Yesterday I posted the team Corsi ratings for 2014/15. Leading the NHL was the Los Angeles Kings with a +760 team Corsi. They had 760 more attempted shots in five on five situations than their opponents. This is almost ten more per game. This makes the kings the top puck possession team in 2014/15.
The problem is these Kings posted 95 points in the season. That placed them ninth place in West Conference. That is one point out of the playoffs. In 2012, the Kings posted 95 points and that was good for eighth place in the West. They won the Stanley Cup. That year the Kings were 12 points out of first place in Corsi. I would argue that both teams are similar in quality.
How can the Kings either miss or barely make the playoffs with the best puck possession in the league? Puck possession isn't everything. They are average at special teams. Their shooting and saves percentages were roughly average last season. Their problem came in overtime and shootouts. Los Angeles had a 3-15 record when a game went to overtime. That was the worst record in the league in that circumstance. That result is so poor that it must be seen as a fluke. It isn't sustainable. In any normal season they would have had several more points in overtime and shootouts. Under those circumstances the Kings would have made the playoffs. As a playoff team they had a good chance to make a serious run. Los Angeles was unlucky to not make the playoffs. They were a team that could have made a significant run.
Today I will continue my summer sabermetrics series by looking a team Corsi ratings. This is the difference between attempted shots (shots on goal, blocked shots and missed shots) between a team and their opponents in 5 on 5 situations. It is important because it measures puck possession. This is a repeatable portion of hockey team success or failure. It doesn't measure everything about a given team. Special teams, goaltending and shooting percentages are not measured at all in Corsi studies. It should be that the better Corsi teams are the better teams in the NHL and the worse Corsi teams are the worse teams in the NHL. Those teams that are clear exceptions to that rule should, assuming roster continuity, likely regress toward the position their Corsi predicts next season.
Here are the team Corsi ratings in 2014/15:
My last post was to list the worst 20 players in 2014/15 by their raw Corsi rating. The worst player in the league was Rasmus Ristolainen of the Buffalo Sabres. He posted a -618 Corsi rating which is over seventy points worse than any other player in the NHL. Let's look at why Ristolainen does so badly. Why does he have the worst Corsi in the NHL? Why is his puck possession so bad? Why does the opposition team take the majority of the attempted shots when Ristolainen is on the ice?
The first reason is that Ristolainen plays for the Buffalo Sabres. They are the worst team in the NHL. They have the worst puck possession in the NHL. Since puck possession is a team based concept, it is logical that the player with the worst Corsi is a Buffalo Sabre. Why is Rasmus Ristolainen that player and not some other Sabre? He is not ready for the NHL. He was Buffalo's first round draft pick in 2013. He came to the NHL immediately. In his first season he split time between the NHL and the AHL. His NHL time was not memorable. He played 34 games scoring four points and posting a -15 +/- rating. Buffalo is a weak enough team that he stayed in the NHL the next season. He played 78 games last year. He was a defenceman who was easily beaten by his opponents. He scored 20 points and posted a -32 +/- rating. Ristolainen had the worst +/- rating on the Buffalo Sabres. He was the ice time leader in Buffalo as well.
Buffalo is such a bad team that their top player by ice time is a young player who isn't ready for the NHL at all. His development is probably being stunted by being forced into a situation where he is failing. If you have a player who is such a poor NHLer leading a bad team in ice time, he should have the worst puck possession numbers in the NHL and Ristolainen does.
I am continuing my summer sabermetrics series by posting the worst 20 players in 2014/15 by their raw Corsi rating. Corsi is getting more accepted by the NHL who started tabulating this stat on the NHL.com website this season. The NHL calls this SAT or shot attempts. It is the difference between attempted shots (shots on goal, blocked shots and missed shots) taken in 5 on 5 situations when a given player is on the ice by a player's team and his opponents. It is a measure of puck possession similar to +/- except that it measures all shots instead of only goals. Hence it gathers a much larger sample size, but there are questions about whether shot attempts are as meaningful as goals scored.
I have already posted the top 20 players by raw Corsi and found Drew Doughty led the league. Here is the flip side. The worst 20 players:
In the past I have criticized the Toronto Maple Leafs because they hired a coach in Mike Babcock without hiring a GM. Babcock was signed to an eight year $50 million contract that gives him more pay and a longer term in his contract than the GM. That looked like a problem. Perhaps the solution to that problem is to hire a GM who has no desire to stay employed for eight more years.
Lou Lamoreillo is the new Toronto GM. He stepped aside this spring in New Jersey after 28 years as general manager. He had been successful. He took over a team that was a bottom feeder and built it into a three time Stanley Cup winner. The problem is he is 72 years old. He basically retired from his last job. He is essentially a short term hire. He will not last a long time because age and health are issues.
A couple days ago, I posted the top 20 players by raw Corsi rating in 2014/15. At number one with a +410 rating was Drew Doughty of the Los Angeles Kings. Doughty had a good season. He finished second in the Norris Trophy voting and had the most first place votes of anyone in the league. He didn't have the same support as Erik Karlsson did among those who didn't place him first overall, so Karlsson won the award.
Karlsson was the top scoring defenceman and led his team into a somewhat improbable playoff berth. Doughty did not score as well and played on a team that missed the playoffs. When all was said and done, Ottawa had for more points than Los Angeles. As Corsi shows, Doughty's team did better in puck possession when Doughty was on the ice than Ottawa did with Karlsson on the ice. Of course there is more to life than puck possession. Actually scoring is important. Karlsson had a full twenty points more than Doughty. Defensively Doughty is the better of the two but can he make up a twenty point difference? Karlsson had the better +/- rating. Both players played against relatively tough competition, although Doughty played against a bit tougher. I would argue that the Norris Trophy voting went right. The right man won. Doughty's statistical argument comes largely from his Corsi, but Karlsson has a better season. His more traditional stats like points scored show that.
The NHL expected that if they put out calls for expansion applications that there would be enough applications submitted that they could be choosy. That looked good when sixteen different groups asked for applications, but only two actually submitted them on time. Las Vegas and Quebec City are the only applicants. The fee for a bid is $10 million and only $2 million is refundable. It is step one of the cash grab of expansion.
The NHL is a ponzi scheme. At least it appears to act like one. Each time teams expand into the league they pay larger and larger fees to the existing ones. This time around teams are being asked to pay $500 million to expand. This is more than some existing NHL teams are worth. The NHL wants this money because it is not money that has to be shared with players according to the bargaining agreement. If we assume two expansion teams with $500 million fees each we have $1 billion to split between thirty teams. This gives about $30 million to each club. That is enough to cover the reported losses of the Arizona Coyotes for a season.
I am continuing my summer sabermetrics series by looking at the top players in 2014/15 by their Corsi. In the past season this information has become available on the nhl.com website when looking at enhanced stats here. This makes Corsi more accessible to fans. On the NHL webpage it is called shot attempts or SAT. It means the difference in attempted shots (shots on goal, missed shots and blocked shots) for a team when a given player is on the ice and those allowed by their opponents in 5 on 5 situations. It is essentially a measure of puck possession as teams need to possess the puck to attempt a shot. In many ways it can be considered like +/- but with a larger sample size since it measures every attempted shot and not merely every goal.
Here are the top 20 players in 2014/15 by their raw Corsi:
I am happy to report that hockey sabermetrics is being taken more seriously by the hockey media. While that is a good thing it is not without problems. Here is one that comes from a Sportsnet post by Jonathan Willis. He writes about the 10 best NHL analytics free agent signings.
The problem with this post is that is that you cannot look at absolutely anything from an analytics standpoint and have it make and sense. The unrestricted free agent class this year is weak as is shown by my 2015 All UFA Team. This year's team is particularly weak. Even in a strong year we get a team that is significantly more expensive than the salary cap allows. The team has problems with its average age and is usually not good enough to be a solid contender. Thus if we signed all the best UFA players we wouldn't expect good results.
Yesterday I posted the worst 20 players by adjusted +/- in 2014/15 as part of my sabermetrics and hockey series. The three worst players on the list are the Toronto Maple Leafs first line of 2014/15 - Tyler Bozak, Phil Kessel and James Van Riemsdyk. Bozak and Kessel have -25.6 ratings and Van Riemsdyk is one point better at -24.6. As the Toronto Maple Leafs fell apart last year, it is clear that they had problems and this clearly shows one of them.
It isn't obvious which of the three Leafs was the biggest puck possession problem. It is hard to isolate individual players who play together most of the time.
Typically when a player has a poor adjusted +/- rating it happened in part because of bad luck. This is shown by a poor PDO (the sum of the shooting and saves percentage of all five players when a player is on the ice). No player can repeatedly control their PDO. Kessel and Bozak have 970 PDOs. This places them in the 50 worst PDOs among players who played 50 or more games last year. Van Riemsdyk at 977 is slightly better. These players are not the most unlucky in the world, but they are among the worst.
About The Puck Stops Here
Who am I? A diehard hockey fan.
Why am I blogging? I want to.
Why are you reading it? ???