Joffrey Lupul of the Toronto Maple Leafs kicked off another round of “watch the game” vs “advanced hockey stats” debate today on Twitter, with the following bromides:
contracts aren’t awarded by this CORSI i am hearing all about. They are awarded for an equal value of skill and depth (at a certain position
— Joffrey Lupul (@JLupul) July 7, 2013
If you bring certain attributes and you play to win. I’ll take you on my team 7 nights a week. Lets not look at this like Moneyball.
— Joffrey Lupul (@JLupul) July 7, 2013
Cam Charron over at Leafs Nation did an outstanding job of breaking down the false dichotomy of “scouts vs stats” that many people seem to be hung up with, but I did want to make a point here that’s been rattling around my head for a while. It tends to get bouncing around again whenever this debate about the utility of “Fancy Stats” comes up.
The thing is, hockey stats aren’t all that advanced, things like Corsi & Fenwick only seek to break the game down into components which are meaningful for projecting future results. Consider the following comparison of stats used in hockey with some of the other major North American pro sports:
| Baseball | Basketball | Football | Hockey | |
| Cro Magnon | Runs Scored, Wins & Losses for Pitchers | Points per Game | Touchdowns (run, catch, pass), Wins & Losses for QB’s | Goals, Assists, Plus/Minus, Wins & Losses for Goaltenders |
| Basic | Batting Average, ERA, OPS | Assists/Blocks/Rebounds per Game, Field Goal %, Plus/Minus | Yards per game (run, catch, pass), sacks, interceptions | Corsi, Fenwick, Goalie Save % by Situation (EV, SH, PP), Penalty Plus/Minus |
| Advanced | BABIP, UZR, FIP, OPS+, ERA+ | Offensive & Defensive Rating, Player Efficiency Rating | Yards After Catch, Expected Points Per Play | Zone Entries & Exits, Shot Quality Neutral Save % |
This isn’t meant to be an exhaustive list, it’s only to show the level of complexity that some of the “advanced” hockey stats like Corsi & Fenwick have relative to commonly accepted metrics used in other sports.
At the most basic of levels we have events which are directly tied to scoring changes within a game, or the ultimate decision of a team’s Win or Loss as assigned to an individual (in baseball the pitcher, in football the quarterback, in hockey the goalie).
Early on it became understood that discussing individual performance only in light of those scoring plays didn’t tell the whole picture – so in baseball we got down to the components of run scoring like hits, walks, stolen bases, and for pitchers the fundamentals of strikeouts & ERA. In basketball they moved beyond the score-related stats to account for things like rebounding and field goal percentage. In football, we talk about individual and team performance in terms of yardage all the time. These are the detailed elements of a game which lead to scoring events, and taking them into account gives us a more complete picture of an individual athlete’s contribution.
But in hockey, where the equivalent is the shot-related measurements like Corsi & Fenwick, many treat this stuff like it’s quantum physics or sorcery. If you don’t care about this stuff and just want to enjoy the game, that’s totally fine. But to deny the utility of tools which have been developed, tested, and honed over the course of several years simply because they don’t offer 100% certainty is, well, nuts.
Hockey is just now catching up with where other sports were decades ago. Most (if not all) NHL teams count scoring chances, and some are actively working with the fancier stuff and integrating it with their “old school” scouting practices. After all, it’s only common sense for organizations with are often spending more than $60 million on their hockey players each year. James Mirtle of the Globe & Mail covered a lot of this ground with Gabe Desjardin (of Behind the Net) in September 2011:
Desjardins met that resistance firsthand when he began meeting with teams, recalling one recent sit-down with an NHL general manager who wrote off his work as merely “doing arithmetic.”
He believes most teams still have a long way to go in terms of embracing what advanced statistics can do for their organizations.
“You can see that there are a lot of decisions made every year – Philadelphia getting [Ilya]Bryzgalov, for one – that pretty much any analytics department would, 100 per cent, advise you against,” Desjardins said, referencing the Flyers netminder’s $51-million contract as an example of inefficient spending.
“As high as player salaries are, if you hired a very skilled analytical consultant from industry, paid them $150,000 to $200,000 a year, and he sits there and works on things all year and comes up with recommendations, what are the odds that he’s not going to be able to find you a player that you can sign for $200,000 less his value or less than you otherwise would have paid?
“It’s pretty obvious that there’s some value in there.”
Blindingly obvious, really. How did that Ilya Bryzgalov contract work out for Philadelphia, anyway?