Newsroom

Unexpected disruptions have a way of shaking up even the most well-established systems, and the world of professional sports is no exception. When the COVID-19 pandemic threw into chaos the traditional schedules of leagues like the NFL, MLB, and NBA, it wasn’t just the games that were affected—it was the very framework that teams and rankings relied on. With fewer games and playoff uncertainty looming, the challenge became clear: how could teams maintain competitive integrity in the face of such upheaval?

This was the question that intrigued UCI Paul Merage School of Business Associate Professor John Turner and his doctoral students, Ali Hassanzadeh and Mojtaba Hosseini. Together, they saw this disruption not just as a problem, but as an opportunity explore new ways of thinking about how professional teams are ranked. They set out to create a solution that would protect the integrity of team rankings the next time broader circumstances force seasons to spontaneously change course.

Their paper, “How to Conclude a Suspended Sports League?” explores their novel approach to this problem. It was recently published in Manufacturing & Service Operations Management.

Ph.D. Student and Basketball Fan Tackles Pandemic NBA Issue

Turner credits the inspiration for the paper to Hassanzadeh, who was deeply affected by the uncertainty surrounding sports leagues during the pandemic. “He was working on other research when he was doing his Ph.D. with me,” Turner says. “After COVID started, he was watching the news, and they kept talking about how no one knew what was going to happen in the NBA or how to resume the league in a shortened time frame.”

Hassanzadeh is a dedicated basketball fan who has a strong background in machine learning and optimization. He began to ponder the NBA’s issue, and his curiosity led to the creation of a research project that would eventually involve Turner and fellow Ph.D. student Hosseini. “The impetus for the paper was to use what we know about predictive and prescriptive analytics to look at this problem,” Turner says. The result was a comprehensive study that provides a framework for concluding sports leagues under extraordinary circumstances.

Model Predicts Fair, Accurate Outcomes as if Teams Played Full Seasons

Rather than focusing on logistical concerns, Turner and his doctoral students homed in on the issue of maintaining fair and accurate team rankings. “The question we asked ourselves was, ‘If we believe the most important thing is to have rankings at the end of our shortened season that are as close to what the rankings would have been if the full season was played, what would we do?’”

This focus led them to develop an optimization model that explicitly accounted for the uncertainty inherent in sports outcomes. They specifically modeled “the uncertainty that exists in the full season ranking,” says Turner. “We compared what the ranking was of our shortened season with the suggested games in the full season as if they had played everything that was left in the remainder of the season. Then we compared the rankings of those two. Our goal was to make those rankings as close as possible.”

The research involved sophisticated mathematical modeling, including the use of stochastic processes to simulate various possible outcomes. Turner is quick to point out that this approach was not about picking winners and losers but about maintaining the integrity of the competition. “We let the model tell us what the solution should be, in a sense,” he says. “It's not that we said, ‘We want to pick the more competitive games.’” Rather, they spent time considering how they should go about modeling the situation.

Implications for the Future of Sports and Other Fields

While the study was born out of the unique circumstances of the COVID-19 pandemic, Turner believes its implications extend far beyond this specific context. “You don’t have to wait around for the next pandemic to use it. There are other instances where leagues are interrupted. Player strikes are probably the largest one,” he says. The researchers’ model could be applied to any sport or competition facing a similar dilemma, from basketball to chess tournaments.

Moreover, Turner sees potential applications in other fields, albeit with some caveats. There are some general principles, he says, but he doesn’t want to go too far to suggest others do exactly what they’ve done. “A lot of work goes into finding the right tweaks that work in a specific environment.”

One of the key takeaways from the research is its focus on fairness and accuracy in rankings, which Turner believes is crucial to maintain the integrity of any competition. The model they developed aims to ensure the teams that deserve to make the playoffs do so—and that home court advantages and other critical factors are preserved. “By having rankings that are similar, we’re implicitly also making sure the right teams are given the chance to compete in the playoffs,” Turner says.

Tool Maintains Integrity of Competitions in Future Disruptions

As sports leagues around the world continue to grapple with the challenges unexpected interruptions pose, Turner and his colleagues’ research offers a valuable tool for maintaining the integrity of these competitions. By focusing on predictive and prescriptive analytics, they have developed a model that not only addresses the immediate needs of a shortened season but also provides a framework for dealing with future disruptions.

In a world where uncertainty is increasingly the norm, their research serves as a reminder of the importance of adaptability and innovation to ensure that, no matter what happens, the game can go on.