Home Sports Statistics: Hockey’s best friend

Statistics: Hockey’s best friend

by Alexander Cole November 14, 2017 0 comment
statistics

The importance of sports analytics were on full display at this year’s JMSM conference

Advanced statistics and hockey analytics were a hot topic at the John Molson Sports Business conference held on Nov. 4 at Hotel Bonaventure.

A panel, titled “Grit and Character: The Evolution of Analytics,” was moderated by Concordia journalism graduate Salim Valji, and featured four guests well-versed in the sports analytics world.

Panelists included Sportsnet writer Dimitri Filipovic; TSN, Sportsnet and Vice Sports contributor Andrew Berkshire; Stathletes co-founder Meghan Chayka; and hockey analyst Mike Kelly.

The talk began with each panelist discussing how hockey analytics have changed over the last few years. In the past, goals, assists and plus/minus were ways to track whether or not a player was performing well. However, new statistical measurements, such as Corsi, Fenwick and PDO, have created new ways to gauge a player’s effectiveness on the ice. Corsi and Fenwick are similar, as both measure how many shots a player takes in relation to everyone else on the ice, while PDO keeps track of a team’s shooting and save percentage.

However, one of the challenges with advanced stats is properly recording what is happening.

“A lot of it is going to come down to video player tracking,” Filipovic said. “Zone entries and loose pucks aren’t something we can quantify right now.”

Due to advanced stats being relatively new, the panelists said players don’t have much respect for advanced analytics. Valji referenced an article in which Montreal Canadiens captain Max Pacioretty said he can’t judge a player based on a pie chart.

Berkshire added that stats like Corsi can make people “roll their eyes,” but that doesn’t mean they should be dismissed. Filipovic agreed, saying a team’s management should care about the new analytics.

“You want to look at it from a manager and coach level,” he said. “Why bring it to the players if they don’t care that much?”

For players to understand their own advanced stats, Chayka suggested the use of visual elements be emphasized rather than the actual numbers.

“Heatmaps and visual maps of the ice are great for showing players their performance,” Chayka said. “You have to know your audience.”

The panel then talked about how a player’s grit and character is different from their stats. Kelly brought up former Ottawa Senators forward Chris Neil, and how he isn’t an effective player on the ice, with 112 career goals in 1028 games, but his grit, character and presence in the locker room helped the Senators win games.

Chayka and Berkshire disagreed, with Chayka stating that because the Senators didn’t win the Stanley Cup with Neil on the team, it doesn’t matter that much. Berkshire added: “If character and leadership actually mattered that much, the Habs wouldn’t be terrible right now.”

The panel ended with each member giving some advice to students on how to make it in the industry and how to be effective at delivering analytics to an audience that might not otherwise know what you’re talking about.

“Make what you’re doing as relatable as possible to those consuming the information,” Kelly said. “The ones who are the most successful are the ones who text you at 10 p.m. with fresh ideas.”

Main photo by Alexander Cole

Related Articles