By the Numbers: A Dominant Performance Wasted

I talk a lot in these columns about a statistic called expected goals. I almost always pull these stats from a website called Corsica.hockey, a site run by hockey stats blogger Emmanuel Perry. You can follow him on Twitter here.

The reason that I am so fond of using this measure is because it bridges the gap between shot quantity metrics (corsi, fenwick, etc.) and arguments relating to shot quality made by many that appeal to the eye test.

In 2013, you might have heard a Maple Leafs fan make an argument that said something like, “Stats like corsi can’t measure how good a team like the Leafs are, since they emphasize looking for good chances over just throwing random shots at the net every chance they get.”

This was a common argument made by people who bought into the success of teams with poor shot quantity metrics. Now, it’s harder for those arguments to be made, since we now have this stat that accounts for this.

You may be under the assumption that a stat like expected goals would only take into account shot distance, as I was when I first heard about it, but that isn’t the case. Corsica’s model of the stat, for example, uses six different factors in determining the goal expectancy of a given shot:

You can read the rest of Perry’s explainer for his expected goals model here.

Why does all of this matter? Well, for a team like the Hurricanes who have long struggled to translate strong shot quantity metrics into actual wins, it’s helpful to know whether or not they are appreciably controlling shot quality as well.

We’ve often seen the Canes lose games in which they’ve dominated possession, but what if that’s just because they haven’t had the talent to create quality shots?

Last night against the Penguins, the Hurricanes controlled 67.19% of the 5-on-5 shot attempts in a game they lost 3-2, but were they able to match that dominance in expected goals? Well, it turns out they actually topped it with an expected goals share of 72.65%. This means that not only were the Canes taking more shots than the Penguins, but the shots that they were taking were also better shots than those taken by Pittsburgh.

Let’s delve a little deeper into what went into that yesterday.

A Performance For the Ages

Last night’s game saw the Hurricanes deliver one of the best offensive performances of the advanced stats era. Last night’s game saw the Hurricanes score two goals.

Both of those things can be true at the same time.

According to Corsica’s model, the Canes were expected to score 5.41 goals last night. Of those 5.41, only 0.03 came with the goalie pulled. That means that at 5-on-5 play, they were expected to score 5.38 goals. As for the Penguins, they were expected to score 2.04 times last night.

To put that into perspective, there have been 9,860 total NHL games played since the start of the 2010-2011 season. Of those, 9,860 last night’s performance against the Penguins ranks sixth in 5-on-5 expected goals. Of the five games that ranked ahead of it, four of them were wins. Two of those four wins were playoff games that went to double overtime.

This means that there have been 19,720 team performances in that time frame, and 19,715 of them (99.97%) have been less prolific offensively than what the Canes did in a loss last night.

So while last night may have been a loss, there really isn’t much that can be taken from that game and seen as a negative. You could argue that Matt Tennyson’s overplay of Patric Hornqvist’s zone entry that led to Carl Hagelin’s game winning goal cost the team a point or two, and you’d probably be right.

You could also say that maybe Ward could have come up with a save or two. Since Kunitz’s goal was his only shot on goal last night, and he individually came up with .12 expected goals, we know that there should have been roughly an 88 percent chance that Ward stopped that shot.

But overall, last night was without question the best road game that the Canes have played all season, and probably the best game, period. They did a good job of limiting the scoring chances of the best offensive team in the league, and they took advantage of a makeshift defensive unit while generating an absurd amount of offense, even though they were stonewalled by two hot goalies.

If they can find a way to come anywhere close to bringing that sort of effort on a nightly basis, they’ll win far more often than they lose, and it’d be hard to imagine this team missing the playoffs.