In the corner of a shared office on the fourth floor of PNC Arena, you’ll find a small desk with enough surface area to house a laptop and an extra screen, a keyboard, a mouse and a pair of books about statistical analysis. It’s the unassuming – and temporary, as a much more appropriately sized desk is en route – home to one of the more incredibly bright and insightful minds in the hockey analytics community: Eric Tulsky.
Tulsky recently joined the Carolina Hurricanes full-time as a hockey analyst, a role in which he will provide and analyze data to assist the hockey operations department and coaching staff.
“In a really broad, general level, my job is to make use of any data we can get and figure out how we can use it to improve our decision making,” he said. “On a daily basis, my job is to collect, assemble and analyze that information to provide useful input to people.”
After working part-time with the organization last season, Tulsky relocated with his family nearly 3,000 miles east. While cross-country transitions are admittedly hard, working with the team in Raleigh felt like the right move for Tulsky.
“There is a lot more that I will learn by being here, just being around the team and hearing about things that I can’t tell from my home in Berkeley about what they’re doing and why and what they’re trying to do and why. I’ll learn a lot of that by being here,” he said.
That communication is a two-way street, too.
“I’ll be more effective in explaining why I think the things I think. There’s a difference in getting the report by email and having a conversation with someone about something,” he said. “It’s easier to make sure that I’m delivering the things that are useful to people and why I think it should be useful to them. That’s inherently a conversation. I know people who work for pro sports teams remotely full-time, but I can’t imagine I’d be nearly as effective that way. I didn’t want to do it and not be effective. I want to do a good job, and to me, that meant coming here, learning what I can and being a part of it.”
And he gets to share an office with the Canes’ Assistant to the General Manager and Video Scout, Darren Yorke.
“That’s really the reason I wanted to move out here,” Tulsky joked.
“With Eric being here full-time, he’s able to spend more time looking at the data, which gives us more of a competitive advantage,” Yorke said. “Hockey analytics is getting more competitive, and everyone is looking for that edge. Having someone like Eric, who has always been on the forefront of everything, and allowing him to spend more time doing research and spending more time with Ronnie and the staff we have, only helps him become better.”
Tulsky and his wife, Karen Sosa, and their nine-year-old son Seth are settling into their new home in Chapel Hill, quite a change of pace from the West Coast.
“California sometimes felt crowded. I spent a lot of time in traffic every day. Here, there is a lush, greenness that we really enjoy,” he said. “We’ve met a bunch of the neighbors in our neighborhood, and everyone seems really friendly and nice. It’s been very welcoming, and we’ve really enjoyed that.”
From Chemistry to … Hockey?
Tulsky has been publishing his advanced hockey statistical analyses online since 2011. Prior to that? It was a bit of a different career path.
Tulsky holds a B.A. in chemistry and physics from Harvard University and a Ph.D. in chemistry from UC-Berkeley. His resume also includes a two-year, post-doctoral study at the Naval Research Lab in Washington, D.C., and a 12-year career in nanotechnology, in which his research enabled unique nanotech solutions to problems in DNA sequencing, solar energy, displays and energy storage. It would be accurate to suggest that Tulsky, prior to diving into the world of hockey analytics, was conducting life-changing research.
It’s not that his work in nanotech led to hockey – “They were separate tracks,” as he pointed out – but the two fields overlapped in an intriguing fashion.
“Both chemistry and sports analytics often have ambiguous and conflicting data. In both fields, it’s important … being able to maneuver through that and form an opinion,” Tulsky explained. “Know what the data can tell you and know what it cannot. I think that applies to both fields.”
Naturally, as a Philadelphia native, Tulsky followed the Flyers. He began reading hockey blogs, including Broad Street Hockey.
“They were one of the early blogs to make use of statistical analysis in their evaluations. I stumbled onto it then,” he said. “It was not the beginning of the analytics movement; people had already done the hard part of just figuring out the basic direction things ought to go, but it was early enough that there was still a lot left, low-hanging fruit, that needed to be done. I found myself with time and interest and just started writing stuff.”
Tulsky also contributed to NHLNumbers.com before launching his own SB Nation website, Outnumbered. His hockey analysis work has also been published by The Washington Post and fivethirtyeight.com, and he has been a featured panelist for MIT’s annual Sports Analytics Conference.
“It was generally well-received, and I found it being a bigger and bigger part of what I was doing,” he said. “I was putting more time into the analysis and eventually started to work with teams. It wasn’t something I expected going in. It was just like, ‘Hey, this looks like fun,’ and here we are.”
“Here we are” has progressed from Tulsky self-publishing his work online to working with a handful of teams on various one-off projects to working part-time with the Hurricanes in the 2014-15 season to today, a full-time gig in the Canes’ front office.
Advanced stats. Analytics. Fancy stats. Whatever your moniker of choice, these metrics have begun to seep into the mainstream hockey conversation. In a partnership with SAP, NHL.com provides “enhanced stats” on its web site.
Enhanced stats and their evaluation in this increasingly digital and information-heavy age seem to be a perfect marriage.
“Our organization has been using analytics for years now. What you’re seeing now is that it’s becoming more public,” Yorke said. “Whether it’s been Jim (Rutherford) and now with Ronnie, we’ve always looked at ways to become better. Bringing on someone like Eric allows us to become better.”
“Every sport has had statistics since the beginning of time. The only thing that is really different now is people are hiring or talking to people with more analytical backgrounds on evaluating which statistics matter most and why,” Tulsky said. “When people started reporting box scores in baseball, they wrote down the things they thought would be important, and they’re right about some of it; some of the things are more important than they thought, and some are less important than they thought. That’s something that’s really hard to pick out with your eye; it’s a lot easier to do analytically. That’s a big part of an analyst’s job: to making sure you’re looking into the right things. It’s sort of a double-check on your intuition about what can be important.”
Information or statistics deemed important can then be passed on to the appropriate parties, and with Tulsky based in Raleigh, that conversation happens more efficiently. Though the Academy Award-nominated film “Moneyball” may pit traditional ideas (scouting) against unconventional ways of thinking (analytics), the relationship between the two is much more symbiotic, and the end goal is much the same: be better.
“It’s not my job to replace a scout or a coach or anyone; it’s my job to help them be better at their job,” Tulsky said. “What ended up happening in baseball in a lot of places was that analysts were able to start identifying things that were important to watch for, and then the scouts and coaches started watching for those things and developing those things. The scouts are still better than anyone else at telling who is good at a specific skill. They just needed some input on which skills were most important to look for. Neither of us is any good without the other; the goal is for me to help them be better at their job, and not in any way replace them.”