Before we get too far into the next NHL season, let’s look at an instant replay of game six of the 2008 Stanley Cup Final:
In those last 40 seconds, things get pretty intense. The Detroit Red Wings have the puck out of the zone, and Pittsburgh Penguins’ Marian Hossa is coming down fast on the ice, lobbing a backhander at goalie Chris Osgood. With only 10 seconds left Osgood moves, and the puck hits his arm. Before it even starts rolling over the side of the net, the horn sounds and the time is up. The Red Wings have won 3-2. The season is officially over, but this is where the software-based analytics begins.
Ryan Lilien doesn’t look like a hockey player. There’s neat dark hair instead of a mullet, and glasses instead of a facemask. As a graduate of Cornell University and someone who earned a Ph.D. as well as an MD, you might assume he’s more egg-head than puckhead. But when you talk to him about Canada’s favourite game, you’d also find it hard to believe he wasn’t born here.
“When I was in graduate school (at Dartmouth College), I got hooked on hockey,” he says. “When I came to the University of Toronto about two years ago, my focus was on computational biology, but this is Canada. I wanted to see if we could do something that could combine both interests.”
The result is a project called the Computational Analysis of Ice Hockey Gameplay. The goal is to develop a system that will learn how hockey is played and help a team improve their performance. Or better yet, help a team like the Penguins understand how Osgood knew just where and how to move in those last 10 seconds, and how they could outthink him.
“Tracking the puck is hard. It’s small,” he says. “As a hockey fan, you don’t always see the puck, but you know based on the position of the players where the puck is and where they’re moving. The computer could do the same thing.”
Lilien’s approach is to use “machine vision,” a subfield of engineering that encompasses computer science, optics, mechanical engineering, and industrial automation. Today it’s mostly used to monitor and inspect packaged goods in a manufacturing setting, but the U of T team project will apply it to video footage of NHL games. The software will study where players move, their habits and play styles. Specially-developed algorithms will then attempt to reason, under uncertain conditions, what kind of patterns emerge and what relationships they have to winning or losing a game.
“It’s not simply to say this shot was taken from this position, but what led up to that situation?” he says. The concept is not far removed from how enterprises use business intelligence (BI) software. Based on sales, customer requests and other information, they try to determine what kind of cause-and-effect relationships their actions have on a company’s growth, and adjust their strategy accordingly. Like hockey teams, they are studying outcomes. Or forecasting.
Real-time analytics
“Think of the draft,” says David Hatch, an analyst at Boston-based Aberdeen Group. “It doesn’t matter what sport you’re talking about. The clock is running, and you have to analyze several things at once. This is true, real-time operational BI.”
BI has slowly been making its way into sports for years. The New Zealand Blackcaps cricket team, for instance, started using a product called Enterprise Miner from SAS Institute Inc. four years ago to analyze matches and in some cases help make decisions during the innings. According to Tammi Kay George, SAS’s BI product marketing manager, it takes a sporting organization of a certain maturity and comfort level with handling information to use the technology successfully.
“I think sometimes we may build some imaginary barriers between the sports world and the business world in their attitudes towards analytics,” she says. “You have to approach it from the perspective of, ‘we appreciate the value of this and are willing to put it in the game.’”
At least one NHL veteran sees some advantages in using software on the ice and off. Steve Thomas played 20 seasons with the Toronto Maple Leafs, the Chicago Blackhawks and even the Red Wings, among several other teams, before retiring last year.
“A lot of assistant coaches spend time in the coaches’ room analyzing video from the games they’ve recently played with opponents,” he says. “They clip out a number of certain plays in regards to a power play. They’ll clip a penalty killing out and one on forechecking and they’ll re-show that video to all the players in the locker room. To be able to put it into a report that can easily be conveyed to the team as a whole would make the whole thing a lot better.”
Although Lilien says U of T hopes to be able to offer the final product to NHL teams, he’s not expecting it to guarantee victories. What you can do is lead the user in the right direction, giving them hypotheses they can validate, he says.
“It does fit with in a more traditional IT setting, in that you’re trying to learn behaviours,” he says. “You might want to have a policy in place to thwart potential attackers, but how do you know who they are? You could build an automated system that’s capable of detecting patterns through a security camera then associating those patterns with different actions that could take place, and send an alert before something happens.” Kaye says the biggest difference between BI and “sports intelligence” may be figuring out what variables to feed into the system.
“You can be limited by data. There’s information we’d like to know but aren’t gathering or don’t know how to get yet,” she says. “Have you just come off a losing streak? What was the temperature when you were playing?”
Hockey may be the litmus test for analytics and sport, Lilien says. “Football is easier in some respects. There are more players on the field, but every play starts with a configuration that is static. It runs and then it stops. You can very easily measure what has happened — maybe it was an incomplete pass or you moved the ball down the field. Hockey has only 10 skaters, but it’s continuous. There are no fixed plays, apart from faceoffs. A lot of these techniques we’re developing would be more applicable to basketball or soccer.”
Until then, however, all eyes — including the computer’s —will be on the ice.
