Are hockey fans ready for advanced stats?
By Allan Muir
"I like him. Great shot, quick release, high compete level, and his Fenwick and Corsi Relative numbers are off the charts!"
-- Something no one in hockey has said to me. Ever.
Bruce Dowbiggen wrote a solid column on Monday for the Globe and Mail about a missed opportunity to explain the true value of Vancouver's Manny Malhotra in the wake of his apparent career-ending injury.
Now, anyone who has paid the Canucks a lick of attention could point to Malhotra's success in the face-off circle and defensive prowess as attributes that the team will miss. And honestly, for a fourth-line center who averaged about 11 minutes a night, that might be as in-depth as the eulogy needed to go.
But Dowbiggen, a vocal proponent of the use of advanced analytics in hockey -- let's just call 'em "fancy stats" -- saw a chance to weave a more nuanced yarn.
He pointed to a fancy stat that highlights zone starts. It painted exactly how effective (and nearly exclusively used) Malhotra was in his role, and from that Dowbiggen was able to correlate how Malhotra freed up the Sedins to save their energy for the offensive zone. And that, he argued, gave a more accurate assessment of Malhotra's impact than the "amiable banter" most television analysts offered.
Dowbiggen might have a point. Hockey has traditionally been an under-analyzed sport. It's only within the past 15 years that the NHL began officially recording stats such as giveaways, takeaways, blocked shots and hits. To this day, things like zone time, passing accuracy and puck possession are concepts left to individual teams to track, if they're so inclined.
That data vacuum led to a new generation of homemade and highly unofficial stats that purport to reveal hidden truths, as well as give fact-based support to concepts that originated in the gut.
One of the most often quoted is the Corsi, which counts all shot attempts-for (on net, blocked and missed) versus shot attempts-against as a proxy for time of possession. (More shots, the theory goes, more control.) That number can then be broken down to show a particular player's effectiveness in relation to his teammates.
It's kind of interesting, right? And by that I politely mean, it's kind of...dense. As in I can't look at it and instinctively figure out what it means, despite being around hockey virtually my entire life.
And that, to me anyway, is a problem.
Maybe I'm not the right audience for these numbers. I mean, there's a reason why I'm a writer. The most complex math I'm willing to do these days is calculating the tip on my bar tab.
But the fevered proselytizing of highly respected writers like Dowbiggen and James Mirtle and countless bloggers notwithstanding, these calculations might just be the greatest product that relatively few people want, the Freaks and Geeks of hockey statistics.
Sure, you can imagine how these might be of use to agents and general managers as they look to quantify a player's value in terms that allow them to increase, or suppress, salary demands. And there may even be some Moneyball-type angle that allows a team that employs an analytical ace to get a leg up in trades and free agency.
But beyond that, the numbers feel alien. Many hockey fans still think of Gary Bettman as a basketball guy, which suggests that influence from other sports is not warmly embraced. Why then would they welcome something that reeks of Astroturf and peanut shells, like baseball's anal obsession with chronicling the full measure of every moment?
Baseball is different. It's a 162-game death march built around four-hour contests pregnant with action-less pauses that cry out for something to break the monotony. Fortunately, the individuality of virtually every play makes it ideal for this kind of analysis.
That's not hockey. The presence of 12 moving pieces makes it impossible to deconstruct as precisely as baseball.
And while Dowbiggen calls this data "manna from heaven," I'm thinking the vast majority of fans are just comfortable with the old Howie Meeker "Stop it right there, OK, watch what Trevor Johansen does here" style of analysis. It's really all they need.
But I'm willing to admit that I could be wrong.
"I don't know if I buy into [advanced stats]," SiriusXM Home Ice host Jim Gordon said to me on the air last week. "I just don't want to be caught on the wrong side of history on this."
That's probably the judicious approach. It's easy to brush these calculations off as message board grenades lobbed by basement dwellers who couldn't tie their own skates on a bet, but there's no reason to go full-on Luddite here without giving these concepts a chance to prove their merit.
It won't be easy to get past the public's aversion to numbers that aren't easily digestible -- it might help if they were branded in a way that was explicit about what they represent rather than naming them after a blogger who whipped up a formula -- but anything that can give us a better understanding of how the game is played should be welcomed by more than just a hardcore minority.
I'm still conflicted. And I'm curious. Do you want this kind of statistical analysis? Have you been using it on a regular basis? Should the media? Has the time come for the league to commit to capturing a broader statistical view of the sport? Or are the postgame highlights and goals and assists all you'll ever need?The floor is open. Comment away.