November 13 2012 03:27PM
We know that Brian Burke doesn't have an eye for analytics. None of what follows in this short post is the suggestion that Burke drop everything and base his player decisions based on things found on The Leafs Nation or NHL Numbers, particularly if it meant we weren't going to collect royalties.
Still, things said by Burke at a panel today attended by friend of our blog and every blog Justin Bourne sort of irk me:
Burke on analytics: "I don't think they have any value at all." Playing role of stodgy old man beautifully today. Sort of tongue-in-cheek.— Justin (@jtbourne) November 13, 2012
Burke has an excellent character, and can be cast in this story the way Grady Fuson was in the movie adaptation of Moneyball, the gruff old scout yelling about how computers can't change the way he thinks. When people first hear about analytics, without fulling getting the picture, it's like they're immediately placed in one of two camps, either "fer" or "aga'nt". I think the best among us drift to the centre and see value in both analytics and traditional viewing methods.
None of Burke's previous words about Moneyball make me think that Burke is an advocate of extra analysis, but I also don't see how the above quote can be construed into Burke identifying Tyler Biggs as a future NHL star because of the way his girlfriend looks. The "stats vs. scouts" debate is a fun narrative, and we do like pointing out the success of some analytical methods, and I always like when a team releases internal information about a player. The best recent example is the Carolina Hurricanes signing Alexander Semin last summer:
Semin's % rank among forwards, last 4 seasons even strength: Goals per 60 Minutes High Pressure situations: 99, 96, 99, 90— Mike Sundheim (@MikeSundheim) July 26, 2012
Now, there was a lot of debate over Semin. "We've heard stories about him," said Jim Rutherford, a few days before the signing. Semin's a guy whose teams do so well with him on the ice, but he's regarded as a liability because of the way he acts around teammates. For Burke, who stresses character when building a team, he's simply shortening his supply of good players who could help him by not assessing all the above information.
I think the balanced approach taken by the Hurricanes was a smart move. They did their research on Semin and determined the good outweighed the bad, particularly if he was on a shorter deal. They balanced their use of analytics with traditional methods of player evaluation, presumably, and made a decision that was probably more informed than many teams'. Perhaps some teams did similar analyses and determined the bad outweighed the good, focusing on different methods. Whatever, they're striving towards information, not necessarily analytics. Scouting information is also important information.
But the balanced approach is sometimes ignored, and I think that a lot of us are too quick to put teams into a "scouts" or "stats" category. A lot of baseball writers love to guffaw when a team that has an executive who talks about VORP stats finishes in last place. Baseball, the sport where critics of hockey analysts suggests best lends itself to advanced analysis, has an even larger debate than hockey, given the scope, its tradition, and the complexity of some of the new numbers being used.
This is not the spectrum:
This is my educated guess at the spectrum:
He makes the point that perhaps in 2002, the above graph was closer to the truth, but as teams expanded their scouting and research staffs in a quest for information, the lines have blurred. The Giants' general manager Brian Sabean is revered as an old school prophet, while the Giants website lists Yeshayah Goldfarb as their "Director, Minor League Operations/Quantitative Analysis" ("he's one of our 'Moneyball' guys, if you will," Giants president Larry Baer said).
Gabriel Desjardins, who updates 'Behind the Net', a leading resource for advanced hockey analysis, has joked that the Los Angeles Kings are "the first team built on Corsi and Qualcomp". He set out a list of five principles that the Kings followed on their way to capturing the Stanley Cup. Executives with the Vancouver Canucks and Phoenix Coyotes have also expressed interest in finding new ways to get ahead. While neither team has won a Cup like the Kings, the Canucks have won more games than any team since Mike Gillis went on board the summer of 2008, and the Phoenix Coyotes have made the playoffs three straight seasons on a shoestring budget.
Again, it's not like Burke needs to fire his scouting staff and spend the rest of the offseason reading old posts at Vic Ferrari's blog, but he should probably add to his scouting staff and also read old posts at Vic Ferrari's blog and see if there's a way to reconcile his old school, business-savvy ways with some modern analysis, and shed some new light on an old problem.