Developing a winning equation with the Chicago Cubs
On an unseasonably warm September afternoon, Joel Shapiro sat in the stands at Wrigley Field as the Chicago Cubs put the finishing touches on a 12-1 win over the St. Louis Cardinals.
Just hours earlier, Shapiro had made the journey by train and foot to the Windy City’s North Side for a meeting with the team’s general manager and president. The three spent an hour whiteboarding, brainstorming and discussing how data can — and can’t — shape decisions.
By the time right-hander Colin Rea threw the first pitch, Shapiro was in the “control room,” watching the franchise’s data team analyze the game in real time.
Looking back, the experience was set in motion months earlier, when Shapiro, a clinical associate professor of managerial economics & decision sciences at Kellogg, presented at a conference about how data is only helpful if it turns into evidence for better decisions. The Cubs’ senior director of baseball data was in the audience. The notion resonated. Connections were made. An opportunity arose.
As executive advisor to the franchise’s president and general manager, Shapiro has spent the 2026 season helping the Cubs improve the team’s approach to high-stakes roster decision making. Specifically, he’s been tasked with advising the team on how to improve and formalize their decision-making to best invest hundreds of millions of dollars a year on the right players. The rigorous and systematic approach relies on complicated data streams and machine learning algorithms, but also human judgment.
“The data are only one part of the story,” says Shapiro. “You can run good models that do a nice job of predicting what players are going to do this year, next year and the year after. But even accurate predictions are not the same as high-quality decision making.”
That gap between what models can predict and the role that human judgment plays is core to what Shapiro teaches, researches and practices. After getting his PhD in public policy analysis, he spent years in education policy before the business world's growing embrace of data pulled him toward a new application. Sports was a natural fit: rich data, clear objectives and a culture that Michael Lewis’ book Moneyball had primed.
In Shapiro’s Analytical Consulting Lab at Kellogg, student teams spend 10 weeks tackling live data problems for real clients across industries. Sports-related clients have included:
- The Northwestern Athletics Department — exploring recruiting in the era of student-athlete compensation
- The Chicago Bulls on stadium operations
- The Chicago Bears on fan engagement
- The streaming service environment for the Bulls, Blackhawks and White Sox (Shapiro’s favorite home team, despite his work on the North Side)
Regardless of their client, students learn how the goal of any analysis is not to build a great model, but to make high-quality decisions.
“A lot of students come in thinking, ‘I can't wait to get this beautiful dataset and run this amazing analysis,’ ” he says. “What they need to understand is that data feeds decisions, and decisions roll up to business outcomes. None of the data or modeling matters if it doesn’t result in good decisions that move the business forward.”
That applies to the Cubs as well. Building a roster, Shapiro says, means solving three distinct problems: who to acquire, how to develop them and how to optimize the current team. Predictive models can be genuinely sophisticated, but they don't resolve questions about risk tolerance, organizational time horizons or how to allocate finite resources across competing needs. “There’s a big difference between trying to make the playoffs this year versus winning the World Series five years from now,” he says.
Shapiro’s research and consulting focus has recently centered on how much player injuries affect winning. His analysis — using data from the National Football League — finds a flatter-than-expected relationship between spending on players who get injured, and wins. He found that teams absorb injuries better than they're typically given credit for. With one exception: A highly paid starting quarterback being injured for even just a few games has an outsized impact on team outcomes.
Back on campus, Shapiro is expanding the school’s new Sports Intelligence Project, an effort to connect Kellogg students with sports organizations to help answer the sports industry’s hardest strategic questions. As artificial intelligence closes the gap between the best and the rest in model-building, he says, the differentiator will be judgment: knowing how to weigh trade-offs, think across time horizons and make consequential calls under uncertainty. “Organizations that hire for that,” Shapiro adds, “will have stronger foundations to build their future.”
Read next: Sports Intelligence Project explores high-end decision making in multifaceted industry