I'd be remiss if I didn't give credit to Vic Ferrari, who posts the Corsi and Fenwick numbers at his site, timeonice.com
From a hockey metrician's point of view, traditional plus/minus is probably THE one stat that will elicit a collective facepalm from the entire hockey metrics community.
And for good reason.
Just look at the Stars penalty kill as a prime example of how useless a stat it's been. The struggles of that unit have been well documented. Yet, if you look at the individual stats, you'll see that Nicklas Grossman leads the club with a +11 rating while Robidas has a +7.
And it's all because those problems on the penalty kill have absolutely no effect on plus/minus. Consequently, if you're a vital part of the overall success on a power play unit, you don't get any credit.
So it's not surprising to see how traditional plus/minus has provided the impetus for other plus/minus ratings. This week, we'll take a look at three such ratings systems and discuss the pros and cons.
Corsi Ratings: Developed by Buffalo Sabres goaltender coach, Jim Corsi.
Japers' Rink has an excellent introduction to Corsi Ratings from last season.
It measures all shots taken in even strength situations, excluding empty net situations, that are either goals, shots on goal, blocked, or missed the net. As you might guess, you get a plus rating if your on the ice when your team scores an even strength goal, puts a shot on net, block a shot, or your opponent misses the net. Consequently, you get a minus if your team surrenders an even strength goal, puts a shot on net, blocks a shot, or if a shot from your team misses the net.
The pros of such a system are that they measure quite a bit more than traditional plus/minus. More to the point, it tries to measure how well your team can apply pressure to the opposition and how well it handles pressure.
Unfortunately, Corsi numbers don't measure power play, short handed, and empty net situations. I understand why as it's almost impossible to correctly weight those situations against regular even strength situations.
Fenwick Ratings: Developed by Matt Fenwick from the blog, Battle of Alberta.
His criteria is almost the same as Corsi's, except it takes blocked shots out of the equation.
Chris Apple and Marc Foster plus/minus: Developed by one of my friends and one of his data analytical friends almost a decade ago. They wrote a series of articles in 2001 that appeared in CNN/SI that you can check out here.
In their system, you get a plus rating for all goals scored by your team when you're on the ice, no matter if it's an even strength tally, short handed tally, or power play goal. Consequently, you get a minus if you're on the ice when your team gives up a goal.
The pro of this system, obviously, is that it takes into account ALL goals that are scored in the course of a game, regardless if it's even strength, on the power play, or shorthanded.
Of course, it does treat all goals the same and thus, doesn't weight even strength or shorthanded goals above power play goals. But to their credit, Marc and Chris also developed a separate weighted goals system that weights goals more that give a team the lead or tie a game than ones scored in garbage time. But that's for another discussion.
So now that you know how each metric is measured, let's look at the Stars best and worst players by each plus/minus rating system after the jump.
First, let's look at the Corsi ratings. You'll notice the secondary header row labeled DAL and OPP for Goals, Saved Shots, Missed Shots, and Blocked Shots. I labeled them the way I did so you'll see them from an offensive POV.
That is, where it says 48 saved shots for Dallas on Brad Richards' ledger, it means that the Stars fired 48 shots that got through on goal and were saved while Richards was on the ice. Consequently, the 41 under Opp on Saved Shots means the Stars gave up 41 shots on goal that were saved while Richards' was on the ice.
No big surprise that Richards is leading here. His line has been the most dominant of all the Stars' lines this season. What sticks out to me, however, is Jeff Woywitka's high rating versus Trevor Daley's low rating.
The 90 shots on goal allowed this season when Daley's been on the ice in even strength situations far and away leads the team. Of course, he also leads the team, far and away, in ESTOI with 186:25. The second highest total in that department belongs to Nicklas Grossman with 160:49.
Also of interesting note is Jamie Benn's -23 rating, despite being on the ice for 8 Stars' goals versus 1 total from the Stars' opponents.
OK, let's take out blocked shots and see what we get with the Fenwick Ratings:
Woywitka shoots to the top of the charts for one. But what stood out to me is that with blocked shots in the equation, Stephane Robidas is a +8 in the Corsi Ratings. Take out blocked shots and he's a -10 player in Fenwick Ratings.
Now onto the Foster/Apple total plus/minus ratings:
Remember when I said Grossman and Robidas were +11 and +7 in traditional plus/minus ratings at the beginning of the post? Well factor in the troubles of the penalty kill, and they're +5 and +4 ratings respectively with James Neal and Brenden Morrow taking the top spots with Brad Richards not far behind.
So what have we learned?
For one, there's no such thing as one metric that's going to perfectly measure a player. And any variation of plus/minus will validate that point. But what it does do is expose new strengths and weaknesses in an individual player's game.