View Single Post
Old
03-27-2013, 04:50 PM
  #40
Cashville
RIP Lindback
 
Cashville's Avatar
 
Join Date: Apr 2011
Location: Washington, DC
Country: United States
Posts: 5,530
vCash: 500
Quote:
Originally Posted by Taco MacArthur View Post
I posted this list in a History forum thread the other day - these show how much better a goaltender did, relative to the leaguewide save percentage in a postseason. Before anyone responds to that point, rest assured that I understand that it's not all about save percentage. It's a starting point for discussion.

Some notable ones better (actually, the top ten in terms of standard deviations above the mean for a particular playoff season):

Johnny Bower, 1963 Toronto (94.9% save percentage, 8-2, 3.2 standard deviations)
Patrick Roy, 1993 Montreal (92.9% save percentage, 16-4, 3.1 SD)
Martin Brodeur, 1995 New Jersey (93.2% save percentage, 16-4, 3.1 SD)
Curtis Joseph, 1993 St. Louis (93.8% save percentage, 7-4, 3.1 SD)
Richard Brodeur, 1982 Vancouver (91.7% save percentage, 11-6, 2.9 SD)
Olaf Kolzig, 1998 Washington (94.1% save percentage, 12-9, 2.9 SD)
Tim Thomas, 2011 Boston (94.0% save percentage, 16-9, 2.9 SD)
J-S Giguere, 2003 Anaheim (94.5% save percentage, 15-6, 2.9 SD)
Bernie Parent, 1976 Philadelphia (96.3% save percentage, 2-3, 2.7 SD)
Rogie Vachon, 1969 Montreal (95.3% save percentage, 7-1, 2.7 SD)

It's harder to get a high standard deviation metric playing in few games (stated differently, it's easier for an average goaltender to put up a seemingly-remarkable performance due to random chance in a smaller number of games).

In terms of goals prevented beyond a replacement-level goaltender, the list is quite similar (but includes my all-time favorite):

Tim Thomas, 2011 Boston (37.1 goals prevented above replacement)
Patrick Roy, 1993 Montreal (34.0)
Olaf Kolzig, 1998 Washington (33.5)
J-S Giguere, 2003 Anaheim (32.2)
John Vanbiesbrouck, 1996 Florida (32.0)
Richard Brodeur, 1982 Vancouver (31.6)
Kirk McLean, 1994 Vancouver (30.8)
Martin Brodeur, 1995 New Jersey (28.9)
Bill Ranford, 1990 Edmonton (27.5)
Curtis Joseph, 1993 St. Louis (26.8)

It's even harder for a small sample performance to end up on this list, since it's weighted by time played (and that's probably appropriate, although it punishes goaltenders from the early era - on the other hand, it was easier to win in those days since you only had to win eight).

These only go back to the 1953 playoffs.

As for Quick, Quick's 2012 playoffs was in an era of inflated save percentages overall:

http://hockeygoalies.org/bio/quick.html

His 94.6% save percentage in 2012 equates to 2.3 standard deviations above average, and 23.0 goals better than replacement.

Still laudable, of course.
Great post; good work. Tough to argue with the analysis here as it views every goalie's stats on a relative basis to their peers at the time thus isolating dead puck era and other similar influences on the raw stats.

You can still argue that the team in front influenced the respective goalie's stats, but there's only so much you can do with the numbers, and this does a pretty good job.

Cashville is offline   Reply With Quote