Two IMS members have been signed to major ice hockey teams in the US — not as players, but as part of the growing number of statistical analysts in professional sports. Sam Ventura, a visiting faculty member at Carnegie Mellon University (CMU), has been signed by the Pittsburgh Penguins as a consultant, and Andrew Thomas has a similar position with the Minnesota Wild hockey club, though he now is on the faculty of the University of Florida.

“In any field, if you can objectively back up your decision with data, you’re doing yourself a favor,” said Ventura. A Pittsburgh native and lifelong hockey player and enthusiast, Ventura’s appointment as a consultant with the Penguins stems from his senior year at CMU, when he took a class about applying statistical methods to the sporting world taught by Andrew Thomas.

Ventura said he and Thomas, a native of Toronto, had a mutual interest in hockey. They put their heads together on a project rating players in the National Hockey League, which incorporated statistical properties that went beyond the standard plus–minus rating, where a player is assigned a value based on the total number of goals his team and the opposing team scores while that player is on the ice. The project led to a paper that was published in The Annals of Applied Statistics and a panel discussion, “Advances in Methods for the Analysis of Ice Hockey,” at the Joint Statistical Meetings.

Ventura said, “It was a good discussion, with a big Q&A with the audience. The conclusion was we thought our work was very good, but the common and advanced fans didn’t have access to it. There was no public presentation of our results.”

Enter, a comprehensive, visually appealing website that Ventura and Thomas co-developed to introduce fans to more statistical metrics. WAR stands for Wins Above Replacement, which measures how many wins a player helps a team achieve over someone who would be his replacement. While fans can go to for player and team statistics, Ventura says is different in two ways. “War-on-Ice is geared to the more advanced fan, who is interested in stats that are good descriptors of what happened in the past but also good predictors of what will happen in the future,” Ventura explained. “That’s what’s behind the core tenet of modern hockey stats. We like to present the metrics that are predictors, like WAR.” Ventura said another big predictor is the total number of shots a team takes during a game, even the ones that are blocked and not officially logged as shots on goal. “The idea is that the more shots a team takes, the more puck possession time they have. That’s a good predictor of wins and losses and future success,” Ventura said.

Ventura feels the website and the Pittsburgh Hockey Analytics Workshop, hosted by War-on-Ice and the Statistics Department at CMU last November, piqued the Penguins’ interest in him.

“We’re excited to add someone of Sam’s talent to our organization,” said Jason Karmanos, the Penguins’ vice president of hockey operations. “Advance stats data continues to emerge and evolve daily, and Sam will be instrumental in helping us interpret new findings in that area.” Ventura said areas in which quantitative analysis could provide value include research about which players play well with each other and against each other, and if it’s advantageous to play your starting goaltender in games on consecutive days.

This article is based on an article in The Piper, Carnegie Mellon’s online newspaper.