In the last 3 posts, we've discussed Corsi and Fenwick, two measures of possessions that rely upon shots in order to measure the effectiveness of players and teams. Neither Corsi nor Fenwick care about the quality of the shots taken; not only do these measures include missed shots (Fenwick & Corsi) and blocked shots (Corsi, but not fenweick), but they include even the worst shots in existence - the shots from angles which have near no chance to score a goal. This isn't necessarily a bad thing - if we're seeking to measure possession (Corsi), then we certainly would like to have even the bad shots, as these shots provide statistical evidence as to which team has control of the puck.
However, if we're talking "effective possession," and really how well teams are playing the game of hockey, you'd think we'd want to exclude bad shots, even if they aren't blocked or do manage to go "on net."
Enter "Scoring Chances." We've all heard of scoring chances before; hell we've all even said at one point in our hockey-watching lives "oh we've had so many great scoring chances this game" (or something thereabout). We may even have heard our Islanders' TV Broadcasts give, during one of the two intermissions, statistics about the amount of scoring chances each team had during a period. Perhaps the broadcasters will say that the Isles outchanced (or were outchanced by) the opponents by some number. In essence, scoring chances are what we'd imagine them to be, the # of shots that are truly a threat to score a goal.
But if you've ever tried to look up scoring chance statistics on a website, through either the NHL's website or the team's official website or some unofficial website, you may have noticed something: the league does not keep track openly of scoring chances. You can't look up which teams lead the league in scoring chance differential, which teams give up the least amount of chances, which players create the most, etc.
So where do the scoring chance numbers that team broadcasts occasionally show come from? Well the answer appears to be this: The TEAMS themselves. Yes, while the NHL isn't tracking scoring chances for us all to see, the teams themselves are tracking them and are using these numbers for their own purposes. Occasionally, that seems to involve sharing those #s with the local TV Broadcast. But not with the public in general.
So a few years back, a few hockey fans decided to do something about this. Led by Oilers bloggers, they started the Scoring Chance Project. The project aimed to do what the NHL was not --> to track the # of scoring chances created and allowed by each team and who was on the ice during each scoring chance (to create a +/- involving scoring chances). Individual fans of teams, starting with the Oilers, would go over each individual game of their teams and track each scoring chance that occurs during those games.* At the end of the year, each fan would have a complete record of the team's scoring chance results.
*IMPORTANT SIDE NOTE: How do we define a scoring chance for this project?
*From Copper & Blue: A scoring chance is defined as a clear play directed toward the opposing net from a dangerous scoring area - loosely defined as the top of the circle in and inside the faceoff dots (nicknamed the Home Plate, detailed below), though sometimes slightly more generous than that depending on the amount of immediately-preceding puck movement or screens in front of the net. Blocked shots are generally not included but missed shots are. A player is awarded a scoring chance anytime he is on the ice and someone from either team has a chance to score. He is awarded a "chance for" if someone on his team has a chance to score and a "chance against" if the opposing team has a chance to score.
As of last season, 20 different teams had at least one person online tracking that team's scoring chances and reporting the results online. The Isles' themselves were not tracked (it's a little time consuming, honestly) but with 2/3 of the league being tracked, nearly every game was in fact being accounted for. This gave hockey analysts a large amount of data to work with; data that the teams had but refused to show the public.
So what does this data show? Well...as noted by Eric T of NHLNumbers.com (and SB Nation's Flyers' affiliate, BroadStreetHockey), scoring chance #s correlate really really really REALLY closely with fenwick #s. In other words, if you simply ranked teams by fenwick, you'd wind up with nearly exactly the same rankings if you used scoring chance +/- instead. We may be using slightly better data - and it's unclear if we are or not by the way - but we're not getting much better results. The same appears to be the case with individual players scoring chance +/-.
This makes sense. The teams that are the best at maximizing their scoring chances while minimizing the scoring chances of opponents are going to be the teams that are the best at controlling POSSESSION! If you keep the puck in the opponents' zone more than your own, well, you're going to have good scoring chance +/- numbers because your opponents will have a hard time out chancing you without controlling the puck for most of the game.
This doesn't necessarily mean that tracking scoring chances is useless or that the scoring chance project should stop --> if scoring chance data was readily available for every team, it might be preferable to simply having corsi/fenwick data by a small amount (again this is unclear). In addition, scoring chance data may have some other value --> for example at telling us which players are getting lucky or unlucky and for whom regression may be coming, for better or for worse.
That said, this finding should serve once again to emphasize to you the importance of possession and the possession metrics we have been talking about these last 3 posts - the teams and players who are the best at maintaining possession are the ones who win the battle of scoring chances and are the ones who, in the long run, will outscore their opponents over a long season.
The Intro to Hockey Analytics/Advanced-Hockey-Statistics Primer so far:
Part 1: - What is the field of Hockey Analytics and Why Might You be Interested?
Part 2.1: - The Importance of Context Part 1 - Time on Ice
Part 2.2: - The Importance of Context Part 2 - Evaluating the Difficulty of Certain TOI through QUALCOMP and Zone-Starts
Part 2.3: - The Importance of Context Part 3 - Evaluating (and Compensating for) the Effect of Teammates via QUALTEAM and Relative Measures
Part 2.4: - The Importance of Context Part 4: The Concept of the Replacement Level Player
Part 3 - The Perils of Sample Size
Part 4.1 - Introduction to Hockey Analytics Part 4.1: Possession Metrics (Corsi/Fenwick)
Part 4.2 - Introduction to Hockey Analytics Part 4.2 - Possession Metrics: The Various Forms of Corsi Available on Hockey Sites
Part 4.3 - Introduction to Hockey Analytics Part 4.3: Possession Metrics: Fenwick a Measure of Effective Possession