Yesterday we looked at the league’s most consistent and most inconsistent goalies, picking at the extremes of an oft-discussed but rarely looked at issue. With those players out of the way, it’s time to consider the whole picture: every goalie in the league (minimum 25 games played) ranked by game-to-game save percentage standard deviation (the length of that phrase is why I just use the word “consistency).

For those without a stats background, standard deviation is just the area on a graph where 34.1% of results fall. A goaltender with a low standard deviation plays most of his games near his average level of performance, while a goalie with a high standard deviation jumps higher or lower than the average.

Player SV% Avg. SV% Std. Dev. (102)
Jimmy Howard 0.924 0.918 4.78
Craig Anderson 0.923 0.921 4.99
Tuukka Rask 0.926 0.923 5.02
Carey Price 0.911 0.901 5.33
Jonathan Quick 0.908 0.904 5.38
Ryan Miller 0.932 0.926 5.46
Jonas Hiller 0.917 0.911 5.55
Henrik Lundqvist 0.920 0.912 5.80
Jean-Sebastien Giguere 0.904 0.897 5.97
Evgeni Nabokov 0.927 0.921 6.07
Miikka Kiprusoff 0.922 0.916 6.11
Michael Leighton 0.908 0.900 6.33
Tomas Vokoun 0.931 0.924 6.40
Dan Ellis 0.911 0.906 6.45
Marc-Andre Fleury 0.906 0.897 6.57
Chris Mason 0.912 0.904 6.61
Brian Elliott 0.908 0.900 6.78
Jaroslav Halak 0.925 0.919 6.83
Ilya Bryzgalov 0.920 0.911 7.06
Antero Niittymaki 0.915 0.904 7.18
Mathieu Garon 0.900 0.888 7.26
Cam Ward 0.913 0.905 7.27
AVERAGE 0.911 0.901 7.34
Niklas Backstrom 0.903 0.893 7.42
Tim Thomas 0.917 0.908 7.45
Ray Emery 0.905 0.900 7.54
Mike Smith 0.900 0.888 7.55
Marty Turco 0.910 0.897 7.77
Jeff Deslauriers 0.903 0.892 7.84
Roberto Luongo 0.917 0.903 8.25
Ondrej Pavelec 0.901 0.884 8.69
Pascal Leclaire 0.892 0.881 8.83
Johan Hedberg 0.914 0.902 8.94
Dwayne Roloson 0.907 0.889 9.04
Martin Brodeur 0.913 0.896 9.14
Vesa Toskala 0.877 0.853 9.43
Jonas Gustavsson 0.899 0.886 9.87
Jose Theodore 0.910 0.895 9.91
Steve Mason 0.899 0.884 10.08
Pekka Rinne 0.901 0.884 10.16
Cristobal Huet 0.898 0.880 10.63

A few things stand out to me here:

- Martin Brodeur’s place on this list isn’t surprising to someone who has followed his season to date, but it certainly isn’t a spot I would have expected him to occupy at the start of the year. He’s been Mr. Consistent from season to season, but he hasn’t been this year.

- There is some definite correlation between consistency and ability; there are some exceptions (Luongo, Hedberg and Brodeur, for instance) but for the most part the good goaltenders have very few off games and bad goaltenders vary more. This isn’t surprising, but it’s nice to see some firm evidence.

- For young goalies with bad numbers, I’d suggest that sitting near the bottom of this list is a good thing; not because inconsistency is a virtue but because it’s much better to be bad with flashes of brilliance than consistently awful.

Comments (5)

  1. You assume goalies are normal and we all know that just isn’t true.

  2. Goaltenders are the most trickiest player to figure out stat wise. Simple because their stats are greatly effected by the team in front of them. You think Brodeur win’s all those games and gets all those shut outs if not for Lou putting all those good defensive teams infront of him? Probably not the same guy if he had played in Florida, (Like Luongo did) or somewhere else. That’s why goaltenders can get a bad rap in a game simply because the defense leaves them hung out to by leaving a guy wide open or missing their assignment and next thing you know you’ve allowed 4 or 5 goals shooting your G.A. up and your S.P. down.

  3. Is that standard deviation multiplied by a thousand or something in order to give a more useful number?

  4. Doogie2K: By 100, yes (that 10^2 didn’t really work in the chart above, but it was supposed to).

  5. Ohhhh. That makes a ton more sense. I would almost present all three as x100, just so they’re consistent (since we understand percentages as a number between 1 and 100, anyway).

Leave a Reply

Your email address will not be published. Required fields are marked *