I am taking a look at sabermetrics and hockey this summer. I have been looking at Corsi ratings and how to turn them into a meaningful individual ranking. I have produced a top 20 and worst 20 list of team adjusted Corsi ratings. This is a gage of which players have the best puck possession in 5 on 5 situations relative to that of their teammates. The main problem with these rankings is that it doesn't take into account the fact that players on the same team play different roles. Some player more offensive roles and others play more defensive roles. One way to gage this is using offensive and defensive zone starts. When a faceoff occurs we record which players are on the ice and in which zone of the ice. Those who are on the ice for more offensive zone starts than defensive zone ones play more offensive roles and are expected to have higher Corsi ratings than those who play more defensive roles with more defensive zone starts than offensive ones. This is an effect that can be included into adjusted Corsi ratings.
First we must know which players have the biggest imbalance in their zone starts. I posted the top 20 players by excess offensive zone starts. Today here are the top 20 players by excess defensive zone starts. These are the players who have the highest number of defensive zone starts minus offensive ones. These are players who have been unfairly given low rankings in the team adjusted Corsi lists.
One of the interesting things to watch this season will be the Vancouver Canucks and New York Rangers since they exchanged coaches. In 2013 Alain Vigneault coached the Vancouver Canucks and John Tortorella coached the New York Rangers. In 2013/14, Vigneault will coach the Rangers and Tortorella will coach the Canucks. In principle it is a great way to evaluate these coaches relative to one another. In practice things will not be so simple because there will be other changes in personnel, injuries etc. that muddy the comparison, but it will still be interesting.
We can look at these moves from the point of view of zone starts. I recently published the top 20 players by excess offensive zone starts in 2013 and it is clear that two teams are more heavily represented on the list and they are the Canucks and Rangers. From the Canucks, Daniel and Henrik Sedin are first and second on the list and their usual linemate Alexandre Burrows is seventh. From the Rangers, traded Marian Gaborik is second, Brad Richards is sixth, Rick Nash is 12th and Michael Del Zotto is 17th. We can see that both teams employ the strategy of giving their top offensive players as many excess offensive zone starts as possible. It was probably a more common thing in Vancouver because the Sedins had a better season than Gaborik, Richards et al did with the Rangers.
This summer I have been looking into sabermetrics and hockey. I have specifically been looking at Corsi ratings and their adjustment. I have produced a top 20 and worst 20 list of players by team adjusted Corsi ratings. The biggest problem with these lists is that they do not take into account the different ways players can be used on the same team. The most easily quantifiable difference comes from zone starts. We record which of the three zones (offensive, neutral and defensive) a player is in when he is on the ice for a faceoff. This is a measure of how defensive or offensive a player's role is. A player used in a defensive role will have more defensive zone faceoffs and a player used in an offensive role will have more offensive zone faceoffs. Similarly a player with a lot of offensive zone starts should have a higher Corsi than one who has a lot of defensive zone starts. This is another factor we can adjust for.
Today I will list the players with the most excess offensive zone starts. This is offensive zone starts minus defensive zone starts. That is the meaningful number when adjusting Corsi for zone starts. These are players who were used in offensive roles either because they have little defensive value and need to be protected or because they are the top offensive player on their team and you want to maximize their offensive value. These are players who are overrated by team adjusted Corsi. That doesn't make them overrated players. Rather they are players who played a role that should maximize their Corsi.
Here are the top 20 players in 2013 by excess offensive zone starts:
Several days ago I posted the 20 worst players by team adjusted Corsi in 2013. This is a measure of puck possession while a player is on the ice in 5 on 5 situations. The players are then ranked against a team dependent baseline so that comparison between teams is possible. It is an imperfect comparison as players will play under different circumstances on the same team, so the same adjustment for each player will be at best an approximation. The worst player in 2013 by this metric was Robyn Regehr, who split the season between the Buffalo Sabres and Los Angeles Kings. He posted a -188.04 team adjusted Corsi.
Regehr was once a top NHL player. He was on Team Canada in the 2004 World Cup and 2006 Olympics. Even in those days he was not the strongest puck possession player. Regehr was good defensively and this translated to his Corsi because he didn't allow his opponents much puck possession. The problem is he no longer does that. Regehr is no longer a strong defensive player and he never had offensive skills. He is the kind of player who will be poor at puck possession and his value to any NHL team is questionable.
Steve Simmons writes a bunch of soundbites and publishes it as a column in the Toronto Sun each week. Here is his most recent example. In it one of his soundbites is
Another reason why I have next to great difficulty for the CORSI analytic statistic in hockey. I saw a team adjusted CORSI ranking for this past season. Tyler Seguin of the Bruins was rated fifth best in the NHL. The same Seguin whom Bruins coach Claude Julien kept on the third line, moving rookie Carl Soderberg from press box to first-line centre when Patrice Bergeron got hurt. I’ll take Julien’s instincts over strange numbers anytime
My first reaction when I read this was perhaps he read my work. After all I published the top 20 players in 2013 by team adjusted Corsi this week. That is unlikely to be a coincidence is it? If I am the source that Simmons used, he definitely did not understand what he read and probably made no attempt to understand it. That is easily shown because Tyler Seguin is not fifth best in the NHL, he is actually sixth. That isn't a big difference and perhaps implies that Simmons saw some other list that team adjusted players in another manner. In a worst case it shows Simmons was not very good at fact-checking. However, whether Seguin is 5th or 6th really makes no difference to the point he is attempting to make.
A few days ago I posted the top 20 players in the 2013 season by team adjusted Corsi rating. This is calculated by taking an individual player's raw Corsi rating which is the difference between the shots attempted by a team and their opponents when the given player is on the ice in 5 on 5 situations. Thus we are looking at the players who have the worst team puck possession when they are on the ice and subtracted off a team baseline based on the team Corsi rating for their team. This allows us to better compare players between different teams. In order to include players who played on more than one team a baseline team adjustment is found by multiplying the team adjustment for each team played on by the number of games a player played with that team and divide by the total number of games played. There are players where this is important in 2013.
This list will give us a group of players who are generally poor players and players who played in defensive situations and hence were not playing a puck possession game. This is the next step to make Corsi ratings into a meaningful individual number. Zone starts are not taken into account yet. Nevertheless, this is a group of players whose team had worse puck possession when they were on the ice than when they were not on the ice.
Here are the 20 worst players in 2013 by team adjusted Corsi:
In my summer sabermetrics and hockey series I have recently posted the top 20 players by team adjusted Corsi ratings. This is an improvement over the top players by raw Corsi ratings, but it still needs some work to be a viable rating for individual players since it does not take into account the circumstances in which the individual player plays. Some players play in offensive situations where it is easier to get a positive Corsi and others play in defensive situations where a negative one is more likely.
At the top of the team adjusted Corsi list are Daniel Sedin and Henrik Sedin of the Vancouver Canucks with +258.8 and +248.8 ratings respectively. While these are players who are in the upper echelon of talent in the NHL, it is hard to argue that either were the best player in the NHL last year. Henrik finished 20th in scoring with 45 points and Daniel finished 31st in scoring with 40 points. Those are solid offensive numbers but they are not league leading and neither player is strong defensively. It is reasonable to expect that the Sedins might be among the best puck possession players in the league in 2013, but it is hard to justify them being on top of the list.
In my summer sabermetrics and hockey posts, I am looking at making sense of individual Corsi ratings. These are the difference between a team's attempted shots and their opponents while a player is on the ice in five on five situations. I posted the top 20 players (due to a tie it's actually a 21 player list) last season and observed that this group of players come from only seven teams and all of them are good puck possession teams. The top raw Corsi players play on good puck possession teams. That probably isn't a surprise. In order to make it more of an individual rating we must adjust for the team in which a player plays. That will not be the final step in making Corsi a meaningful individual rating because not all players play under the same circumstances. Even on the same team some players play in more offensive roles and some play in more defensive roles and that should affect puck possession while they are on the ice.
In order to make a team adjustment, we first must know team Corsi ratings. These can be used to find a baseline for a player on a given team. The baseline is the team Corsi divided by five (since five position players are on the ice at a given time). This baseline is subtracted from each player. In order to ensure that the adjustment is meaningful only players who played at least 30 games in the lockout shortened 48 game season are included.
Here are the top 20 players by team adjusted Corsi in 2013:
A while ago I published a list of the NHL teams by their team Corsi. This is the difference between the team's attempted shots and those taken by their opponents. Attempted shots include shots on goal, missed shots and blocked shots. This is recorded only in five on five situations. It is an indicator of puck possession and a repeatable portion of hockey. It is certainly not the only piece of success as special teams, goaltending, shooting, luck and other factors also can influence how well a team does.
The worst team in the league in the 2013 season was the Toronto Maple Leafs by their team Corsi. They posted a -522 team Corsi. This means that more than ten extra shots were taken by their opponents than by the Leafs. Despite this handicap, the Toronto Maple Leafs made the playoffs for the first time since 2004 and they forced the Stanley Cup finalist Boston Bruins to a hard-fought seven game comeback series win. As a result, most people think of the Leafs as a team on the rise but a more accurate representation may be as a flawed team that maximized their success last year and will be hard-pressed to ever match that success
There are three reasons for the Leaf success despite their poor puck possession. Toronto had the highest shooting percentage in the NHL in 2013 with 11.5%. Toronto had the second best penalty kill with an 87.9% success rate. Toronto had remarkably strong goaltending from James Reimer and Ben Scrivens despite the fact this was seen as a weakness and required an upgrade to bring in Jonathan Bernier via trade in the off-season.
During my summer look at sabermetrics and hockey, I have posted the team Corsi ratings. This is the difference between shots attempted for and against for a team in 5 on 5 situations. The team with the second best team Corsi was the New Jersey Devils. They posted a +417 rating. While doing that they missed the playoffs. They finished 11th overall in the East Conference with a 19-29 record with 10 regulation tie points. This was good enough for 48 points. That was seven points back of the last playoff spot in the East Conference. How can it be that a team with such good puck possession had such a poor record?
The simple answer is goaltending. Martin Brodeur and Johan Hedberg were a very weak goaltending tandem. The only goalie to play more minutes than Brodeur and post a saves percentage worse than his .901 was Ilya Bryzgalov. Similarly only one goalie played more minutes than Johan Hedberg and posted a worse saves percentage than his .883 and he was Miikka Kiprusoff. New Jersey had the worst goaltending in the NHL in 2013 - ranked by saves percentage. Very few people realized this because Martin Brodeur is a legend. He has had a great career and people have been slow to question him even though his play has deteriorated. Johan Hedberg has been even worse and he has not had the same great career. Hence he has been questioned and bought out by the Devils. He is likely going to be forced into retirement because nobody is interested in signing him.
About The Puck Stops Here
Who am I? A diehard hockey fan.
Why am I blogging? I want to.
Why are you reading it? ???