field for practice and games, and maintained equipment. And though he was paid for his time, the real value came from insight gained.

“It’s helped him to be around football, and see the game every day,” said Sam Turner, equipment manager for the Tartans. “Whenever coaches come into the equipment room, I’d see Abhi try to pick their brains about schemes, and coaches would take him over to the white board to offer their knowledge about anything he wanted to learn."

Varadarajan is also an officer in CMU’s Sports Analytics Club, a student-run sports statistics group that keeps the conversation going outside of class. Like in a chess match — or a pre-snap disguise — he’s setting up a long-term position for success.

“I’d really like to work in a team’s front office, helping to win games every week,” he said.