This is the second in a series of articles that we’re calling “How I got here.” We’ll be talking to McGill alums in their twenties and thirties about their careers and how their time at McGill played a role in what they’re doing today.
When Jonathan Arsenault, BEng’17, MEng’19, PhD’23, left high school, he wasn’t sure what to do with his life. But he was good at math and curious about how things worked, so he gravitated towards mechanical engineering. “Then I met [McGill mechanical engineering professor] James Forbes, who became my master’s and PhD supervisor,” Arsenault recalls. “He convinced me to do a master’s degree. And it was a similar story for the PhD – but it gets a bit crazier.”
That’s because the focus of Arsenault’s doctoral studies was prompted by a request from Eric Tulsky, the general manager of the Carolina Hurricanes, an NHL team based in Raleigh, North Carolina. Tulsky has a highly unusual background for a hockey GM – he was a scientist with a PhD from the University of California, Berkeley, working in the nanotechnology field.
Tulsky wanted to take a more analytical approach to understanding how hockey is played and he wanted to work with people able to sift through data to uncover useful information about how to help improve the performance of his team. The insights gleaned from this strategy have played a significant role in establishing the Hurricanes as one of the NHL’s best teams.
Arsenault has been an important part of that story. A little more than two years after completing his doctorate at McGill, he works for the Hurricanes as an in-house data scientist. We asked him about his experience.
How exactly did you end up doing a PhD in hockey data?
I was doing my master’s in robotics, specifically underwater robotics. We were developing algorithms for autonomous robots. [Eric Tulsky] was embarking on an ambitious search for potential academic partners that have experience working with spatial data – position, location, velocity and orientation over time.
In the NHL, they were starting to collect this kind of data for players on the ice. So, [Tulsky] reached out to Professor Forbes, who is very eager to work on every kind of problem that is presented to him. I was always a hockey fan, and I was also approaching it from an analytical perspective, so my prof was like, ‘I know you’re a good fit for this.’ All the stars aligned.
What’s your background in hockey?
It’s always been a big, big part of my life. To this day, I play once a week – and as you might imagine, I watch a lot of hockey because it’s part of the job.
My dad put me on skates when I was four. I played organized hockey – low-level, nothing hyper-competitive – until I was 17. We’d watch every time the Canadiens were playing. You know, a classic Canadian upbringing.
As I was getting older and the hockey analytics field was starting to become more mainstream in online communities, I naturally gravitated in that direction. So, my hockey fandom took on an even more analytical turn.
How does hockey compare to other sports in terms of using data and analytics?
Compared to basketball, the other North American sports, even soccer, we’re definitely behind. But at one point the NHL started collecting location data of shots, which started to have a big impact in terms of being able to measure things like shot quality. Now the tracking data is really the next frontier. The NHL started putting out some aggregate statistics that show which players are the fastest, which players shoot the hardest. So, we’re moving towards what other leagues have, but it’s definitely a work in progress.
What’s your day-to-day work like?
We collect a pretty extensive dataset from various vendors, including the tracking data from the NHL. Part of my day-to-day is just to make sure that that data is processed smoothly and transformed into a usable state. After that, my core job is to use that information to answer questions that people have about hockey games – essentially player performance, patterns, tactics and whatnot.
How has all of this data changed the way hockey is played?
We’ve seen changes in how people play based on concepts that have emerged through data analysis. One example is related to the power play, when one team has more players on the ice than the other because of a penalty. Teams used to use three forwards and two defensemen on the power play, but then people came out with data saying, ‘Hey, it’s better if you use four forwards and one defenseman.’ So, every team in the league now uses four forwards.
Another example is pulling the goalie. If you are down by one goal late in the game, you’re allowed to take your goalie off the ice and put in another player. People did the math and realized everyone was pulling their goalie too late in the game [for it] to make a difference.
Those are very basic, fundamental things that you can measure that have changed how teams play. You could have said five or 10 years ago that some teams were just completely ignoring data but now, in 2026, nobody’s ignoring it.
How has your work changed your relationship to hockey?
When I’m watching Hurricanes games, there’s still that emotional reaction – we need to win this game. And if we don’t, I’ll be very distraught. But of course, there are also patterns and trends that I pick up on because I’m constantly looking at the data. It’s always in the back of your mind, things that you’ve been working on that day or that week. And then you point it out and message someone on the team and you’re like, ‘Hey, did you see that? Because that was exactly what we were talking about yesterday.’
