Maple Leafs take next step in analytics movement by teaming up with SAS

TORONTO – When Kyle Dubas began using advanced stats and analytics as general manager of the Ontario Hockey League’s Sault Ste. Marie Greyhounds, he and his staff had to track, import and compare everything manually.

“That would take hours and hours, and there would be lag time between games,” Dubas said. “We didn’t have any budget to do anything good enough to make it real-time.”

Now, as assistant GM of the Toronto Maple Leafs, Dubas has just about all the information he could want at his fingertips. Over the past couple months, the Leafs have forged a partnership with analytics firm SAS and on opening night began using the technology they hope will eventually improve their on-ice results.

“We’re able to collect the data and have it in real-time, be able to use it to give to our coaches in the game and to analyze for ourselves in decision-making,” Dubas said. “To have those tools at your disposal on your phone or iPad, it’s been great and a massive and welcome change.”

The SAS technology captures stats from every NHL game and team. Dubas and Toronto’s hockey research and development staff can then access that information, which means Leafs analytics specialists Cam Charron, Darryl Metcalf and Rob Pettapiece — all hired over the summer as part of Dubas and president Brendan Shanahan’s shift to a more analytical approach — don’t have to manually keep track of things like shot attempts, ice time or quality of competition.

Meanwhile, the Leafs track their own stats — proprietary information they won’t share — which is also inputted into the new system to be cross-referenced later. The result is a series of graphs and charts that attempt to explain aspects of the sport that go beyond traditional statistics like goals, assists and plus/minus.

“When you start to use a deeper form of analysis, it kind of just allows you to question what you’re seeing and allows you to better understand what’s actually happening in the game, which players will play well together, which defence pairings will play well together,” Dubas said. “It encompasses around everything because it adds context and powerful information to everything that we can do.”

SAS analytics lead Tim Trussell said the purpose is “to be validating assumptions or validating hypotheses in an unbiased way.”

Hockey executives, coaches, players and fans can watch a game and have a good idea of who’s playing well and who’s not, but having the data to back it up adds another dimension to measuring hockey.

Carl Farrell, SAS Canada’s executive vice-president, said several other teams are doing similar things. SAS has worked with the NBA’s Orlando Magic for some time.

“Sports analytics now is growing in all the different sports segments and hockey being one of them,” Farrell said. “I think you’re going to see more of this. People are realizing: This data’s important. It brings extra context, extra depth to the kind of decisions that Kyle and the team have to make.”

Dubas said when he was hired, Shanahan and general manager Dave Nonis were eager to learn as much about analytics and obtain as much information as possible. Over the previous couple of years, the Leafs were a punching bag for the advanced-stats converts, who criticized them for being a bad puck-possession team.

Centre Nazem Kadri has noticed more of an emphasis on holding onto the puck this season than in the past.

“I think they want you controlling the puck a little more and teams have really started to do that,” Kadri said. “Obviously there’s times and places where you’ve got to dump the puck and try and get retrievals, but really it’s all about the neutral-zone and offensive-zone possession.”

In an effort to join some of the better teams in the departments of analytics, puck possession and sustained winning, the Leafs hired stats bloggers Charron, Metcalf and Pettapiece.

“They’re watching the game and tracking different things that we’re doing either that we’ve keyed on and want for every game or different things that we’re trying to test to see if there’s any correlation between how we play and our ability to score more goals than the other team,” Dubas said.

The SAS partnership, announced Thursday morning, seems like the next step in the process for the Leafs.

“Over time it’s not so much the technology that differentiates. It’s how an organization uses it, how they incorporate it into their business process,” said Cameron Dow, vice-president of marketing for SAS Canada. “It’s all that softer stuff: How do you change the culture to be more analytically oriented, those sorts of things, is really one of the keys to unlocking the value.”

Part of changing the culture is taking ideas and information and selling it to the coaching staff. Dubas said that was one of the more difficult things in Sault Ste. Marie.

“You can’t just come in and be abrasive and try to overnight have everybody just nod in agreement with everything that happens,” Dubas said. “It just adds context and information to what they’re doing as coaches and try to be helpful to them.”

Coach Randy Carlyle said it could take some time to put into effect what the analytics data says.

“There’s been information that’s been supplied to us that’s helped us, and we’re going to continue to use it as a tool,” Carlyle said. “We think we’d be crazy not to. If there was no value in it — in our minds there is value in it, so we’re going to use it.”

So far and through the very small sample size of four regular-season games, Dubas said, the most noticeable impact has been on analyzing upcoming opponents. The Leafs could begin to anticipate how the Pittsburgh Penguins were playing under new coach Mike Johnston or how new players were fitting in for the New York Rangers and Colorado Avalanche.

“You’re trying to analyze, even though it’s a small sample: How are those teams changing or how are they playing, what effects are different guys having with different players, and be able to apply that to our preparation for every game,” Dubas said.

So far, only Dubas, Charron, Metcalf and Pettapiece have access to the raw data, though that could change over time. SAS also has the ability to add more depth to the information the Leafs can unearth and sift through.

“We see a lot of potential in the future to add and complement to what Kyle and the team are doing right now,” Dow said. “I think they’re just scratching the surface of what is possible.”

Integrating analytics is part of a long-term process for the Leafs, Dubas said. And unlike junior hockey with its high turnover, the NHL has components like a salary cap and roster construction that mean any such turnaround like the one he led with the Greyhounds will require patience.

Dubas already got a taste of impatience when the Leafs lost their first two games.

“Being an active social-media user I was fascinated: Saturday night we lost and went to 0-2 and it was hilarious to kind of watch everyone say, ‘I thought analytics was supposed to make the team an overnight success,'” Dubas said. “The key thing for me is I’ve lived through it before in Sault Ste. Marie. We went from doing nothing to collecting information and data and tracking it and integrating it into the organization. It takes time.”

Kadri spent some time learning more about advanced stats but hasn’t changed his style.

“I’m a puck-possession player, so it’s not like I’m opposed to it,” Kadri said. “I’ve seen charts and things like that and graphs and all that. I’m not a math major by all means. I just go out there and try to play my game, and whatever the numbers show, they show.”

Follow @SWhyno on Twitter