As I am sure everyone on this blog is well aware of, the Canucks are having a very solid season thus far and are looking like a solid bet to make the playoffs this year. The Western Conference has so many great teams along with Vancouver such as the Ducks, Blackhawks and shockingly, the Nashville Predators. Rather than reiterating some of the points this post makes, I am going to take a brief look at how the Canucks are performing according to advanced statistics. Keep in mind that hockey is very far behind other sports such as baseball and basketball in terms of analytics, mostly due to the fluid nature of the sport and the lack of current technology. Soon, however, this will change.
As westy99's post from a couple weeks ago mentioned, an easy way to assess the likelihood of making the playoffs is to look at current record and how it compares with making the playoffs historically.
As westy99 also notes, individual stats do not tell the full story on the ice. We are far from determining how to quantify individual success in hockey aside from traditional box score stats. These stats are used as a supplementary representation of what happens on the ice. A combination of both areas are key in making the best estimation of team and player performance.
Pythagorean Winning Expectation: Developed by Bill James, Pythagorean winning expectation believes that run differential (or in hockey's case) goal differential, is a better predictor of future success than wins and losses. This method fails to account for blowouts, which can have a large effect on the calculations.
The formula for hockey's Pythagorean winning expectation is: (Goals for)^2/([Goals for]^2+[Goals against]^2)
How do the Canucks fare?
The Canucks, from this formula, are expected to have a winning percentage of 51.8 percent of their remaining games. This fails to account for shootout losses, which can easily determine seeding or who gets a final playoff spot. What I prefer to look at, rather than calculating various winning percentages, is to compare the goal differentials of teams that are fighting for the same spots. In our division, the Canucks are ranked third with a +6 goal differential. What's interesting here is that:
1. San Jose is (barely) second in the division despite a -1 goal differential.
2. Calgary, a team that isn't talked about very often, has a very strong +14 goal differential. Playoff spots are so tight but it is a bit odd that Calgary hasn't distinguished themselves.
Other Predictors of Team Performance:
I won't get too deeply into other predictors such as Corsi percentage (not a huge fan of Corsi personally), but here are a few small predictors to keep in mind while qualitatively assessing the team's performance:
1. Shootout record: This is definitely arguable, but I tend to believe that shootouts are basically a crapshoot. If someone has a study to disprove this please let me know. A team that has a particularly strong or weak shootout record is likely to regress to a 50% winning percentage for their remaining shootouts, since it is assumed that all shootouts are independent of one another.
Notable teams affected: The LA Kings are 1-7 in shootouts this season. I doubt this will continue.
2. Number of home/road games played: Knowing a team that is very strong at home or on the road has either a very high or very low amount of games in their preferred environment should affect projections for the rest of the season.
Notable teams affected: The Canucks have 20 home games and 16 road games, which may not be helpful considering Vancouver has been stronger on the road this season(15-8-2 vs. 11-9-1). The LA Kings have 13 home games and 21 road games left, which hurts their chances significantly since they are much better in LA than on the road (12-6-6 vs. 5-9-6).