Kukla's Korner

The Puck Stops Here

Lowest Scoring Team

For most of the season the Los Angeles Kings have been the lowest scoring team in the league.  Likely that is what motivated them to trade for Jeff Carter.  Recently they have lost their position as the lowest scoring team in the league.  The Minnesota Wild has even fewer goals per game.  LA has 2.12 goals per game (that is a reduction in their scoring rate since I first wrote about them in December).  Minnesota has 2.08 goals per game. 

Minnesota’s top scorer is Dany Heatley.  He has 46 points on the season.  That makes him the 68th highest scorer in the league right now.  The only other player on the team with a significant scoring rate is Mikko Koivu.  He has 37 points in 45 games.  He has a shoulder injury and is out and indefinitely.  Nobody else on the team is scoring more than a point every other game.  Minnesota is not a team with much scoring power.

Minnesota has been an interesting team this season.  They started off extremely well.  They were first overall in the league in late November.  Since then things have fallen apart.  They are currently third last in the West Conference.

The general explanation for things is that Minnesota overachieved at the beginning of the season.  They were winning at an unsustainable rate.  Corsi analysis showed this.  They were getting badly outshot in their games.  They were allowing their opponents possession of the puck.  That is not the way to remain ca first place team.  A team getting badly outshot (including missed and blocked shots for Corsi analysis) is a bad team and will regress to that level in time.  Kent Wilson of Puck Daddy wrote a great piece explaining this.  This prompted the people at hockey wilderness to write a juvenile response article to say we don’t care that you stats people predicted what would happen.  As often happens on the internet, fans are partisan.  When somebody says something negative about their team they see it as an attack and fight against it.  It doesn’t matter if the negative thing is factual; they still deny it and fight against it.  In this case the best they could do is go through the trouble of writing a lengthy article to tell you they “don’t give a damn”.  Their actions contradict their words.

Minnesota had a strong start to the season.  It was an improbable result that was not sustainable.  Now they find themselves the lowest scoring team in the league and well back in the playoff race.  The underlying numbers always suggested this is where they would most likely be even when they were first overall.  This is a success of Corsi analysis.

Filed in: | The Puck Stops Here | Permalink
 

Comments

Avatar

I appreciate how you look into how statistics can be used to predict a teams success, but every time I read your articles I feel like something is missing. I can understand and respect the conclusion that the Wild’s early season success was an aberration, which can be seen in the statistical analysis. The question you don’t seem to touch though is why the success? It is strange that for 40 games the wild produced better than any other team in the league despite the poor in game statistics. Where is the stat that can explain their prolonged success in the early part of the season. Saying it was just dumb luck feels lazy and irresponsible.

Posted by BNWild on 03/05/12 at 09:54 PM ET

PuckStopsHere's avatar

It was far fewer than 40 games when the Wild looked good.  They were in first place with a 13-8 record (three regulation tie points).  That is 21 games played. 

Let’s say that over the longterm Minnesota should win 45% of their games and lose 55% (we will neglect regulation ties).  There is a slightly over nine percent chance (using a binomial distrubition) that Minnesota would have 13 or more wins in 21 games.  That result is quite probable.  Given that there are 30 teams in the NHL, in any given 21 game stretch of a season 2 or 3 teams should overachieve by that amount.

There really isn’t anything else needed to explain things.

Posted by PuckStopsHere on 03/05/12 at 10:17 PM ET

Avatar

At 35 games the wild Were 20-10-5 which placed them 2nd in the West and 4th in the league. By saying it was far fewer than 40 it feels like you may be picking the data so it fits nicely in your hypothesis. If I still follow your theory that it was just overachieving during that stretch, and that at anyone time 3 teams in the league have the possibility to produce those numbers in your data, then I wonder if you have looked into figuring out if their is any way to predict which teams are most likely to overachieve in a given period?

Posted by BNWild on 03/05/12 at 10:39 PM ET

PuckStopsHere's avatar

35 games doesn’t really make it harder to explain.  A .450 team will win 20 of 35 games or more about 10% of the time by the binomial distribution.

If this over or underachieving is drivwen by luck then it is not predictable.  Which team will be lucky?  How do you predict that?

Posted by PuckStopsHere on 03/05/12 at 11:13 PM ET

Avatar

I guess what I’m saying is that luckiness doesn’t work as an explanation for me. You are not flipping a coin with limited variables. A hockey team should have more factors involved that can be used to predict the level of play the Wild sustained at the start of the year.

Posted by BNWild on 03/05/12 at 11:25 PM ET

tuxedoTshirt's avatar

It’s true Puck.  I like the way you’ve rigged your jib-sail, but the un-quantifiable aspects of hockey are both more interesting and more important than stats.  Not to say that your number juggling doesn’t have an important role, especially when attempting to validate a hypothesis about intangibles.

Posted by tuxedoTshirt from the Home of the 1937 World Champions on 03/06/12 at 03:59 AM ET

Avatar

Posted by BNWild on 03/05/12 at 09:25 PM ET

I would say that, even though hockey teams are complex constructions composed of complex individuals with a potentially infinite number of variables, not all variables are equal.

A team’s ability to control the puck and outshoot its opponents is one of the most important variables in a game. It’s not the only variable; just because one team is slightly better in overall Corsi doesn’t necessarily mean it’s better overall, or better when both teams are playing their best. Certainly, other factors are involved—and not all of them are quantifiable, at least using current methods.

But I think it’s safe to say that if your team is routinely among the worst in the league in both puck possession and shot differential, no amount of other factors besides dumb luck can make you one of the best teams in the NHL, with the possible exception of an absolutely dominant goalie.

“Luck” isn’t coin flipping. “Luck,” let’s say, is any outcome that cannot be predicted, controlled or easily repeated through normal human processes. That happens a lot in hockey. With so few goals scored per game, and so many goals the result of chance or individual fortune rather than sustained possession, it’s relatively common for a team that gets outplayed for most of the game to win anyway, and vice versa. Over long time scales, those instances tend to cancel out, but over shorter samples one may win or lose much more often than one’s level of play would otherwise “deserve.”

For example, if you have the talent of a .500 free throw shooter, that doesn’t mean you’re always going to make every other bucket. You’ll probably go through stretches where you hit three or four straight, or 13 of 21, or 25 of 40. Over small samples, variation can be huge.

Just because there are more variables involved within a hockey team doesn’t mean that same principle isn’t true. More variables doesn’t mean luck becomes less important, it just means there are more variables.

Posted by Sven22 from Grand Rapids on 03/06/12 at 02:50 PM ET

Avatar

The question you don’t seem to touch though is why the success?

Insanely good goaltending from two average-to-above-average, but not elite, let alone .940+, goalies.

Posted by Ralph on 03/06/12 at 03:04 PM ET

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.

Smileys

Notify me of follow-up comments?

Feed

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