More than a hundred sultans of swing mechanics brought their A-game to Kline Tower last week.
This metrics-laden cohort of sports analytics experts — a group that included undergrads, professors, consultants, stats folks from ESPN, a data engineer for the National Football League, and the principal data analyst for the Detroit Tigers — had convened for the 2025 Connecticut Sports Analytics Symposium, hosted by Yale’s Institute for Foundations of Data Science.
“Your work brings order to the unscripted chaos of sports,” Meg Kirkpatrick, Yale’s associate provost for research, told the group. “Every exciting, seemingly spontaneous home run, touchdown, buzzer-beater goal is built on thousands of scrutinized data points informing every single decision, from equipment purchasing and training methods to roster development and beyond.
“In a world of new and bigger challenges,” she said, “we will rely more and more on the evolution of foundational methodologies and techniques to help shed light on challenging problems, and where can we see that relationship between chaos and order more than in data analytics and sports?”
Theo Epstein, Sean Ahmed (principal data analyst for the Detroit Tigers), and Brian Macdonald
Over the course of the symposium, held April 11-12, the group pondered an array of stats-related questions:
Is there a better way to track and model tackling ability in football? What behavioral factors influence the success of penalty kicks in soccer? How has the extra-inning “zombie” runner affected Major League Baseball? What effect does resting a player have on their performance at different stages of their career?
The symposium highlighted research and training workshops from a wide range of academic institutions, including the University of Connecticut, Syracuse University, Carnegie Mellon University, the United States Military Academy at West Point, the University of North Carolina-Charlotte, Simon Fraser University, Smith College, Williams College, The Wharton School, and Yale.
Shane Sanders and Alivia Uribe of Syracuse, for example, presented data on how relatively rare it is for soccer players to kick toward the upper quadrants of the goal — even though kicks to those quadrants have a higher success rate. One possible reason? A missed kick to those areas, sailing over or wide of the net, is more embarrassing than a failed kick lower in the goal.
“Our recommendation is to remove these behavioral consequences,” said Uribe, who, in addition to being an undergraduate studying sports management, is a forward on the Syracuse women’s soccer team.
Toby Moskowitz, the Dean Takahashi Professor of Finance at the Yale School of Management, presented research on the role of luck in professional tennis.
Using a single data point — the likelihood that a particular player will win the next point in a tennis game — Moskowitz is able to extrapolate the probability of that player winning games, sets, matches, and tournaments throughout their career. As for luck, Moskowitz explained that shorter matches are more prone to have unexpected occurrences, a.k.a. luck, than longer matches.
And what would this mean, potentially, for tennis great Serena Williams, who won 23 grand slam tournaments playing shorter, 3-set matches? Would she have won even more tournaments if she’d played five-set matches?
“Looks like more,” Moskowitz said.
Winning posters from the event’s research poster session.
One of the main draws of the symposium was a conversation with Theo Epstein ’95, whose front office leadership brought championships to the Boston Red Sox and Chicago Cubs. Brian Macdonald, a senior lecturer and research scientist in Yale’s Department of Statistics and Data Science, led the discussion.
“It’s a great time to be working in data science and sports,” said Epstein. “You can walk into any sports franchise now and you’ll see players working with analysts, trainers, and conditioning coaches, with the data right there on a laptop. It’s so refreshing to see.”
As general manager for the Red Sox, Epstein embraced advanced metrics as a tool for drafting players and developing talent. For example, the team valued players who could hit for power and hold strikeouts to a minimum. That’s how they drafted future star Dustin Pedroia.
Epstein and the Red Sox also worked with neuroscientists to identify key neural pathways exhibited by successful players — and used that information to draft (now perennial All-Star) Mookie Betts.
After winning the World Series with the Red Sox (2004 and 2007) and the Cubs (2016), Epstein eventually took a job with Major League Baseball as a consultant. He helped usher in rule changes — buoyed by analytics — that sped up the game and generated more dynamic action on the field.
Those changes included the establishment of a pitch clock, a ban on defensive shifts, and the use of bigger bases (which made them easier for runners to steal).
In 2024, Epstein joined Fenway Sports Group, which owns the Red Sox, hockey’s Pittsburgh Penguins, and the Premier League’s Liverpool soccer club. He continues to advocate for the use of analytics — along with empathy and respect for the human beings those analytics represent — to bring out the best in athletes.
“Back when I started out, it felt like we were just rolling the bats and balls out onto the field and expecting the best players to make their way up,” he said. “It created this organizational tension and dynamic. Now, with this young generation of players coming up, there’s such great collaboration and transparency. And the data is at the center of that.”
Sponsors for the symposium included Ceros Financial Services, the Connecticut Sun, ESPN Analytics, the Red Sox Foundation, the Connecticut Statistical Data Science Lab, the National Science Foundation, the UConn Data Science Club, the UConn Department of Statistics, the UConn Joint Statistical Club, and the Yale Institute for Foundations of Data Science.
