COHEN’S CONDEMNATIONS: NHL Statistics
Greetings hockey fans!
Jesse Cohen here again to tell you why the things you love are horrible and why you’re horrible for loving them!
At this point I think it’s fair to assume that most of you are at least peripherally aware of the “Advanced Metrics” (aka “Fancy Stats” aka “Moneyball”) Revolution taking place in modern athletics. For those that aren’t familiar with the Jock Civil War threatening to tear us all apart, here’s the gist of the conflict: Nerds realized the people running sports had no idea what they were doing.
I won’t bore or confuse you with the details of the new “Fancy Stats” but trust me when I say there’s really nothing complicated about them. They just require someone (usually with bad posture and complexion) to obsessively peruse every single moment of every single game and track every single mundane detail about what happens on the field of play and then objectively compile a record of every single event for later statistical analysis. That data usually manifests itself as a completely illegible spreadsheet of numbers that, without any kind of familiarity or context, will completely turn off and enrage your typical sports fan or human being.
But what if I told you that the traditional statistics the average NHL fan parrots back with confidence should be just as questionable and controversial as these new “Fancy Stats”? Would that blow your mind?
THE 5 WORST THINGS ABOUT
TRADITIONAL NHL STATISTICS
This the easiest stat to tackle. The crowd of people who turn up their nose at +/- is growing exponentially and if you’re comfortable instantly judging people by a single opinion just listen to them talk about a players +/- rating. Take a look at this goal and tell me if everybody on the ice for the Hurricanes deserves a “+”.
Hockey is a team sport but scoring moments can hinge on spectacular individual efforts both sublime and comical. A closer look at how +/- is distributed discovered that up to 30% (and possibly up to 40%) of +/- stats are attributed to players that had no significant impact on a scoring event.
4. Goalie Win/Loss Records
During the 2011-2012 NHL Regular Season, Jonathan Quick played in 69 games. He compiled a Win/Loss record of 35-21-13, including 10 shutouts.
In 4 of his 13 OT/SO losses he allowed only a single goal. In 5 of his 21 Regulation losses he allowed only a single goal.
That’s 9 of 34 (26%) games he wasn’t credited with a win while allowing only a single goal.
The total number of games that Quick allowed 2 or fewer goals was 46. Only 5 goalies that played 30 games or more had a GAA of fewer than 2.00 in 2011-2012 (Quick was one of them) and of those goalies, Quick faced the most shots and played the most minutes. Yet his record shows only 35 “Wins” and he finished second to New York Rangers goalie Henrik Lundqvuist in “Wins”, and Vezina Trophy voting. Lundqvuist played in 41 games where he allowed 2 or fewer goals.
Now I’m not suggesting that number of goals allowed is somehow a more perfect system for determining how well a goalie played on any given night. I’m simply demonstrating that attributing a “Win-Loss” record to one player who can only impact one half of the score is a particularly ridiculous system.
3. Game Winning Goals
When I was growing up and learning the game of hockey, there were a number of details that caused me to tilt my head like a confused dog and wonder “Who the hell came up with that idea?”
One of those that I still don’t understand is the Game Winning Goal. It’s a simple enough process to determine which goal was the “game winner” but it’s hardly an effective method of determining which players possess any kind of “clutch” or “ability to rise to the moment” and isn’t that really the only point of recording “Game Winning Goals”?
Take the 2012 Stanley Cup clinching Game 6 between the Los Angeles Kings and New Jersey Devils. A 6-1 victory where the Devils were never really a threat to win. It was 4-0 before the Devils finally scored late in the 2nd period. So why is Jeff Carter credited with the Game Winning Goal just because he scored the second goal in a rout?
Now THIS is a “Game Winning Goal”.
2. Secondary Assists (and sometimes even Primary Assists)
The method used to distribute assists in the NHL is simple,. That’s good. However it leads to absurd situations where players not even remotely involved in scoring a goal are credited with an assist. That’s bad.
Here is a perfect example.
Pavel Datsyuk starts behind his own goal line. He picks up the puck near the hashmarks, carries the puck out of his own zone, through the neutral zone, into the offensive zone, past 3 members of the Predators and scores the goal purely as a result of his own talent and determination, having traveled the entire length of the ice and passing every other player.
Daniel Cleary and Jonas Gustavsson, THE GOALIE, were both credited with assists.
1. Screening The Goalie
If the NHL simply handed out assists to players who didn’t deserve them, one could question the usefulness of traditional statistics as a way of judging a players worth, but the NHL compounds the problem by NOT handing out assists to players who play a huge role in scoring goals.
Helm is credited with the assist on this goal, but what did he do that contributed most to the goal being scored?
Lazily dish off the puck to Franzen or screen the goalie.
Not only does a screening player obstruct the goaltenders view of the incoming shot, they also put themselves at risk of injury. Yet there is no stat (that I’m aware of) that tracks this contribution. I realize that there is a subjective element to what contributes to a screen and occasionally the screening player has already been credited with an assist (like the play above), but even if a player doesn’t clearly obstruct the goalies vision his mere presence should distract the opposition defence and the attention of the netminder as a potential scoring opportunity. That alone should merit some kind of statistical attention.
So there you have it puck fans! The next time someone asks “Do you even Corsi, bro?” just let him know that you refuse to address “Advanced Metrics” until the NHL fixes the “Remedial Metrics”.
Then tell them it’s all about PDO anway.