Eastern Edge: Regression in Save Percentage and Goalie Targets

by Cam Metz on June 26, 2018
  • Eastern Edge
  • Eastern Edge: Regression in Save Percentage and Goalie Targets

It pains me every year to start looking at goalie stats, picking goalies is notoriously difficult.  Do you hate picking goalies so much that you’ll sacrifice a first round pick on Carey Price just so you can set it and forget it? Right, and then last season happens are you’re sitting there with the thoughts of what could have been a great fantasy season.

Regardless of my disdain for goalies, they do end up for better or worse making up a significant portion of your draft strategy and ability to compete. For example when I was drafting under pressure last year Marc Andre-Fleury very nicely fell to me in the 12th round of a 14-team league. I would say that’s a nice way to have things fall, but if we’re being completely honest MAF was actually on my list of players to draft in that position – but so was Robin Lehner. Oh I also over drafted Scott Darling last year as well, so finding one goaltender out of three is not particularly useful; however this logic would have helped you stay away from Henrik Lundqvist.

The goalie stats I’m interested in are shorthanded SV% compared to the league average -it is a great way to find someone who was unreasonably lucky and accordingly someone you should tank in your rankings.  The flip side of that is someone who has a great 5v5 SV% but couldn’t get any luck on a team you suspect is improving (again Buffalo was a swing and a miss here). With that preface I’d like to introduce you to a table that breaks down the league average 5v5 and shorthanded save percentages for goalies with a minimum of 20 games played.

I’m going to include four stats that I believe have value in picking goaltenders of value for next year. The first stat is adjusted save percentage; this stat takes into account the location of a shot and is a simplified version of saying that a net front shot has more danger than a soft wrister from the point.

The other three stats are Low, Medium, and High-Danger save percentages. These percentages help make up the adjusted save percentage but by breaking them out we can see if a goalie was unreasonably lucky or unlucky in any one of the three categories.

2017-2018 Average Adjusted Save, Low-Danger, Medium-Danger, and High-Danger Save Percentage



    LD SV%

       MD SV%

     HD SV%

5 v 5











Knowing that save percentage as a whole is tough to repeat year after year if a goalie had a down year last year but was at least league average at 5v5 they could provide some value.  So a shorthanded save percentage drastically above 77.77% in the High-Danger area would sound the alarm that most likely unless you are John Gibson there is absolutely no way you’re going to be that lucky next year (regression is always a bet I am willing to make).

Unfortunately for the sake of consistency – most of the really fun goalies to talk about are in the West.  But for arguments here are some of the goalies I am interested in going into next year:


Carey Price – A lot of this is going to depend on ADP and just how low Bergevin can sink this team, but for my money Carey Price is not as bad as his High-Danger SV% differences from the league mean represent.  This is bound to regress and those willing to take the early round gamble will gain handsomely in SV%.  Granted the wins won’t be there this year – so you’ll just have to live with that.  Compared to his peers in the East he was also one of the bottom-three in save percentage – almost unfathomable. 

Tuukka Rask and Henrik Lundqvist – I’m sorry to break the news to Bruins fans but it feels like a coming back to earth party is in store this year, much like it was for the Rangers with Lundqvist last year.  Each player greatly outplayed the league average in terms of High-Danger SV% shorthanded (Rask led the league 9 points higher than his peers). Any reversion to the mean for these goalies and they won’t even have a spot on their own teams let alone your fantasy team.

Sergei Bobrovsky – I would say that Bobrovsky is the one player that I should relinquish my prior disappointments with and move forward with what he offers next year. One of the more exciting aspects of Bob’s SV% comparison is that he played shorthanded Medium-Danger scoring chances very poorly, regression could be his friend considering his HD SH% was not that out of line with league averages.

Frederik Andersen – I like Anderson to return value next year, I think he’ll be one of the best options in net league wide. His numbers shorthanded are not outrageous for a top ten goalie and his 5v5 HD SV% compared to the league has the opportunity to improve.  If he is able to swing that 5v5 HD SV% into the positive one percent it would result in him saving almost 32 more goals.  Granted his shorthanded HD SV% could certainly tank and add those goals back to his total allowed.  Either way I think Anderson is in a sustainable situation – confidence in net is worth drafting.

I know, I know, John Gibson plays in the West – but if you’re doing a goalie piece talking about shorthanded SV% you have to mention his work over the last couple years.  Gibson holds and Adj Sv% SH of 91.88; to put that in perspective he was better shorthanded than 14 goalies at 5v5 in the East last year. INCREDIBLE.

Below is some of the data adjusted to show the difference between the goalies performance and the league average last year.

2017-2018 Average Adjusted Save, Low-Danger, Medium-Danger, and High-Danger Save Percentage Difference from League Average



More from Eastern Edge: Looking at Rebounds Created