A few weeks ago I took a look at the Corsi rating and Corsi %, specifically if a team’s Corsi % ranking could predict how well they would do in the playoffs. It seems the highest ranked Corsi teams have a better chance of making it further into the playoffs and winning the Stanley Cup. There is room for improvement though, and with that in mind a modification called the Fenwick rating was created. Named after Matt Fenwick of The Battle of Alberta hockey blog, this alternative rating removes blocked shots from the calculation of the Corsi rating, leaving just shots on goal and shots that missed the net. The main idea behind the modification was that blocked shots generally come from lesser scoring chances and are therefore not as relevant to how well a team is playing. There are people who prefer one statistic to the other for different reasons, but is one any better than the other?
The equations to calculate Corsi and Fenwick are the same, but the Fenwick rating doesn’t include blocked shots in the “Total Shots” numbers. The particular subset of data used in this article for the Fenwick rating is taken from 5-on-5 situations with the score close (score close means the score is within one goal in the first two periods or tied in the third). The reason for this is that teams tend to play differently when the game isn’t close, be it the coach telling the leading team to play a more defensive game or the players changing their game subconsciously when ahead by a few goals. If sticking with stats when the game is “close”, we ensure that we get the most accurate look at a team.
As with the Corsi rating, I wanted to see if a high rank would make it more likely to go further into the playoffs. If the Fenwick rating of a team had no correlation with their playoff outcomes, we would expect all teams to have a 53.3% (16/30) chance of making the playoffs, 26.7% (8/30) chance of making it past the first round, and etc.
Since the NHL began recording data on missed and blocked shots in 2005, 80% of the teams ranked in the top 10 by Fenwick % made the playoffs, so successful teams usually have a good Fenwick rating. The shape of this graph is very similar to the Corsi ratings from last time:
With such a similar shape to the Corsi ratings, I compared them directly to see how close they really were:
Other than one deviation they are virtually identical, which shouldn’t be too surprising considering their only difference in calculation is blocked shots.
I reseeded teams once they made the playoffs based on their regular season Fenwick rank, which proved to be even more similar to the Corsi rank:
There is only 9 seasons of data to work with, so any variances could still appear, but for now there doesn’t seem to be a difference between the two when considering the whole team statistic.
The next graph just shows how each individual seed/ranking performed over the past 9 seasons. Both ratings again agree on a bit of an advantage to teams in the top 5 or so of the rankings, but the rest is mostly noise around the 50% average. Comparisons of the later rounds brought similar results.
At around the midway point of the 2014-15 NHL season complete, Tampa Bay and Chicago are both ranked in the top 3 for Corsi % and Fenwick %, and Detroit, Nashville, Winnipeg, and the New York Islanders all rank in the top 6 for both statistics. All the current playoff teams lie within the top 19 Fenwick % teams, with Minnesota, Florida, and Dallas also in that mix. Those 3 non-playoff teams are within striking distance of the playoffs, especially Florida being 3 points behind Boston with 4 games in hand.
There are many of us “hockey nerds” out there who will debate that Corsi is better than Fenwick, and vice-versa. However, the two are virtually the same in terms of the outcomes they predict. The only upper hand that I would give to the Corsi rating is that since it includes blocked shots in it’s calculations, there is a larger sample size, leading to more precise results. Otherwise though, the difference between the two is the difference between two shades of grey .
All Corsi and Fenwick statistics taken from war-on-ice.com