March 06 2012 05:10PM
At the Sloan Sports Conference over the weekend, former Canucks General Manager Brian Burke cast doubt on the usefulness of applied analytics in hockey. Burke was quoted as saying, "Statistics are like a lamp post to a drunk, supportive but not very illuminating," that "moneyball doesn't work for hockey" and he even described advanced stats and "moneyball" concepts as "aggravating."
In general, Brian Burke's term as the general manager of the Vancouver Canucks was solid, albeit marked by his failure to surround an exciting young team with competent goaltending. The one area in which he was absolutely exceptional, however, was his acquisition of top-end talent. In the Sedin twins, Kesler and Bieksa - Burke managed to amass an impressive array of top-end talent, and that core has been essential in transforming the Canucks from the "LA Clippers of the NHL" into a top-team.
For all of Burke's success, however, I couldn't help but cringe when I read his description of how Burke decided to draft Ryan Kesler in the 2003 NHL draft. This tidbit comes from Scott Cullen's excellent recap of the conference:
Burke also said that, when he was with Vancouver, he decided he liked Ryan Kesler after seeing him play one shift at Ohio State, telling his scouts he had seen enough and that they should do whatever they needed to get him in the first round.
That's outrageous! I mean, we've seen Kesler have some pretty dominant shifts in his time with the Canucks, but that must have been some demigod-like stuff that Kesler showed in that game.
Though Burke's story makes me shake my head, it's fair to say that the Canucks have reaped the benefits of his impulsiveness for seven seasons now. Kesler has become a perennial Selke nominee, and one of the leagues best two-way forwards so clearly Burke nailed that pick. But results aside: drafting a player in the first round based solely on ninety seconds of ice-time is absolutely absurd.
I'm a believer in the power of observation to evaluate talent, especially if the observer in question is Brian Burke. That said, his Kesler story is undeniably extreme, and as a long-term drafting philosophy: relying solely on your observations over the span of a tiny sample is a loser proposition. Sure, every now and then you might strike oil, like Burke did when he picked Kesler. On the whole, however, evaluating talent this way will yield decidedly mixed results.