One of the greatest predictors of future success is monitoring a team’s Score Adjusted Fenwick percentage.  

This stat is basically a metric that takes the percentage of unblocked shot attemps a team takes relative to league average at different score situations.  What that means is that instead of taking Fenwick data from specific score situations, it takes data from all even strenght situations and weighs each Fenwick event accordingly.  By using data from all scoring situation you are collecting a larger sample size from each game.  Because it uses all of the even strength Fenwick events you don't need as many games to make accurate predictions 

Travis Yost of TSN wrote a piece about the predictive powers of Score Adjusted Fenwick percentage (SAF%) for the Sporting News last year that highlighted the predictive powers of this statistic  Here is an excerpt: 

We only have seven years of data available, but the results are impressive. If you simply picked the better Score-Adjusted Fenwick% team in the last 20 games of the regular season, your record since 2008 is 71-33 (68.2%). Since picking the home team in every series would net you a 56% success rate, I'm inclined to believe there's something here. 

Last year the teams with the higher regular season SAF% were an impressive 12-3 in winning playoff series.  That is a winning percentage of 75%.

The only upsets were Montreal (48.33%) over Tampa Bay (51.76%), Pittsburgh (50.03%) over Columbus (50.29%), and Montreal (48.33%) over Boston (54.17%).

The Stanley Cup Champion Los Angeles Kings were #1 overall at 56.24% and the Chicago Blackhawks were #2 overall at 55.79%.

Given this predictive power I thought I would take a gander at how things shaped so far this season through the first two weeks of game play. 

Rank

Team

SAF%

1

Minnesota

59.91%

2

Pittsburgh

59.08%

3

Chicago

58.89%

4

St. Louis

58.14%

5

Boston

56.12%

 

While it may be surprising to see Minnesota at the top of the list they have been actively shifting from a dump and chase team into one that values the puck more.  In fact, Zach Parise spoke specifically about his efforts to study the data over the summer.  The other four teams are not very surprising and you could certainly expect them to be near the top. 

 

Rank

Team

SAF%

26

Toronto

45.07%

27

Colorado

43.48%

28

Calgary

41.58%

29

Ottawa

41.50%

30

Buffalo

33.29%

 

Much like the top of the list the bottom of the list does not come with any true surprises.  Buffalo is a special kind of bad and that is probably something they are completely content with.  Toronto and Colorado seem to be plagued by the same issues that impacted their play last year.  Ottawa is going to be held back by their below average defense corp. this season.  They need to do something about that if they want to improve (Andrej Sekera anyone?).  

These sample sizes are incredibly small, each of these teams have only played between 4-7 games.  This exercise is more or less meant to highlight the importance of the statistic and how you can make accurate predictions with it.  As the season wears on and more data is collected we should get a better idea of where the teams stand.  Starting at the 20 game mark you can start to figure out which teams are for real and which teams are pretenders.

For those folks who like wagering on the games, SAF% should be one of your best friends.

 

Thanks for reading!

 

Do you like the content at Hockey Hurts? Would you like to support us? Do you buy stuff on Amazon? If the answer is yes to all of the above click on the Amazon banner at the top of the page anytime you are looking to make an Amazon purchase. It doesn't cost you anything and we get a little kick back for you going through our link. Thanks!

FacebookTwitterDiggRedditNewsvine