Tampa Bay Lightning

Statistical Analysis Helps Lightning Organization Navigate Free Agency

Wednesday, 06.27.2012 / 10:24 AM / Best of the Web
By Alex Peckham
X
Share with your Friends


Statistical Analysis Helps Lightning Organization Navigate Free Agency
Moneyball: A look at the Lightning\'s approach to statistical analysis

The book that sits most prominently on the desk of Lightning Statistical Analyst Michael Peterson is Moneyball, the story of Oakland A's general manager Billy Beane and how he assembled a winning team despite his organization's minute payroll through an analytical, stats-based approach.

Beane's work revolutionized baseball; changing the way players are valued, teams are constructed, and prospects are scouted. This sabermetrics movement soon had a similar effect on other sports, but as of now, hasn't found its way into NHL front offices.

While this season, the New York Mets became the last team to employ statistical analysis in Major League Baseball, Peterson and the Lightning organization are helping to pioneer the field in hockey, where most scouts still use the same methods as in years past and trust their instincts over numbers.

Peterson came to the Lightning as an unpaid intern in 2008, but within months became a full-time member of the club's front office. He now has an important role to play in nearly all aspects of hockey operations, and is frequently consulted by general manager Steve Yzerman and the coaching staff for his unique take on statistics, and how it might impact free agent decisions.

"I'm somewhat old-school when it comes to evaluating process," Yzerman said. "You ultimately watch players on tape to form your decisions, but with Michael, he takes it one step further from a statistical point of view, creates analysis and uses it to reaffirm each and every decision we make on a player. The more information we have, the better decisions we can make, and it is also useful in that it makes us go back and look at something different that we might have missed the first time around."

His focus is on the development of advanced statistical metrics that can tell the true value of a player and his production rather than traditional stats like goals, assists, and hits, which can be subjective and ultimately deceptive in judging a player's effectiveness.

Advanced stats in hockey, very much in their infancy, "can't measure a whole lot right now, but in the next few years, they'll be more evident and useful," says Peterson.

Peterson's methods aren't yet widely used in the NHL and conflict with those of many NHL scouts who believe the best way to value a player is to watch him play and make judgments largely based on intuition. Despite at times being skeptical of Peterson's advanced statistics, Yzerman has made him a key cog in the Bolts' front office and consults him regularly for information on a player.

"Steve's a good manager because he wants to get as much information as possible before he makes a decision," Peterson said. "He's not going to exclude information he has access to. He really does take everything into consideration."

"It's useful to receive additional information," Bolts assistant general manager Julien BriseBois said. "It's a compliment to what our eyes and what our scouts' eyes find in terms of information. Sometimes it's like a double check on what our eyes are telling us."

Statistical Analysis further supports traditional analysis conducted by GM Steve Yzerman and his staff to gain a full understanding of a player's value.

For Yzerman and the front office staff, Peterson's job is to provide information on players the Lightning may be looking to acquire, but with Guy Boucher and the coaching staff, he is asked to analyze the current members of the team.

"I look at faceoffs for [assistant] coach Martin Raymond each game," said Peterson. " I've started measuring a couple of extra things just to see if there would be anything surprising about our own players, and it turns out the coaches have liked that."

With the NHL draft finished, the front office will now shift its focus toward the start of free agency. With so many tough decisions about who to keep, who to let go, and who to go after, all while weighting how these choices will affect the makeup of the locker room and eventually the product on the ice, Peterson's advice is highly valued by Yzerman and his staff.

While he is hesitant to call himself a trendsetter, Peterson does acknowledge the importance of the work he and his handful of colleagues around the league have done in the development of advanced statistics in hockey.

He graduated from Texas Tech University with bachelor's degrees in mathematics and computer science, along with a master's degree in mathematics, but at the time, had not considered the possibility of working in sports.

"I was starting a PHD program in math and I wanted to work for the government, but I stopped and thought about if that's what I would enjoy doing," he said. "I had always been interested in sports, so I decided at that point that was what I wanted to do."

What he eventually did was move on to the University of Central Florida, where he earned a master's in sports business. Despite initially not even considering hockey as a career path, Peterson decided about half way through his time at UCF that he did want to work in hockey because not much had been done at the time in the field of advanced statistics.

He says that people like Gabriel Desjardins, who runs the hockey blog "Behind the Net," will be the faces of this stats revolution because all of their work is public, while most of Peterson's own methods are kept within the Lightning organization. If his ideas can help the Bolts achieve success though, teams around the league will likely follow suit.

"What he does is a work in progress as far as advanced statistics go," Yzerman said. "It originated in baseball, but we in hockey are trying to figure it out, and Michael's trying to figure it out. What he does is study how else we can find value in a player, and for the past two years, it's been very useful."

Peterson's work is underappreciated around the NHL, he says, because "the game is so fast and fluid that it's hard to decide what needs to be measured as opposed to a sport like baseball where you don't even have to think about it."

"Advanced statistics in the other sports are pretty well developed too," Peterson added, "but I think within five or ten years, hockey will be fully caught up."