Interview with Globe&Mail: Hockey's Numbers Game
by Hawerchuk on Mar 27, 2010 10:33 AM EDT
Email
PrintHere's a link to the interview. My quote:
"In baseball, you had Neanderthals working as GMs, and they were ignorant of a lot of things that helped their team win. Hockey teams, by and large, do know what they are doing. Coaches have a very good understanding of their players."
And I firmly believe that. The inefficiencies in baseball were massive - 15 years ago, not one major-league team thought Matt Stairs had any value. And he was hardly the only valuable player who was freely available. Those inefficiencies closed quickly as a handful of general managers began to understand the obvious things that Bill James had written about in the 1980s.
But hockey? Not so much. Coaches and GMs in hockey have been systematically recording important on-ice events for decades. We don't know a whole lot about what goes on inside hockey, but Roger Neilson appears to have pioneered quantitative scouting in the 1980s: watching video and recording the important things that go on during a hockey game. We're talking about subtle things - not just the real-time stats that the NHL publishes, but rushes, passes, defensive miscues, scoring chances. Everything.
In baseball, you could sit down with a spreadsheet for 20 minutes and prove that on-base percentage was the most important thing in the game. What are you going to tell a hockey coach who has already been breaking down video for decades?
12 comments
|
0 recs |
Do you like this story?
Comments
" What are you going to tell a hockey coach who has already been breaking down video for decades?"
Something along the lines of: “Can you show some of that to me? Pretty please? I won’t tell anybody, I promise!”
The big difference between hockey and Baseball is the available data. Most of the significant data recording is made by teams and has been going on for quite a while; Darcy Jennish’s book about the habs related that, even tough they started using plus-minus in the 50s, they actually started systematically recording on-ice events in the 40s. My guess is every one of those original-6 had some stuff going on and it seems the 60s habs had not only a pretty large organization but had pretty extensive data collection processes; Nielson, after all, started as a coach for a habs affiliate team in the late 60s.
Yet despite this, there’s still the statistical dead-weight of the goon in hockey. Your anti-Matt Stairs.
Bettman's Nightmare: A Blog Where Hockey Aficionados Dismantle That Mighty Empire, One Balsillie at a Time
http://bettmansnightmare.blogspot.com/
by Bettman's Nightmare on Mar 27, 2010 11:40 AM EDT reply actions
And many teams are going without goons these days.
I agree totally about the stupidity of the average person. Horrifyingly, nearly 50% of people are even stupider!
While others blow third rounders on them.
Not bitter.
SNN Sports - A theoretical Oilers blog (i.e. theoretically, I write stuff there). Link now 100% less broken.
I could have sworn I read an article one time talking about the Red Wings and their use
of statistical consultants or analysts. Yet, they were not even mentioned in the article. I must be way off, though.
Interestingly, when Mike Gillis was hired a couple of years ago, he talked about finding the “marines,” players who don’t get much attention, aren’t necessarily the biggest or most offensively talented (boxcars), but who played hard and were extremely effective.
To extrapolate, what I think he meant was they are better bang for buck. He also said he would look for players who were underutilized by their teams. I think of the various players he’s brought in, Ryan Johnson and Mikael Samuelsson epitomize such a player. Johnson has been injured a ton, which is a problem with players who play so hard, and esp block a ton of shots. Samuelsson, I think, has lived up to his salary (top shots on goal on Canucks, 30 goals, plays next to Kesler vs hard competition, high GVT, etc).
But my point is, with Kesler’s recent signing, there were some great blog posts on CoppernBlue and MC79 on him, where the discussion led to evaluating the 2M Gillis signings (Burrows, Samuelsson, and Bernier). Interestingly, Burrows created the most debate, but almost everyone wrote off Samuelsson as a waste of money, using the usual stats. But your post makes me think that the evaluation of players might be more sophisticated than what we fans have, even the readers of sophisticated stats-heavy bloggers.
Interestingly enough, you probably only think Samuelsson is any good because he’s scored 30 goals, so in other words, the usual stats.
Interestingly enough, it’s actually not all that big a secret why Samuelsson is scoring so much (luck and circumsances). It doesn’t take “stats-heavy” analysis, just common sense and a keen eye towards his track record.
I mean if we’re going to talk about GMs who have properly evaluated players who have not generated much offense but are useful at helping you win, then Gillis is pretty low on that list. I mean the Canucks have some of the worst bottom six forwards in the league, of course they can get away with it because of the strength of their top players (who were obtained under Nonis and Burke) and a year of good bounces.
Interestingly enough, you missed my point entirely. (But I’m not surprised that you have decided that it’s only one player on each line who drives the bus entirely. Samuelsson has been injured, so now’s a good time to check.)
I’m not arguing about Gillis. I’m wondering about whether NHL or NHL managers have more statistics that we don’t know about. Do you have any info on this? That’s suggested by the article linked and also by Gabe’s post.
Btw, I believe the Canucks have among the league’s highest man-games lost to injury. They’ve had their best forwards, defencemen, and goalie out for significant chunks of time this season, including their best d-man for a third of a season and counting (after an ugly hit from behind by Gino). How is that “good bounces” or “puck luck”? Oh, that’s not included in “puck luck”—my bad.
Assessing defensive contribution seems to the thing in baseball now, with Range Factor and Zone Rating getting a lot of press. Defensive contribution in hockey seems hard to measure. What do you think are the best measures? Where do you think defensive assessment is headed?
by DoctorMyBrainHurts on Mar 27, 2010 3:31 PM EDT reply actions
1980s? Heck, wasn’t he doing that all the way back with Toronto in the 70s?
SNN Sports - A theoretical Oilers blog (i.e. theoretically, I write stuff there). Link now 100% less broken.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
