Why Is Possession Important?

There have been a few questions about why EOTP, among others, are so negative about the hiring of Michel Therrien. Most of this has to do with the possession numbers. Possession is still a bit of a new term for hockey fans and people who aren’t familiar with it likely don’t understand its importance.

I wrote about it briefly last season, explaining how possession is the surest way to control performance, and account for random variance in things like shooting and save percentages. But the site has new readers now, and I think once in awhile revisiting topics to explain them in better ways is necessary.

Many people get thrown off by the terms Fenwick, Corsi or PDO (I won’t be using PDO today, but there’s a good explanation of it in the previously linked post), because they’re unfamiliar and not intuitive. You can’t read the name of those stats and know what they are. EOTP created a simplified glossary to help with that, but the concepts are actually very simple once you understand them.

Corsi is the easiest to understand, just think of it as total shot attempts. The more often a team has the puck in the opponent’s zone, the more shot attempts that team will have, and this is why it is a measure of possession. The more shot attempts a team gets, the more likely they are to get more shots on net, more scoring chances, and more goals.

Fenwick is similar, but shots that are blocked are not included. Because of this, Fenwick is a less reliable indicator of overall possession, but a better indicator of how effective that team’s possession is. In order to further refine this measure, we can look at the Fenwick score of a team while the score is tied or close (defined as within 1 goal in the 1st two period or tied in the 3rd), or when the game is on the line. This gives a better indication of how good a team is, because when the score is lopsided, the team with the lead is unlikely to worry about shooting much. This is what’s called ‘score effects’.

Due to the continued year to year drop in penalty calls in the NHL since the lockout’s initial outburst of powerplays, being adept at even strength play is all the more important. Most of the game is played at even strength, and it’s becoming harder and harder to make the playoffs without positive possession.

Unfortunately we can only go as far back as 2007-08 on Gabe Desjardins’ Behind The Net, but 5 years of data is enough to show the trend of even strength prowess becoming more and more important. To illustrate this, let’s examine how many teams made the playoffs each year with a negative Fenwick in both close and tied situations, and how many positive Fenwick teams missed the playoffs in the same situations.

*The asterisk is what’s troubling about Michel Therrien. Technically the Pittsburgh Penguins were a negative Fenwick team when they won the Stanley Cup in 2009, but that was only because they were heavily in the red during Therrien’s tenure. Once he was replaced, as this article explains, Dan Bylsma had them playing like a much better team.

As you can see from the table, 07-08 was a bit of a cutoff point. Before then, about 50% of the teams that made the playoffs got in with poor even strength play and great special teams and goaltending. The Habs were one of these teams in 2008, 2009, and 2010, but Jacques Martin had them playing within the top 7-10 teams in the NHL in 2011. This continued in the first 20 games this season, until Brian Gionta was injured and the team began to play poorly.

The Canadiens making the playoffs 4 times in 5 years while employing a strategy that hinged on special teams and goaltending after the lockout is likely why Habs fans are fairly reluctant to accept possession as a measure of team talent. But as you can see from the table, the results from 2008-09 onward are fairly consistent.

2-5 poor even strength teams make the playoffs each year, with the help of special teams, goaltending, and some good luck, while 1-4 good even strength teams miss the playoffs each year. To put it bluntly, a poor even strength team has just a 25.8% chance of making the postseason in the last 4 years, while a strong even strength team has an 81.3% chance of making the playoffs over that time.

Faced with these odds, it is beyond obvious that even strength play should be the primary focus in building a team. Now comes the big question, can Michel Therrien do it?

That question is why there has been so much negativity about the hiring. We can only use Fenwick for the last two years of Therrien’s tenure in Pittsburgh, and what we find isn’t encouraging.

With this roster in 07-08, Therrien lead the Pittsburgh Penguins to to a 45.76% share of total Fenwick shots while the game was tied (Worse than Montreal this season), and a 46.51% share while the game was close (Also worse than Cunneyworth’s Habs). Those scores are both 26th overall in the NHL that year.

With this roster in 08-09, Therrien lead the Penguins to a 49.45% share of total Fenwick shots while the game was tied, and a 49.97% share while the game was close. This includes the 31 games that Dan Bylsma coached, which brought up the possession numbers considerably. As was mentioned before, Therrien’s Pens had a Fenwick score of about 47% in 08-09. Bylsma’s boost brought the Penguins up to the 17th rank while the score was tied, and the 13th rank when the score was close.

For the rest of Therrien’s career, let’s take a look at shots alone:

The seasons with asterisks were seasons where Therrien only coached part of the season, so it’s more difficult to draw conclusions from them. However none of these seasons are inspiring much confidence. On the surface it looks like Therrien improved when he was in Pittsburgh, but the Canadiens team he coached didn’t have Sidney Crosby, Evgeni Malkin and Sergei Gonchar to lead it.

What we can see is that Therrien’s teams are consistently outshot, every year he’s coached, and by a wide margin. Therrien lead teams seem to be destined to suffer 30+ shots against per game as well, which isn’t a good thing. Only 5/16 teams to make the playoffs this season allowed more than 30 shots per game. Only 2 playoff teams allowed more than 31 shots against per game.

As you can clearly see, there are abundant reasons why the hiring of Michel Therrien has stats-oriented fans worried. Is it possible that while sitting in RDS’s studios that Therrien may have enhanced his knowledge of the game? Of course it is. Perhaps the time to reflect and learn about newer strategies of successful teams will yield a better result, but as yet we have no evidence of this. Everything we can actually see, looks bad.

It is my hope that Therrien makes these worries look foolish, and proves me wrong.