analytics

Moneyball-ization of hockey pays off for Penguins

TORONTO — The Globe and Mail

Pittsburgh Penguins James Neal comes in on a breakaway against Toronto Maple Leafs goalie Jonas Gustavsson (R) during the third period of their NHL game in Toronto, October 29, 2011. (© Mark Blinch / Reuters/REUTERS)

It’s a deal that, nearly two years later, looks like a true no-brainer.

But when the Pittsburgh Penguins went through a lengthy search for a power forward leading up to the 2011 trade deadline, the organization was split, with a handful of candidates in front of them and arguments for and against from various members of their scouting staff and management.

Story continues below ad

And that’s where the analytics came in.

Speaking at the Predictive Analytics World Conference in Toronto on Thursday afternoon as part of a panel on hockey statistics, Penguins director of player personnel Dan MacKinnon explained that the team’s trade for James Neal was the first time the organization referenced the work of a company called The Sports Analytics Institute before pulling the trigger.

The end result of that unique behind-closed-doors process has been widely visible on the ice ever since, as the player they eventually chose has scored more goals since the start of the 2011-12 season than all but Tampa Bay Lightning star Steven Stamkos.

“I don’t think we’ve made an impact decision since then without consulting the analytics,” MacKinnon said. “I’ll put it that way.”

While Neal was just 22 years old and viewed as a rising talent when the Dallas Stars moved both he and Matt Niskanen to the Penguins in exchange for defenceman Alex Goligoski back in February of 2011, what tipped the balance for Pittsburgh in terms of going for Neal over anyone else was what MacKinnon calls his “conversion rate.”

And even though the trade was ultimately a huge win – with Neal scoring 40 goals and being named a first-team all-star in his first full season – MacKinnon admits there were anxious moments when he didn’t provide immediate results.

“We made the deal and in the final 27 games of that season he scored a total of two goals,” MacKinnon said during his presentation. “I remember just like it’s yesterday having this conversation with my GM – and, believe me, scouts have been fired for less in the business – and he said ‘you know, Dan, everyone likes James Neal. He plays hard, he hits guys, but he scored two goals.’ But he never got to play with [Evgeni] Malkin and [Sidney] Crosby [who were injured at the time]… and I said, to truly evaluate this guy, we’re going to have to give him time playing with these centremen.”

What had set Neal apart for MacKinnon was his ability to produce goals at a high rate based on where he was shooting from, something SAI analysts Mike Boyle and Kevin Mongeon felt meant he could score far more often if elite players were getting him the puck in better areas on the ice.

“It looks like a good deal now, but at the time, out of all the possible players to get, it wasn’t that simple,” Mongeon said.

“They had other choices,” Boyle added. “Neal wasn’t necessarily the most obvious.”

SAI’s analysis relies on breaking the offensive zone into sections based on the probability of scoring from those areas and weighting for other factors such as what type of shots players are taking.

The use of shot-quality data is still hotly debated within analytics circles and some of the concern is over the accuracy of the league-tallied location information, which can vary from building to building. Much of the other advanced statistics work being done, both for teams and independently, is more focused on puck-possession metrics that use shot attempts for and against to measure the amount of time teams and players spend in the offensive zone.

Boyle and Mongeon, however, have found converts for their “predicted goals scored” system with the Penguins and one other undisclosed NHL team they said is also among the best in the league.

And a lot of their success making inroads in the typically old-school NHL came as a result of the Neal trade.

“To see the first deal we really used analytics with come to fruition the following year was just a win internally where it led us to more comfortably work them into future discussions,” MacKinnon said, adding that signing backup netminder Tomas Vokoun was a more recent move influenced by SAI’s metrics.

“It gave them a sort of a confidence,” Boyle said.

While other organizations using advanced statistics continue to keep their work top secret, the Penguins have taken the stance that the more they talk about what they’re doing, the better the available analysis and data will eventually be.

Boyle, meanwhile, notes that their anecdotal evidence reveals the better teams in the league are following Pittsburgh’s lead by applying Moneyball-type principles developed in baseball to hockey.

“As far as we can see, the clients we work with as well as the organizations who call us back and ask the right types of questions happen to be the teams that are in the upper ranks of the league,” Boyle said.

“We’re just trying to be an early driver on this and get a bit of a leap on the rest of the league,” MacKinnon said. “Maybe we can be a little bit better than the competition at every turn.”