How Seattle Is Taking a Data-Driven Approach to Building Its New NHL Team
Illustration by Martin Laksman
On a Friday morning in October, Alexandra Mandrycky, director of hockey strategy and research for the NHL expansion franchise in Seattle, posted a thread of tweets soliciting applications for three jobs in the hockey operations department, hoping for a few nibbles over the weekend. Instead, the interest was so overwhelming—the thread’s first message had 241 retweets and some 1,300 likes at last tally—that Mandrycky had to shut off her notifications.
“I was not,” she says, “expecting that to blow up.”
Then again, Mandrycky probably should have known better than to underestimate Seattle’s willingness to answer her call to arms. Or, more precisely, computers. Except for Silicon Valley, no region of the country boasts such a large pool of qualified candidates for the available positions: developer, data engineer and quantitative analyst. “Microsoft, Amazon, Expedia—all these companies here are tech giants,” Mandrycky says. “They certainly have a lot of employees. We hope some of them may be interested in working for a hockey team.”
Mining local talent is nothing new; the Vegas Golden Knights, for instance, tabbed longtime Cirque du Soleil trainer Jay Mellette as their director of sports performance before their inaugural season in 2017–18. But the resources available in Seattle might prove especially beneficial given the increased—and entirely overdue—emphasis put on data collection, analysis and R&D by NHL front offices. “Hopefully at the end of the day it gives us a competitive edge,” Seattle general manager Ron Francis says. “A lot of cities don’t have this tech base.”
The connections are already paying off. Not long ago, Mandrycky and Francis met for breakfast with Andy Jassy, a member of the Seattle Hockey Partners ownership group and CEO of Amazon’s cloud computing subsidiary, AWS. “Just to pick his brain,” Mandrycky says. “When we talk about how do we want to build our team, who are the types of people that we should be hiring, what sort of technology should we be looking at?”
They also attended sessions of a monthly gathering of area tech executives, discussing the team’s plan to utilize data in the pursuit of a Stanley Cup. “Machine learning, artificial intelligence—every team in every league is looking at those,” says Todd Humphrey, senior VP of digital and fan experience for NHL Seattle, and a regular member of what he dubs the “innovation and efficiency” committee. “So how do we leverage new tech to make the best sense out of data? As the new kid on the block, a lot of people have been willing to help.”
At least part of the enthusiasm is thanks to the man in charge. “They’re sports fans,” Humphrey says, “so when they see Ron Francis in the room, their eyes light up.” A reliable two-way centerman whose 23-year Hall of Fame career ended in 2004, Francis refers to himself as an analytics “dinosaur, in the sense that it wasn’t around” during his playing days. But he always enjoyed math in school, aspiring to become an accountant or financial representative if hockey never panned out. “If I was going into a contract year, I’d calculate numbers to see what my case looked like against others,” Francis says.
Decades later, as GM of the Hurricanes, Francis created one of the NHL’s first full-time analytics positions in August ’15 by hiring Eric Tulsky, a nanotech chemist and popular hockey blogger who now works as Carolina’s vice president of hockey management and strategy. After leaving the Canes—and following a brief stint in commercial real estate—Francis popped onto Seattle’s radar as a forward-thinking candidate whose past fit their mission. Even so, though, the franchise needed to make sure. “I interviewed with eight different owners,” Francis says, “and it was fairly blunt. Like, ‘Where are you on analytics? What’s your interpretations?’”
Appropriately enough, Francis wasn’t even the first hockey ops employee on Seattle’s staff.
Mandrycky didn’t take to the sport until college, when she began attending Thrashers games while at Georgia Tech. After earning her industrial engineering degree in 2013, though, she accepted a travel-heavy gig as a software consultant, averaging two weeks out of each month on the road. “Instead of going out to the bar at night, I would sit in my hotel room and work on hockey stats,” Mandrycky says. “It was an easy vehicle to improve my programming and statistical method skills and enhance my non-hockey career.
“Then ironically it ended up turning into my career.”
The following year, Mandrycky linked up with then–Carnegie Mellon professor Andrew C. Thomas and Ph.D. student Sam Ventura to help launch the advanced stats website War on Ice. It was a pivotal time for the hockey analytics community, which had just seen Extra Skater founder Darryl Metcalf join the Maple Leafs in the first of what became many such blogosphere-to-boardroom hires. Sure enough, Thomas and Mandrycky had scored gigs before long too, both as analysts for the Wild. “I remember Andrew sent me a note one day, and he was like, ‘I’m talking to Minnesota and I told them they should talk to you,’” Mandrycky says. “I thought that he was kidding. That was never the motivation, never the goal.”
And yet look at her now. After four seasons in Minnesota, Mandrycky leapt at the chance to start an entire analytics department from scratch in Seattle. (It certainly didn’t hurt that her husband, Christian, has been working towards his doctorate in bioengineering at the University of Washington.) “From an analytics, R&D perspective, I would expect that we’re, as a team, one of the teams at the forefronts of the league in working with data,” Mandrycky says. “We certainly hope to be on the cutting edge as tracking data comes onboard with the league. We see that as an advantage for us, to be able to work with that early. But at the same time, we’re not just hiring people to say we have a certain number of hires. We’re going to do it very strategically.”
Indeed, the first puck won’t drop at KeyArena until October 2021, an extra season of runway compared to Vegas, but the front office is slowly taking shape. Five pro scouts were added in September to join Mandrycky, Francis and assistant GM Ricky Olczyk, and amateur scouts won’t be far behind. Plus, the team recently filled one of the three positions that Mandrycky posted, bringing aboard Dani Chu as a quantitative analyst.
“I look at this year as our demo, trial year,” Mandrycky says. “As we create reports, or as we’re talking about different player evaluation metrics, we can pass those on to our scouts and management and think about what we want to do moving forward. As we get into that second year and really have a full staff and we’re preparing for the expansion draft, that’s phase two. I expect we’ll have things more ironed out then.”
No doubt that Seattle is ready for whatever Mandrycky has in store. In late September, she appeared along Francis and Olczyk at an event titled Hockey Analytics: The Numbers Behind the Game. After a panel discussion, the trio took questions from fans for 45 minutes, ranging from “How do you specifically evaluate goaltenders?” to “How do you plan to use machine learning?”
Noticeably absent, however, was one popular query. “The one we always get is: What’s the team name going to be?” Humphrey says. “And we did not get that.”
The fans, it seemed, were just too focused on the numbers.