December 28, 2009
In The Crease
Even Strength and Penalty Kill Save Percentage

by Philip Myrland

Last week I looked at situational shots against on the penalty kill. This week I'll take a closer look at goalie performance on the penalty kill.

The first thing to be aware of when looking at special team statistics is that they tend to vary considerably because of the small sample size. Every team spends much more time playing at even strength than killing penalties, and over an entire season a goalie will only face a few hundred shots with his team down a man. Goalies that go on hot or cold special teams streaks or play on streaky units can have very different results.

To show this, it is necessary to first establish the baseline. Here are the top 10 goalies in the league in shorthanded save percentage from 2005-06 through 2008-09 (min. 75 GP):

Rank Goalie             Shorthanded Save Percentage
1.   Niklas Backstrom  .912
2.   Henrik Lundqvist  .890
3.   Chris Mason       .889
4.   Cristobal Huet    .886
5.   Martin Biron      .886
6.   Dwayne Roloson    .885
7.   Roberto Luongo    .883
8.   Martin Brodeur    .882
9.   Craig Anderson    .881
10.  Pascal Leclaire   .878

The league average over that span has been .865. The worst qualifying goalies were John Grahame (.825), Mikael Tellqvist (.833), and Andrew Raycroft (.835).

Niklas Backstrom is a major outlier, but there is good reason to believe that his team has a lot to do with it. Backstrom is in the top tier of goalies in even strength save percentage, but he does not stand out from the pack like he does in shorthanded situations. His primary backup since the lockout, Josh Harding, has a .885 shorthanded save percentage, meaning he would have made this list with more playing time. Dwayne Roloson also had a better save percentage in Minnesota (.887) than in Edmonton (.881). All this supports Gabe Desjardins’ shot quality research that suggests that Minnesota prevents its opponents from getting many high quality shots against in comparison to the average NHL team. If we treat Backstrom as an exception, then over four seasons of play the established performance range is approximately .830 - .890.

Here are the current leaders in penalty kill save percentage (minimum 10 games played):

Rank Goalie             Penalty Kill Save Percentage
1.   Ty Conklin        .951
2.   Tuukka Rask       .932
3.   Antti Niemi       .929
4.   Jaroslav Halak    .923
5.   Ondrej Pavelec    .912
6.   Evgeni Nabokov    .907
7.   Brian Elliott     .906
8.   Niklas Backstrom  .901
9.   Tim Thomas        .901
10.  Mathieu Garon     .900

And here are the trailers:

Rank Goalie             Penalty Kill Save Percentage
40.  Semyon Varlamov   .844
41.  Jonathan Quick    .841
42.  Chris Osgood      .841
43.  Dan Ellis         .840
44.  Jose Theodore     .827
45.  Mike Smith        .823
46.  Ray Emery         .821
47.  Martin Biron      .786
48.  Vesa Toskala      .773
49.  Brian Boucher     .769

Some of those names are perhaps unexpected, but others are whom we expected to see. It's no surprise to see the defending Vezina Trophy winner in the top 10, for example, or to see goalies like Vesa Toskala or Chris Osgood near the bottom of the list given their play this season.

There is some evidence of potential team effects. Boston has both goalies in the top 10, while Chicago places both of their netminders in the top 14. At the other end, the goalie tandems in Philadelphia and Washington both have had very poor performances so far. Backstrom is again among the leaders, which is likely aided by the Wild’s excellent penalty kill unit, which I discussed in a previous post article on how the L.A. Kings’ poor penalty kill likely has had an adverse effect on Jonathan Quick in his career so far.

The goalies with results above .890 or below .830 have been either really hot or really cold so far this season, and will likely not maintain the same pace for the rest of the season. Ty Conklin is not going to continue to save 95% of shorthanded shots against, and Philadelphia’s opponents won’t continue to score on 1 out of every 5 shots they take while the Flyers are shorthanded.

In general, a goalie's save percentage on the penalty kill will tend to follow their save percentage at even strength. On average, save percentages on the penalty kill are around .050 lower than save percentages at even strength. Chris Osgood is an example of a goalie who has had a low save percentage this year while shorthanded (.841), but his even strength save percentage is also pretty low (.908), which suggests that he is simply playing poorly this year in all situations. On the other hand, Semyon Varlamov is narrowly ahead of Osgood in shorthanded situations (.844), but is up among the league leaders in even strength save percentage (.941), which indicates that he has mostly played well but has struggled a bit on the penalty kill. If Varlamov continues to play well it is likely that his penalty kill performance will also follow suit.

Based on these examples, perhaps a better way to look at who is underperforming or overperforming is to compare goalies in terms of their differential between even strength save percentage and penalty kill save percentage.

Largest differential:

Legend:

ESSV%: Even Strength Save Percentage

SHSV%: Shorthanded/Penalty Kill Save Percentage

Differential: Difference between Even Strength Save Percentage and Shorthanded/Penalty Kill Save Percentage

Rank Goalie             ESSV%  SHSV% Differential
1.   Brian Boucher     .933   .769   -.164
2.   Martin Biron      .926   .786   -.140
3.   Vesa Toskala      .889   .773   -.116
4.   Manny Legace      .919   .809   -.110
5.   Ray Emery         .923   .821   -.102
6.   Semyon Varlamov   .941   .844   -.097
7.   Mike Smith        .914   .823   -.091
8.   Jose Theodore     .915   .827   -.088
9.   Johan Hedberg     .939   .853   -.086
10.  Tomas Vokoun      .937   .854   -.083

Smallest differential:

Legend:

ESSV%: Even Strength Save Percentage

SHSV%: Shorthanded/Penalty Kill Save Percentage

Differential: Difference between Even Strength Save Percentage and Shorthanded/Penalty Kill Save Percentage

Rank Goalie             ESSV%  SHSV% Differential
40.  Jeff Deslauriers  .909   .893   -.016
41.  Antti Niemi       .941   .929   -.012
42.  Cam Ward          .895   .884   -.011
43.  Jaroslav Halak    .926   .923   -.003
44.  Tuukka Rask       .931   .932   +.001
45.  Mathieu Garon     .899   .900   +.001
46.  Scott Clemmensen  .885   .886   +.001
47.  Ondrej Pavelec    .906   .912   +.006
48.  Brian Elliott     .897   .906   +.009
49.  Ty Conklin        .925   .951   +.026

Philadelphia and Washington both have a much better save percentage at even strength than on the penalty kill, which indicates that the teams have struggled on special teams. However, since the goalies have been able to make saves during 5 on 5 it seems likely that their performance will likely improve as well while shorthanded, and that should help the Flyers and Capitals going forward.

Goalies like Martin Biron and Tomas Vokoun are very likely to see their penalty killing results improve over the rest of the season, since they have an established track record that is substantially higher than what they have managed to date. I would also expect goalies like Conklin, Rask, Niemi, Halak, Pavelec and Elliott to be unable to maintain their pace on the penalty kill. Over a short sample size goalies can get hot, but the more shots they face in dangerous power play situations the more likely it is that their performance will regress to the mean.

Philip Myrland is an author of Puck Prospectus and runs the statistical hockey website Brodeur Is A Fraud. You can contact him at BrodeurIsAFraud@Inbox.com.

Philip Myrland is an author of Hockey Prospectus. You can contact Philip by clicking here or click here to see Philip's other articles.