Analytics Taking Hold in North American Sports Organizations

Hockey teams have access to various kinds of in-game data, using player tracking and shot maps to observe player and shot speed, shot locations and distances traveled. Teams across the NHL and professional sports have made analytics part of their front office operations (Photo by Matthew Page)

By: Matthew Page

Games in professional sports are often defined by the performance of the players on the field. Stats in the box score are the main way fans are able to understand each sport. However, in organizations, the data goes beyond these stats and takes the form of performance-based analytics. 

These statistics have taken hold across the four major sports, from roster building to decision-making. Billy Beane’s “Moneyball” Oakland Athletics was a popular example in 2000’s baseball, with his work starting a revolution in the sport. Nowadays, big and small markets now use analytics to find a competitive advantage.

Trailer Courtesy: Sony Pictures

“The interesting thing that I noticed while in baseball was the data was changing and coming at you faster than we were able to make use of it,” said Chris Fonnesbeck, a former quantitative analyst with the Philadelphia Phillies and New York Yankees.

Fonnesbeck cites tools such as Hawkeye, a system that measures the metrics behind pitched and batted balls and TrackMan, which provides biomechanical data behind the pitchers, tracking release points and the arm angles, as examples used by Major League Baseball today.

Case Studies

In football, the analytics help not just with roster decisions, they also assist with how teams approach in-game situations. A 2025 study from Digital Access to Scholarship at Harvard notes, "Teams pass more, run less, and take more risks than they used to. Fourth down aggression, in particular, has become way more common over the last decade. What used to be seen as “risky” is now often standard practice, especially for teams that buy into analytics."

Younger coaches in football use analytics as a base to take more risks in-game, whether it's going for it on fourth down or kicking longer field goals (Photo by Muyuan Ma on Unsplash)

As a result, the number of fourth down attempts per game have noticeably increased over the last decade:

• 2015-19: 2.01 attempts per game

• 2020-24: 2.80 attempts per game

• 2025: 552 attempts through 178 regular season games, 3.10 attempts per game

Today's data examines the player’s position, speed and direction on a play, such as expected points added, expected yards per carry and expected yards per catch. Brad Congelio, assistant professor at Kutztown University’s School of Business and author of Introduction to NFL Analytics with R, values rushing yards over expected, which factors metrics to how running backs perform on paper compared to in game. 

“It’s really fascinating to dig into it, trying to determine on any given play why that running back didn’t get the expected rushing yards or why did it go over the expected rushing yards,” he said. “The reason I really like rushing yards over expected in terms of a teaching perspective is that ultimately it’s easy to understand.”

In hockey, analytics can be used in a similar way, with a player’s offensive and defensive performance can be evaluated by expected goals for and against. Goalies can be evaluated by goals saved above average, and these metrics apply to both NHL players and prospects. 

Michael Schuckers, professor of data science and sports analytics at University of North Carolina Charlotte, said creating models for prospects is complicated, with evaluations being about relativity. 

“The great challenge there is dealing with players who are playing, say, U.S. high school on Long Island versus someone who is playing in the second division in Sweden against grown men,” he said. “How do you deal with and adjust for their performance in those two different situations along with all of the other leagues that are out there?”

Imperfect Analytics?

While analytics certainly give teams an advantage, that may not always translate into how games are played. Regardless of how hard the ball is hit, the scoring chances created, or the explosive plays, it doesn’t always translate to victories. 

Derek Grifka, creator of the MLB Deserve-to-Win-O-Meter on X, built models that factor exit velocity and contact to see which teams deserved to win or, in some instances, got lucky. However, there are shortcomings that may not factor in who deserves to win, such as player speed, pitching or fielding. Grifka says the models are not meant to recreate games, but they get people to think about it differently. 

“We deal with the randomness, obviously, stuff that will never be able to qualify,” Grifka noted, citing examples such as clubhouse leadership or morale. He added a popular data science saying that, “all models are wrong but some are useful.”

In college basketball, the same can occur. Evan Miyakawa, creator of college basketball database evanmiya.com, says upsets and early season data can be a cause for concern among fans, but his model is to adjust based on preseason projections and regular season performance. 

“You can properly start to adjust and get more accurate in your understanding of how good a team is or how they play stylistically in this one area, but not overreacting too quickly to what’s going on in the beginning of the season,” Miyakawa said. 

Recent Advancements, Why Analytics Matter

While today’s sports analytics tools has pros and cons, it’s a growing industry with the technology continuing to evolve. As methods change, the ways analytics can store and share information will continue to grow. 

Notable examples include new NHL player tracking technology, which Schuckers claims can give access to new data. Fonnesbeck believes AI is worth monitoring in sports data, with companies like Sportlogiq using it for hockey and soccer data.

Analytics are also making their way into leagues beyond the four majors. Notably, advancements are being made for women's sports to have their own data analysis strategies. In the National Women's Soccer League, Kitman Labs, a sports science company, has partnered with teams to provide insights to player performance, training and recovery.

However, methods such as machine learning and forecasting can be applied not just by teams, but also by businesses at large to help reach their goals. Meredith Somers, formerly of the Massachusetts Institute of Technology, wrote in 2020, "companies want to be at the top of their respective markets — hiring the best employees, maximizing productivity and efficiency, and building customer relationships. And one way to meet those business goals is by adopting a sports-analytics approach."

Whether it's sports, business or other realms of data analysis, Fonnesbeck notes it’s a great experience for those who look to learn more about the world of data. “I think it’s great for an analyst to be able to work in baseball for sure, but there are other sports as well that would be a useful training ground for new analysts, even if their long-term goal isn’t necessarily to be working in sports,” he said.