Every few days, a person of some public renown mentions Corsi, which is closely followed by the entire hockey Twitterverse exploding into a brouhaha. People then proceed to squabble and bicker for a while until things eventually die down.
Rinse and repeat.
But what is Corsi? What is Corsi not? What can it and can't it do? What relationships does it have with players?
If you were to put it simply: it is a slice of the pie that is what makes a player or team effective but not the whole pie.
There have been numerous of articles on Corsi, from investigative research pieces to layman introductories on hockey statistics. With all the commotion lately, plus Arctic Ice Hockey's large growth as a Winnipeg Jets community, one more gloss over won't hurt.
The math that goes into Corsi is quite easy and not very advanced at all; it is simply adding up all shot attempts together, whether they be on net, misses, blocked or goals.
In single game samples it is often shown as a raw number similar to +/-. For example: if Jets get six shot attempts while Jacob Trouba is on the ice and 3 shot attempts are taken by the opposition, Trouba would be a +3 for that game.
For larger samples the number is often per 60 minutes of play, as seen at www.behindthenet.ca. For example, Dustin Byfuglien has +1.21 Corsi for the 2012-13 season. Another option is making it a percentage of Corsi-for events divided by all events, which is the way it is shown at extraskater.com. For the same 2012-13 season Byfuglien had a 50.3% Corsi percentage.
Corsi is sometimes shown as relative to the team, where the player's Corsi value is subtracted by the team's value when the player was not on the ice for the same games. Tobias Enstrom had a relative Corsi percentage of +6.2% in the 2011-12 season.
Corsi is not an acronym. It is named after Jim Corsi, the goalie coach for the Buffalo Sabres. Corsi started tracking all shot attempts to better measure a goaltender's workload, since a goalie must react to all shot attempts.
'Vic Ferrari' -- an Edmonton Oiler's blogger -- one day decided to pull numbers on a few games and he noted that the differential at even strength seemed to mirror his opinion on puck possession dominance. At the time there had been a lot of blogger conducted research on scoring chances, in an effort to deal with the variability in +/- and goal%. This new stat opened new areas of research where Corsi% was shown to have a close correlation to scoring chances. The rest, as they say, is history.
Corsi can proxy scoring chances reasonably well, although once a larger sample is established Fenwick (Corsi without blocked shots) does hold the strongest relationship. The advantage to shot metrics over scoring chances is with the data being tracked and readily available by the NHL already, so there is no need for the huge manpower needed in manual tracking and there is also league wide information to compare to when making context.
These relationships are why Corsi predicts future goal differentials better than past differentials can. Out scoring your opponent the primary goal and leads to wins, which is why Corsi then also predicts future wins better than past goal differentials or winning percentage.
The numbers can also be split into particular situations. WOWY's (with or without you) are commonly used to show how certain players drive possession by improving their linemates, while anchors do the opposite. With chopping the numbers down to tiny samples, it is better to look at trends rather than individuals.
Corsi is not everything and all things; however, this does not diminish what it is. Winning in puck possession and scoring chances is important and will lead to wins but does not encompass the full game. The largest factors outside of possession and chances are luck (ie: bounces), special teams, and combination of goaltending and shot quality (probably in that order). Gabriel Desjardins once estimated that about 75% of winning percentage is attributed to a combination of Corsi/Fenwick and luck; this means that by accounting for just those two factors, you can tell a great deal of what is going on... almost -but not quite- everything.
The other major limit to Corsi is the need for contextual nuances. Recently Tyler Dellow stated: a player’s Corsi% can’t be divorced from the context in which he plays. There are many different things that can affect a player's Corsi; who they line up against (Quality of Competition), who they line up with (Quality of Linemate), which zone they are predominately deployed in (Offensive Zone Starts), and what line they are on (Time On Ice) are the major ones. There are likely other minor ones as well, which Dellow alludes to in the article his quote is from.
What It Is Not
Corsi is not something that gets inflated simply by shooting more often.
Recently Jordan Eberle depicted (very accurately) the basics of Corsi. In the sound clip he mentions that when players do things correctly -like breakouts and zone entries- a player will naturally improve Corsi; he also states that once a player starts throwing the puck as soon as they cross the blue line, things will start to go south.
This is correct. Looking at the last three seasons combined, there is a relationship between a player's Corsi and their personal shot attempt rate, albeit a very weak R^2 of 0.27. Keep in mind that being in another team's zone more will naturally give more opportunities to shoot and that better players will be able to create relatively more as well.
Corsi on its own was, is and will never be a perfect indicator of how good or bad a player is. It cannot perfectly predict the future or tell you who will win.
But neither can a fan's eyes be a perfect indicator of how good or bad a player is, nor will traditional scouting ever be perfect in its ability to predict a player's future success.
What they can all do is tell you something.
Statistics can help guide your eyes to what is going on, who is helping and by how much.