Charts & Graphs - Dec. 5, 2013
If you're a hockey fan, you probably haven't escaped hearing the terms "Corsi" and "Fenwick". It may be enough to make you mute the television, or click the 'x' on the browser tab. That's been one of the problems with hockey analytics—the confusing names sound a little bit like acronyms, and there are few people that explain what 'Corsi' is when bringing up to a large audience that may not be receptive to new ideas in the game.
'Corsi' and 'Fenwick' are very similar. They measure relatively identical things, and if you have played a small sample of games, they're very helpful for determining how a team has played relative to its success.
Essentially, 'Corsi' and 'Fenwick' measure shot differential. A team's (or player's) Corsi percentage is sum of all the shots: on net, missing the net, and blocked shots, divided by the total amount of attempted shots for both teams, and almost always measuring just 5-on-5 situations. That number correlates very well with the amount of time spent in the opponent's zone, and expressed as a percentage, it can be thought of as a good approximation of the amount of time a team spends in the attacking zone.
Fenwick is almost the same as Corsi, but it excludes blocked shots completely.
It's interesting. Most analysts that work with these numbers use Fenwick almost interchangeably with Corsi. In the last 82-game season, in 2011-2012, the top five Corsi teams were Detroit, Los Angeles, Pittsburgh, Boston and St. Louis. The top five Fenwick teams were Detroit, Pittsburgh, St. Louis, Los Angeles and Boston. The same five chairs, just arranged slightly differently around the table.
Here's an example of a Fenwick graph from the Canucks game against the Predators on December 3 courtesy of ExtraSkater.com.
Shot blocking is still brought up as a relatively important part of the game, but there's become a serious debate as to whether it's better to have a shot blocked or whether to block a shot. On one hand, you're preventing an opportunity for your opponent to get a puck on your goalie. On the other hand, the circumstances leading up to the blocked shot mean you were playing in your zone, and you can't score from your own zone (unless the opposing goaltender is Vesa Toskala or Jonathan Bernier).
Also, blocked shots factor in just a quarter into the full Corsi equation. There's a lot less math than you'd think, but lay it out this way. Over the long-term, having the puck in the zone is a positive. Think of it like a dartboard: a slightly larger bullseye may not give you a noticeable advantage if you're only taking 20 throws, but if you take 1000 you'll notice a difference in the amount of times you hit the centre. Part of the reason a lot of traditionalists dismiss these analytics off-hand is because in the small samples, things don't always line up with "what the metrics say" when you're looking at just a single game, or a small handful of games. Corsi is not good to use to attempt to explain the past.
SO WHAT DOES THIS TELL US ABOUT THE CANUCKS?
The consensus in the analytics community has been that the Vancouver Canucks are due for a regression to the mean… upwards. Through Tuesday's games, the Canucks have the 8th best Corsi in the league counted in score-close situations*, and 7th best Fenwick.
* - "score-close situations" means when the game is tied, or where one team is within one goal in the first or second period. The reason for this is that teams that are behind (especially late) tend to take a disproportionate amount of low-quality shots.
The problem, as we all know, is that the team's they're chasing are also exceptionally good at puck-possession. The Pacific Division is stacked with good teams, so the Canucks are still behind Los Angeles and San Jose in the division.
That said, the Anaheim Ducks have fallen to below 50% Corsi in recent days, so they've crossed the line into a negative possession club. The Phoenix Coyotes have also been there for some time, and the Canucks are closing in on them, just a point back of the Coyotes for the final playoff spot in the West, although Phoenix has three games in hand.
There are some reasons to be positive. The powerplay has begun to score a lot more recently, as I thought it might when I focused on that in a previous Charts & Graphs column, and I have a feeling they'll end a couple of points higher than their projected 96-point pace that you'd use if you looked at the standings today (and I *definitely* would have said this before their two-game win streak. The Canucks were on pace for 91 points over 82 games after Saturday's game in New York).
For the Corsi and Fenwick, well, all I can say is that there's a higher observed correlation between those two rates and "wins" at the end of the season than there will be in early December.
Hopefully this clears up any confusion you may have had about Corsi and Fenwick, and how and why they're used.