Kukla's Korner

The Puck Stops Here

Why Edmonton Should Miss Playoffs

When I gave my 2008/09 predictions, one of the most criticized picks I made is that the Edmonton Oilers should finish last in the Northwest Division and 11th in the West Conference.  The argument against this is simple.  Last season,  Edmonton finished 4th in the Northwest Division and 9th in the West Conference.  They finished strong with a 14-6 record in their last 20 games (one game was a shootout loss and counts as a regulation tie).  During the off season, they added Lubomir Visnovsky and Erik Cole, giving up Joni Pitkanen.  They have many talented young players such as San Gagner and Andrew Cogliano, who should continue to improve.  How can anyone say they will be worse this season?  The answer is that last season’s result was misleading.  I argue that they will be a better team and not be as lucky in their point total.

Edmonton’s 2007/08 record is significantly distorted by their shootout success.  They appeared in a league leading 19 shootouts last season and they won 15 of them.  The New York Rangers were second in shootout appearances with 17.  The median NHL team had 10 or 11 shootout appearances.  In all likelihood, the Oilers will appear in less shootouts this season than they did last year.  If they depend on shootout wins for points, some of their opportunity should go away.

It is worse than that; I am not convinced there is a skill that teams can carry over from one season to the next to win in shootouts.  In 2005/06, the three teams with the best shootout winning percentages were the Dallas Stars (12 for 13), Los Angeles Kings (6 for 7) and Carolina Hurricanes (8 for 10).  In 2006/07, those three teams combined for a 13 for 28 record in the shootout.  They lost more shootouts than they won.  The new top teams were the Tampa Bay Lightning (10 for 12), Buffalo Sabres (10 for 14) and Phoenix Coyotes (5 for 7).  Last season, those teams were a combined 11 for 27 in the shootout.  Last year, Edmonton’s 15 for 19 record was the best shootout winning percentage in the league (with the highest single season total of shootouts since they were added to the league).  What is the chance that it is a skill that can be retained into next season?

It looks like the best model for shootouts is that they are random events - at least on a longer term basis from one season to the next.  Teams that do well one season tend to not do well the next year.  In fact, teams that lead the league one year tend to lose more than they win the next year.  Is there any reason to expect that Edmonton should have the best shootout winning percentage this season?  No.  There is no evidence at all that teams can maintain this “skill” from one year to the next.  It is certainly a stretch to imagine that Edmonton will be the first team ever in NHL history to do so. 

How good was Edmonton really, if we take out the shootouts?  One way to check this is to use the Hockey Analysis Power Rankings.  They calculate winning percentages when shootouts are considered ties and there are no points for overtime losses.  They have Edmonton with a .390 winning percentage.  This puts them in 26th place (before they do some further normalization for schedule strength).  If Edmonton was a 26th place team last year that was lucky to play in a lot of shootouts and lucky to win those shootouts, it makes the most sense to start them as a 26th place team in our predictions and move them upward as their improved roster allows.  They didn’t make sufficient moves to jump from a 26th place team into a playoff berth. I think my ranking of them is approximately right.

When making predictions, people often use the simple method of starting with the final standings from last year and adjusting a team up or down based upon expected contributions from new and departing players.  This can be over-simplistic.  It makes the assumption that last season was repeatable.  It assumes that no distortions in the statistics occurred for any reason.  In reality, of course distortions exist.  Rarely are they as large or as easy to identify as the 2007/08 shootout record of the Edmonton Oilers.  Upon identifying that distortion, it is very reasonable to predict that Edmonton will not be a playoff team in the upcoming season.

Filed in: | The Puck Stops Here | Permalink


Be the first to comment.

Add a Comment

Please limit embedded image or media size to 575 pixels wide.

Add your own avatar by joining Kukla's Korner, or logging in and uploading one in your member control panel.

Captchas bug you? Join KK or log in and you won't have to bother.


Notify me of follow-up comments?


Most Recent Blog Posts

About The Puck Stops Here

imageThe Puck Stops Here was founded during the 2004/05 lockout as a place to rant about hockey. The original site contains over 1000 posts, some of which were also published on FoxSports.com.

Who am I? A diehard hockey fan.

Why am I blogging? I want to.

Why are you reading it? ???

Email: y2kfhl@hotmail.com