Round 2, Vote 1 (HOH Top Goaltenders)

[TheDevilMadeMe]

Please take a look at the adjusted save percentage thread - I've posted the all-time leaders using the revised methodology (adjusting to the regular season, rather than playoff average). The differences are, for the most part, quite small. It primarily benefits goalies from the 1980s when there was often a large difference between RS and PO save percentage - interested in if anybody has theories why.

I think TCG has a good theory. In the 1980s, 16/21 teams made the playoffs, so the good teams just played to have fun and didn't really care about defense until the playoffs.

Yes, that's a fair point. The Sabres' offense was notoriously streaky (the fact that Buffalo averaged 2.81 goals per game in the 1999 postseason means little given that they scored just 6 goals over the last 5 games of that postseason, including the infamous triple-OT game) but they performed better than expected in the playoffs.

Right. My point is that the commonly used refrain "Brodeur didn't do anything in the playoffs without help from his team" is just as true when talking about Hasek. Brodeur tended to get help from his defense. When Hasek advanced, his teams always played well offensively - uncharacteristically so for the Sabres, as expected for the Red Wings.

If two goalies (from the same era) stop the puck at virtually the same rate, but one of them wins more games, that tells me that he's either very clutch (having a bad performance when the game's outcome is essentially decided and playing well in critical situations) or he plays on a much stronger team, or he's very lucky. How much is dependent on the goalie, the team and luck is up for discussion.

True

Also - in case my first post wasn't clear, I wasn't suggesting that Hall was better than (or equal to) Sawchuk in the postseason. I agree that Sawchuk was superior in that regard. I was simply noting that, given the vast difference in their reputations, their effectiveness at stopping the puck was remarkably similar. Does anybody have any theories as to why?

Sawchuk played a pretty large sample of playoff games after his 1951-1955 peak, and while I think he was better in the playoffs than in the regular season after 1955, he still wasn't as good as he was before. A large sample of less impressive games would drag his averages down. Also, even during his peak, he had the one really awful playoffs that would drag his averages down.

 

[Dennis Bonvie]

Martin Brodeur's awards recognition in a league without Europeans

Brodeur is the goalie affected most by the European influence on awards. Here's a closer look at him season by season to figure out how many of those 9-10 hypothetical 1st Team All Star nods he's actually likely to get.

1997
VEZINA: Dominik Hasek 120 (22-3-1); Martin Brodeur 73 (3-18-4); Patrick Roy 25 (1-3-11); Guy Hebert 10 (0-2-4); Curtis Joseph 2 (0-0-2);
ALL-STAR: Dominik Hasek 237 (40-12-1); Martin Brodeur 174 (13-35-4); Patrick Roy 47 (0-5-32); John Vanbiesbrouck 4 (0-1-1); Jeff Hackett 4 (0-0-4);

1998
VEZINA: Dominik Hasek 126 (24-2-0); Martin Brodeur 57 (2-14-5); Tom Barrasso 26 (0-7-5); Ed Belfour 10 (0-1-7); Patrick Roy 5 (0-1-2);
ALL-STAR: Dominik Hasek 259 (50-3-0); Martin Brodeur 159 (3-47-3); Ed Belfour 40 (0-2-34); Tom Barrasso 15 (0-1-12); Olie Kolzig 3 (0-0-3)

1997 and 1998 were Hasek first, a gap to Brodeur in second, then a gap to the field. Brodeur wins the 1997 and 1998 Vezina and 1st Team nods if it weren't for Hasek's two Hart-winning seasons

2000
VEZINA: Olaf Kolzig 110 (14-13-1); Roman Turek 79 (9-9-7); Curtis Joseph 23 (2-3-4); Ed Belfour 14 (2-0-4); Martin Brodeur 8 (0-1-5);
ALL-STAR: Olaf Kolzig 197 (29-14-10); Roman Turek 186 (23-21-8); Martin Brodeur 48 (2-9-11); Ed Belfour 42 (1-7-16); Curtis Joseph 32 (2-5-7);

2000 was Kolzig, then Turek, then the Joseph/Belfour/Brodeur trio. It's uncertain which of those three finished first without the Europeans.

2001
VEZINA: Dominik Hasek, BUF 85 (9-12-4); Roman Cechmanek, PHI 65 (7-9-3); Martin Brodeur, N.J. 42 (7-2-1); Evgeni Nabokov, S.J. 29 (3-3-5); Patrick Roy, COL 19 (2-1-6);
GOALTENDER: Dominik Hasek, BUF 160 (19-18-11); Roman Cechmanek, PHI 154 (19-17-8); Martin Brodeur, N.J. 143 (18-15-8); Patrick Roy, COL 46 (3-7-10); Sean Burke, PHX 20 (2-1-7)

There is a fairly big gap between Brodeur and the 2nd best Canadian (Roy) in both Vezina and All-Star voting in 2001. It's more likely than not that he wins the 2001 Vezina and 1st Team All-Star in a European-only league.

Brodeur probably (but not definitely) picks up the 2001 Vezina/1st Team nod if it weren't for Hasek and Cechmanek. It's uncertain who wins the 2000 Vezina/1st Team behind Hasek and Turek.

2003&2004
Brodeur actually won the Vezina/1st Team combo

2006
VEZINA: Miikka Kiprusoff, CGY 140 (25-5-0); 2. Martin Brodeur, NJD 48 (2-10-8); 3. Henrik Lundqvist, NYR 41 (2-9-4); 4. Tomas Vokoun, NAS 15 (1-1-7); 5. Manny Legace, DET 6 (0-1-3)
All-Star: Kiprusoff - 599 (113-11-1); Brodeur - 232 (6-59-25); Lundqvist - 128 (5-27-22); Turco - 47 (1-7-21); Vokoun - 44 (2-5-19)

2006 is tougher than it looks, because Kiprusoff was much better than any other goalie, and Brodeur probably got a lot of reputation second place votes since everyone knew it was "Kiprusoff then the field." Still, the guys close to him in "the field" are almost all Europeans, especially in Vezina voting. Brodeur wins the 2006 Vezina without Europeans, and has a good shot at 1st Team All Star

2007
Brodeur won both the Vezina & 1st Team All Star nod

2008
VEZINA: 1. Martin Brodeur, N.J. 113 {15-12-2}; 2. Evgeni Nabokov, S.J. 106 {13-13-2}; 3. Henrik Lundqvist. NYR 13 {1-0-8); 4. J.S. Giguere, ANA 11 {0-1-8); 5. Miikka Kiprusoff, CGY 7 {1-0-2)

ALL-STAR: EVGENI NABOKOV, S.J. 523 (71-55-3); Martin Brodeur, N.J. 475 (56-61-12); J.S. Giguere, ANA 82 (3-8-43); Henrik Lundqvist, NYR 74 (1-6-51); Roberto Luongo, VAN 12 (1-1-4)

2008 was Brodeur and Nabokov, the the field. Brodeur picks up the 2008 1st Team nod to go with his Vezina without Nabokov

2010
VEZINA: Miller Ryan BUF 126 (23, 3, 2); Bryzgalov Ilya PHO 79 (5, 16, 6), Brodeur Martin 32 (1, 6, 9); Nabokov Evgeni 9 (1, 1, 1); Anderson Craig COL 9 (0, 2, 3)

ALL-STAR: Miller Ryan BUF 598 (105 24 1); Bryzgalov Ilya PHO 371 (20, 85, 16); Brodeur Martin NJ 136 (6, 16, 58); Nabokov Evgeni SJ 13 (0, 2, 7); Kiprusoff Miikka CGY 12 (0, 0, 12)

2010 was clearly a case of Ryan Miller, then Bryzgalov, then Brodeur. Brodeur was the only Canadian Top 5 in voting for either category. Brodeur likely is a 2nd Team All Star without Bryzgalov, and a 1st Teamer without any Americans

Conclusions

• Brodeur won 4 Vezinas and 3 First Team All Stars.
• He was clearly #2 behind Hasek in 1997 and 1998. Without Hasek's Hart-winning seasons, Brodeur wins 6 Vezinas and 5 First Team All Stars
• Brodeur is almost certain to win the 2006 Vezina and 2008 First Team without any Europeans. In a Euro-less league, Brodeur wins at least 7 Vezinas and 6 First Team All Stars.
• Brodeur probably wins both the 2001 Vezina and First Team and the 2006 First Team without any Europeans. In a Euro-less league, Brodeur probably wins 8 Vezinas and 8 First Team All Stars.
• Brodeur might win both the 2000 Vezina and First Team without any Europeans, but Joseph and Belfour had about the same chance. In a Euro-less league, Brodeur has a chance at 9 Vezinas and 9 First Team All Stars.
• Brodeur is probably a 2nd Team All Star in 2010 without any Europeans. Without any Americans either, he is probably the Vezina winner and First Team All Star. In a Euro-less league, Brodeur adds a 2nd Team All Star in 2010 to his 6-9 First Teams. In a league with no Euros or Americans, adds another Vezina, giving him at most 10 Vezinas and 10 First Team All Stars.
• Brodeur was also the leading Canadian in Hart voting by fairly wide margins in both 1998 and 2003, but it's always possible that a Canadian forward who finished well below him wins the award if you just remove all the European forwards, given the preference for forwards.

Conclusion: If you remove all Europeans, Brodeur likely wins 7-9 Vezinas and 6-9 First Teams. Remove all Americans and he likely wins 8-10 Vezinas and 7-10 First Teams. He also has a shot at both the 1998 and 2003 Hart trophies.

Standard disclaimer: This entire study is about who would have won, and is making no judgments about who should have won.

Having seen tons of games played by Brodeur, Hasek and Roy I didn't realize how much better Brodeur really was if I only considered throwing out Euros & Americans.

This one might top the "they don't know how to count in New Jersey" argument.

 

[Canadiens1958]

Other Factors

:
If two goalies (from the same era) stop the puck at virtually the same rate, but one of them wins more games, that tells me that he's either very clutch (having a bad performance when the game's outcome is essentially decided and playing well in critical situations) or he plays on a much stronger team, or he's very lucky. How much is dependent on the goalie, the team and luck is up for discussion.

Also - in case my first post wasn't clear, I wasn't suggesting that Hall was better than (or equal to) Sawchuk in the postseason. I agree that Sawchuk was superior in that regard. I was simply noting that, given the vast difference in their reputations, their effectiveness at stopping the puck was remarkably similar. Does anybody have any theories as to why?

SV% is global.

Glenn Hall enjoyed a career relatively free of major injuries. Sawchuk did not. Pre and post injury performance is an important consideration. Sawchuk drops from .917 to .900 he is still better than Bassen, Riggin or a young Crozier.

Matching them when they are both healthy produces different results than global numbers tend to predict.

 

[Hockey Outsider]

Not sure if this is off-topic but worth a shot.

From what I remember, the Conn Smythe was clearly a two-horse race between Lidstrom and Yzerman. This Sports Illustrated article (written June 12, the day before the Red Wings won the Cup) suggests that the field is wide open and that any of the aforementioned, Hull, Hasek or Fedorov could win. Was this the errant opinion of one author, or was the 2002 Conn Smythe really considered wide open?

http://sportsillustrated.cnn.com/hockey/2002/playoffs/news/2002/06/12/offday_notebook_ap/

 

[Ohashi_Jouzu*]

Not sure if this is off-topic but worth a shot.

From what I remember, the Conn Smythe was clearly a two-horse race between Lidstrom and Yzerman. This Sports Illustrated article (written June 12, the day before the Red Wings won the Cup) suggests that the field is wide open and that any of the aforementioned, Hull, Hasek or Fedorov could win. Was this the errant opinion of one author, or was the 2002 Conn Smythe really considered wide open?

http://sportsillustrated.cnn.com/hockey/2002/playoffs/news/2002/06/12/offday_notebook_ap/

Quick search brought me to a Red Wings fan discussion forum in 2007 discussing the 2002 playoffs. General consensus is that Lidstrom, Yzerman, Hasek, and Fedorov were all suitable candidates, with most favouring Yzerman or Lidstrom.

 

[TheDevilMadeMe]

Having seen tons of games played by Brodeur, Hasek and Roy I didn't realize how much better Brodeur really was if I only considered throwing out Euros & Americans.

The point is that awards only show how good a player is when you consider the competition for the awards. No Europeans competed for awards in the Original 6 era, so I find it useful to estimate what the leaderboards might look like with no Europeans.

Edit: The point wasn't to compare Brodeur with directly Hasek and Roy by the way. My point is that you can't just compare All Star Teams from the Original 6 era with All Star teams from the modern era and assume that they are as easy to come by.

This one might top the "they don't know how to count in New Jersey" argument.

Considering the fact that the undercounting in New Jersey has been proven mathematically, I'll take that as a complement. But I don't think the effects of adding competition from Europe are quite as cut and dry.

 

[Canadiens1958]

Sawchuk Injuries

I went to confirm this, and just noticed that on my Sawchuk page, the "injuries" section is blank.

Looks like I've got some things to add once the database work is completed.

Reference to the 1962-63 tendon injury and others suffered by Terry Sawchuk below:

http://www.hhof.com/htmlSpotlight/spot_oneononep197103.shtml

 

[TheDevilMadeMe]

Not sure if this is off-topic but worth a shot.

From what I remember, the Conn Smythe was clearly a two-horse race between Lidstrom and Yzerman. This Sports Illustrated article (written June 12, the day before the Red Wings won the Cup) suggests that the field is wide open and that any of the aforementioned, Hull, Hasek or Fedorov could win. Was this the errant opinion of one author, or was the 2002 Conn Smythe really considered wide open?

http://sportsillustrated.cnn.com/hockey/2002/playoffs/news/2002/06/12/offday_notebook_ap/

I put as much stock into that as I do the articles from the 2012 finals that listed any of 5 Kings who could win. The media loves a discussion. To the extent Hasek was a candidate, it was because he set the new shutout record (broken by Brodeur in 2003). Hull and Fedorov had some nice stats, but were behind Yzerman by pretty much any measure.

Lidstrom and Yzerman were the only serious candidates IMO. Those other three played very well, but didn't have much of a chance IMO.

 

[TheDevilMadeMe]

If you want to continue discussing Hasek vs Roy in the 2002 Western Conference finals, please do so here:

http://hfboards.mandatory.com/showthread.php?t=1272027

I doubt it will be the last spinoff thread that this project inspires.

 

[JackSlater]

I put as much stock into that as I do the articles from the 2012 finals that listed any of 5 Kings who could win. The media loves a discussion. To the extent Hasek was a candidate, it was because he set the new shutout record (broken by Brodeur in 2003). Hull and Fedorov had some nice stats, but were behind Yzerman by pretty much any measure.

Lidstrom and Yzerman were the only serious candidates IMO. Those other three played very well, but didn't have much of a chance IMO.

I believe CBC had a segment during the finals that basically said as much. Looking at the Conn Smythe candidates from Detroit they favoured Yzerman, then Lidstrom. Hasek received credit for the shutouts but his inconsistency did him in.

 

[TheDevilMadeMe]

Here's a look at Terry Sawchuk, Jacques Plante, and Glenn Hall from the pages of Sports Illustrated over the years. It includes a lot of subjective detail that the statistical record doesn't show - filling out the bones of the statistical record, as it were.

I don't plan to look at the four more modern goaltenders in the same way, because many of us have experienced their careers as they happened and because there's a lot more material to dig through.

...

Excellent job. I'm hoping more people comment on this.

One thing I take away is how often the 60s Chicago Blackhawks are described as a "run and gun" team. Relevant to this project, it really helps explain how Hall was the one who broke the mold of "1st Team All Star = lowest GAA."

Relevant to future projects, it really helps put the offensive stats of Bobby Hull and Stan Mikita into perspective.

 

[Dennis Bonvie]

The point is that awards only show how good a player is when you consider the competition for the awards. No Europeans competed for awards in the Original 6 era, so I find it useful to estimate what the leaderboards might look like with no Europeans.

Edit: The point wasn't to compare Brodeur with directly Hasek and Roy by the way. My point is that you can't just compare All Star Teams from the Original 6 era with All Star teams from the modern era and assume that they are as easy to come by.



Considering the fact that the undercounting in New Jersey has been proven mathematically, I'll take that as a complement. But I don't think the effects of adding competition from Europe are quite as cut and dry.

Sorry, I don't buy it.

 

[Bear of Bad News]

Sorry, I don't buy it.

It's not a question of whether you buy it or not. It's been exhibited statistically.

You're free to not buy it.

 

[Theokritos]

2000
VEZINA: Olaf Kolzig 110 (14-13-1); Roman Turek 79 (9-9-7); Curtis Joseph 23 (2-3-4); Ed Belfour 14 (2-0-4); Martin Brodeur 8 (0-1-5);
ALL-STAR: Olaf Kolzig 197 (29-14-10); Roman Turek 186 (23-21-8); Martin Brodeur 48 (2-9-11); Ed Belfour 42 (1-7-16); Curtis Joseph 32 (2-5-7);

2000 was Kolzig, then Turek, then the Joseph/Belfour/Brodeur trio. It's uncertain which of those three finished first without the Europeans.

Side note: Kölzig was born in South Africa to German parents, but he moved to Canada when he was three years old and grew up there and learned to play hockey there. Yes, he decided to play for Germany in 1996, but for the purpose of this project he shouldn't be considered a European, his case is similiar to Stan Mikita for example.

 

[ContrarianGoaltender]

Sorry, I don't buy it.

I understand the skepticism because I needed evidence to convince me of it as well, but the difference in home vs. road shooting percentage, save percentage and total shots counted do in fact make a compelling argument, although the exact magnitude may still not yet be certain.

Brodeur's save percentages are higher on the road, which is what the undercounting theory predicts, particularly during the seasons where he won his Vezinas. You mentioned before his low save percentage rankings in those seasons, which is correct, but the counterargument would be that his official scorer cost him some saves and he also had a slight disadvantage for playing relatively more on the road. The reality is that if you look at his performance on the road in those seasons he does rank quite well relative to the competition.

I'm going to bring up another factor as well, shot prevention. If you compare Brodeur to the backup goalies who have played with him for longer periods of time and played more minutes per game, he typically faced fewer shots against than they did. My estimate of the effect is one shot per 60 minutes against, based on comparing their New Jersey numbers relative to Brodeur with their shots against relative to other goalies on other teams. Overpass also once did an adjustment of New Jersey backups based on their strength of opposition that got a similar number.

The adjustment is therefore to add one extra save and one extra shot against to Brodeur's numbers per 60 minutes of play. The logic is that if Brodeur prevented a shot from happening that another goalie would not have, whether it was through puckhandling or by freezing the puck less often for defensive zone faceoffs (the two most likely factors, IMO), then he should be credited for that in his stats.

For save percentages in the .925-.930 range, crediting one extra save and shot per 60 minutes adds .003 to a goalie's save percentage, so if you want to see what Brodeur's numbers look like without that adjustment then just subtract .003 from his totals below.

Road Save Percentage Leaders, 2002-03:
1. Marty Turco: 730 SA, .937
2. Roberto Luongo: 1084 SA, .928
3. Olaf Kolzig: 874 SA, .928
4. Ed Belfour: 828 SA, .926
5. Dwayne Roloson: 732 SA, .926
6. Martin Brodeur: 933 SA, .925

Road Save Percentage Leaders, 2003-04:
1. Dwayne Roloson: 609 SA, .939
2. Roberto Luongo: 1129 SA, .935
3. Andrew Raycroft: 716 SA, .930
4. Martin Brodeur: 1025 SA, .930
5. Miikka Kiprusoff: 424 SA, .927

Road Save Percentage Leaders, 2006-07:
1. Martin Brodeur: 1222 SA, .929
2. Roberto Luongo: 1032 SA, .927
3. Tomas Vokoun: 592 SA, .926
4. Ryan Miller: 862 SA, .922
5. Marty Turco: 856 SA, .919

Road Save Percentage Leaders, 2007-08:
1. Dan Ellis: 566 SA, .929
2. Martin Brodeur: 1059 SA, .924
3. J.S. Giguere: 804 SA, .924
4. Pascal Leclaire: 627 SA, .920
5. Carey Price: 689 SA, .920

Brodeur's finishes in save percentage go from 3, 8, 13, 14 in overall save percentage to 1, 2, 4, 6 on the road only, which is a massive improvement. I didn't run the numbers for all goalies, just ones with a lot of games played, so there might be a few in the 20-30 game range who would have snuck in there ahead of him in road save % for a few of those seasons but they wouldn't exactly be Vezina threats anyway.

Take those results, emphasize Brodeur's durability, and that's probably the most pro-Brodeur case that can be made for those four seasons.

I'm not saying he necessarily deserved those Vezinas (I would have gone Turco, Luongo, Luongo, Brodeur, although the 2007 one was pretty much a toss up), just pointing out that he very likely did have more of a legitimate case than the overall save percentages indicate. He also doesn't exactly dominate the field like Roy or Hasek did, although as TDMM points out the depth of the talent pool has been rapidly increasing over the last two decades with the influx of European goalies which is an important consideration.

It should be noted that this does not account for shot quality or special teams factors, which were generally to Brodeur's advantage. In '03, '04 and '07 New Jersey faced 28% fewer PP against than the average team, and under Pat Burns in '03 and '04 Brodeur's backups had a .944 save percentage on 375 shots, an extreme number which suggests possible shot quality effects even though it is a very small sample size (and they did play against weak opposition). I also think the ideal method is to adjust the home sample for overcounting rather than remove it completely. However, this is at least one pretty good argument in support of Brodeur.

 

[GuineaPig]

I put as much stock into that as I do the articles from the 2012 finals that listed any of 5 Kings who could win. The media loves a discussion. To the extent Hasek was a candidate, it was because he set the new shutout record (broken by Brodeur in 2003). Hull and Fedorov had some nice stats, but were behind Yzerman by pretty much any measure.

Lidstrom and Yzerman were the only serious candidates IMO. Those other three played very well, but didn't have much of a chance IMO.

I think another part of the Smythe discussion re: Hasek was that his best play definitely came against Carolina. When people started talking about Smythe candidates (seeing as it was pretty much assumed that the Wings would demolish the 'canes) Hasek hadn't played all that great up to that point (0.914 save percentage). And then once he did start playing like the Dominator of old, the series was so one-sided anyways that he didn't stand out.

 

[Canadiens1958]

1961 to 1963 Playoffs - Jacques Plante

Post five consecutive SC championship seasons. Jacques Plante's las three with the Canadiens. Note mid season Jacques Plante suffered a knee injuryNov 26 - Feb 9) that did not require surgery but caused him to miss 30 games including an eight game rehab stint in the EPHL with the Montreal Royals. Data culled from the HSP project and reference to BM67 post #123 this thread.

1961 Semi Final vs CHICAGO

Game 1 - Home, MTL 6 CHI 2,(25/27) (12/9/6), First goal 11:04 (1 - 1).
Game 2 - Home, MTL 3 CHI 4, (19/23) (9/6/8), First goal 29:02 (0 - 1)
Game 3 - Away, MTL 1 CHI 2 3OT, (43/45), (7/6/7/8/13/4), First goal 38:33 (0 - 1).
Game 4 - Away, MTL 5 CHI 2, (19/21) (4/11/6) First goal 4:24 (1 - 1).
Game 5 - Home, MTL 0 CHI 3, (32/35) (14/7/14) First goal 28:34 (0 - 1).
Game 6 - Away, MTL 0 CHI 3, (24/27) (7/8/12) First goal 21:15 (0-1).

Jacques Plante SV%

1st period .962 51/53
2nd period .809 38/47
3rd period .943 50/53
Overtime .969 24/25
Away .925 86/93
Home .894 76/85
Series .911 163/179

Notes,5PPGs - (0/4/0/0/0/1)

1962 Semi Final vs CHICAGO

Game 1 - Home, MTL 2 CHI 1,(36/37) (17/9/11), First goal 58:06 (2 - 1).
Game 2 - Home, MTL 4 CHI 3, (26/29) (10/10/9), First goal 5:26 (1 - 1)
Game 3 - Away, MTL 1 CHI 4, (23/27), (10/9/8), First goal 9:05 (0 - 1).
Game 4 - Away, MTL 3 CHI 5, (37/42) (22/7/13) First goal 1:51 (0 - 1).
Game 5 - Home, MTL 3 CHI 4, (25/29) (8/9/12) First goal 2:55 (0 - 1).
Game 6 - Away, MTL 0 CHI 2, (29/31) (13/11/7) First goal 4:02 (0-1).

Jacques Plante SV%

1st period .900 72/80
2nd period .927 51/55
3rd period .883 53/60
Away .890 89/100
Home .916 87/95
Series .903 176/195

Notes,5PPGs - (2/1/2), 1 SHG - 3rd period.

1963 Semi Final vs TORONTO

Game 1 - Away, MTL 1 TOR 3,(31/34) (10/12/12), First goal 3:30 (0 - 1).
Game 2 - Away, MTL 2 TOR 3, (24/27) (10/8/9), First goal 6:54 (1 - 1)
Game 3 - Home, MTL 0 TOR 2, (20/22), (7/7/8) First goal 33:14 (0-1).
Game 4 - Home, MTL 3 TOR 1, (23/24) (7/9/8) First goal 6:12 (0 - 1).
Game 5 - Away, MTL 0 TOR 5, (28/33) (12/11/10) First goal 24:37 (0 - 1).


Jacques Plante SV%

1st period .869 40/46
2nd period .872 41/47
3rd period .957 45/47
Away .883 83/94
Home .935 43/46
Series .900 126/140

Notes. 1 PPG - 2nd period

 

[BM67]

A 13 year record of goals, shots, and shooting% for both teams in Devils' home and road games.

Year|H G|H SOG|R G|R SOG|H S%|R S%
1998-99|225|2256|219|2335|9.97%|9.38%
1999-00|225|2368|229|2450|9.50%|9.35%
2000-01|237|2244|253|2370|10.56%|10.68%
2001-02|200|2023|192|2322|9.89%|8.27%
2002-03|206|2205|176|2329|9.34%|7.56%
2003-04|194|2101|183|2333|9.23%|7.84%
2005-06|225|2278|233|2520|9.88%|9.25%
2006-07|194|2196|205|2490|8.83%|8.23%
2007-08|197|2268|194|2351|8.69%|8.25%
2008-09|231|2476|214|2637|9.33%|8.12%
2009-10|210|2205|192|2467|9.52%|7.78%
2010-11|192|2247|186|2281|8.54%|8.15%
2011-12|213|2208|208|2244|9.65%|9.27%
Total|2749|29075|2684|31129|9.45%|8.62%
T/GP|5.16|54.55|5.04|58.40|9.45%|8.62%

0.12 more goals in home games, 3.85 more shots in road games, and S% almost 10% higher in home games. More goals at home in 9 of 13 seasons, but not once more shots at home. The only season with a higher S% on the road is a complete outlier powered by the Devils' 150 road GF at a S% of 11.52%.

Let's look at this for Brodeur's Vezina winning seasons.

Total shots both teams home and road in Devils' games.

Year|HG SOG|RG SOG|DIF/GP
2002-03|2205|2329|3.02
2003-04|2101|2333|5.66
2006-07|2196|2490|7.17
2007-08|2268|2351|2.02

In 06-07, there are 11.8% more shots in Devils' road games, but only 5.4% more goals.

2006-07 looks even worse in Brodeur's H/R splits. He faced 25.47 SOG/60 at home and 30.30 SOG/60 on the road. That's a difference of 4.83 or 18.96%. The league average for 06-07 was 2.02 or 7.16% more SOG/60 on the road.

 

[Canadiens1958]

Away vs Home

I understand the skepticism because I needed evidence to convince me of it as well, but the difference in home vs. road shooting percentage, save percentage and total shots counted do in fact make a compelling argument, although the exact magnitude may still not yet be certain.

Brodeur's save percentages are higher on the road, which is what the undercounting theory predicts, particularly during the seasons where he won his Vezinas. You mentioned before his low save percentage rankings in those seasons, which is correct, but the counterargument would be that his official scorer cost him some saves and he also had a slight disadvantage for playing relatively more on the road. The reality is that if you look at his performance on the road in those seasons he does rank quite well relative to the competition.

I'm going to bring up another factor as well, shot prevention. If you compare Brodeur to the backup goalies who have played with him for longer periods of time and played more minutes per game, he typically faced fewer shots against than they did. My estimate of the effect is one shot per 60 minutes against, based on comparing their New Jersey numbers relative to Brodeur with their shots against relative to other goalies on other teams. Overpass also once did an adjustment of New Jersey backups based on their strength of opposition that got a similar number.

The adjustment is therefore to add one extra save and one extra shot against to Brodeur's numbers per 60 minutes of play. The logic is that if Brodeur prevented a shot from happening that another goalie would not have, whether it was through puckhandling or by freezing the puck less often for defensive zone faceoffs (the two most likely factors, IMO), then he should be credited for that in his stats.

For save percentages in the .925-.930 range, crediting one extra save and shot per 60 minutes adds .003 to a goalie's save percentage, so if you want to see what Brodeur's numbers look like without that adjustment then just subtract .003 from his totals below.

Road Save Percentage Leaders, 2002-03:
1. Marty Turco: 730 SA, .937
2. Roberto Luongo: 1084 SA, .928
3. Olaf Kolzig: 874 SA, .928
4. Ed Belfour: 828 SA, .926
5. Dwayne Roloson: 732 SA, .926
6. Martin Brodeur: 933 SA, .925

Road Save Percentage Leaders, 2003-04:
1. Dwayne Roloson: 609 SA, .939
2. Roberto Luongo: 1129 SA, .935
3. Andrew Raycroft: 716 SA, .930
4. Martin Brodeur: 1025 SA, .930
5. Miikka Kiprusoff: 424 SA, .927

Road Save Percentage Leaders, 2006-07:
1. Martin Brodeur: 1222 SA, .929
2. Roberto Luongo: 1032 SA, .927
3. Tomas Vokoun: 592 SA, .926
4. Ryan Miller: 862 SA, .922
5. Marty Turco: 856 SA, .919

Road Save Percentage Leaders, 2007-08:
1. Dan Ellis: 566 SA, .929
2. Martin Brodeur: 1059 SA, .924
3. J.S. Giguere: 804 SA, .924
4. Pascal Leclaire: 627 SA, .920
5. Carey Price: 689 SA, .920

Brodeur's finishes in save percentage go from 3, 8, 13, 14 in overall save percentage to 1, 2, 4, 6 on the road only, which is a massive improvement. I didn't run the numbers for all goalies, just ones with a lot of games played, so there might be a few in the 20-30 game range who would have snuck in there ahead of him in road save % for a few of those seasons but they wouldn't exactly be Vezina threats anyway.

Take those results, emphasize Brodeur's durability, and that's probably the most pro-Brodeur case that can be made for those four seasons.

I'm not saying he necessarily deserved those Vezinas (I would have gone Turco, Luongo, Luongo, Brodeur, although the 2007 one was pretty much a toss up), just pointing out that he very likely did have more of a legitimate case than the overall save percentages indicate. He also doesn't exactly dominate the field like Roy or Hasek did, although as TDMM points out the depth of the talent pool has been rapidly increasing over the last two decades with the influx of European goalies which is an important consideration.

It should be noted that this does not account for shot quality or special teams factors, which were generally to Brodeur's advantage. In '03, '04 and '07 New Jersey faced 28% fewer PP against than the average team, and under Pat Burns in '03 and '04 Brodeur's backups had a .944 save percentage on 375 shots, an extreme number which suggests possible shot quality effects even though it is a very small sample size (and they did play against weak opposition). I also think the ideal method is to adjust the home sample for overcounting rather than remove it completely. However, this is at least one pretty good argument in support of Brodeur.

Your data makes a very compelling argument favouring goalies that are strong(stronger) on the road. In the four seasons studied above, other than Robert Luongo no other goalie comes close to Martin Brodeur in terms of appearances on the road and shots against.

Comparing Brodeur's road numbers to his total numbers

2002-03 Total 1706 SA .914 SV% Away 933 SA .925 SV%
2003-04 Total 1845 SA .917 SV% Away 1025 SA .930 SV%
2006-07 Total 2182 SA .922 SV% Away 1222 SA .929 SV%
2007-08 Total 2089 SA .920 SV% Away 1059 SA .924 SV%

Even without the extrapolation to home numbers, to conclusions are very clear. During these four seasons Martin Brodeur was a significantly better goalie on the road than at home, he was used accordingly and countered the oppositions home advantages - last change, etc. The shot undercounting issue while founded does not impact this difference since Martin Brodeur produced a SV% > .925 cumalative over four seasons while playing a solid majority of the time on the road.

 

[Canadiens1958]

Road Performance

A 13 year record of goals, shots, and shooting% for both teams in Devils' home and road games.

Year|H G|H SOG|R G|R SOG|H S%|R S%
1998-99|225|2256|219|2335|9.97%|9.38%
1999-00|225|2368|229|2450|9.50%|9.35%
2000-01|237|2244|253|2370|10.56%|10.68%
2001-02|200|2023|192|2322|9.89%|8.27%
2002-03|206|2205|176|2329|9.34%|7.56%
2003-04|194|2101|183|2333|9.23%|7.84%
2005-06|225|2278|233|2520|9.88%|9.25%
2006-07|194|2196|205|2490|8.83%|8.23%
2007-08|197|2268|194|2351|8.69%|8.25%
2008-09|231|2476|214|2637|9.33%|8.12%
2009-10|210|2205|192|2467|9.52%|7.78%
2010-11|192|2247|186|2281|8.54%|8.15%
2011-12|213|2208|208|2244|9.65%|9.27%
Total|2749|29075|2684|31129|9.45%|8.62%
T/GP|5.16|54.55|5.04|58.40|9.45%|8.62%

0.12 more goals in home games, 3.85 more shots in road games, and S% almost 10% higher in home games. More goals at home in 9 of 13 seasons, but not once more shots at home. The only season with a higher S% on the road is a complete outlier powered by the Devils' 150 road GF at a S% of 11.52%.

Let's look at this for Brodeur's Vezina winning seasons.

Total shots both teams home and road in Devils' games.

Year|HG SOG|RG SOG|DIF/GP
2002-03|2205|2329|3.02
2003-04|2101|2333|5.66
2006-07|2196|2490|7.17
2007-08|2268|2351|2.02

In 06-07, there are 11.8% more shots in Devils' road games, but only 5.4% more goals.

2006-07 looks even worse in Brodeur's H/R splits. He faced 25.47 SOG/60 at home and 30.30 SOG/60 on the road. That's a difference of 4.83 or 18.96%. The league average for 06-07 was 2.02 or 7.16% more SOG/60 on the road.

Not in dispute at all. Fact remains that goalies that can sustain a SV% > .925 on the road while leading (top 3) in road starts/appearances are far from commonplace.

 

[seventieslord]

I understand the skepticism because I needed evidence to convince me of it as well, but the difference in home vs. road shooting percentage, save percentage and total shots counted do in fact make a compelling argument, although the exact magnitude may still not yet be certain.

Brodeur's save percentages are higher on the road, which is what the undercounting theory predicts, particularly during the seasons where he won his Vezinas. You mentioned before his low save percentage rankings in those seasons, which is correct, but the counterargument would be that his official scorer cost him some saves and he also had a slight disadvantage for playing relatively more on the road. The reality is that if you look at his performance on the road in those seasons he does rank quite well relative to the competition.

I'm going to bring up another factor as well, shot prevention. If you compare Brodeur to the backup goalies who have played with him for longer periods of time and played more minutes per game, he typically faced fewer shots against than they did. My estimate of the effect is one shot per 60 minutes against, based on comparing their New Jersey numbers relative to Brodeur with their shots against relative to other goalies on other teams. Overpass also once did an adjustment of New Jersey backups based on their strength of opposition that got a similar number.

The adjustment is therefore to add one extra save and one extra shot against to Brodeur's numbers per 60 minutes of play. The logic is that if Brodeur prevented a shot from happening that another goalie would not have, whether it was through puckhandling or by freezing the puck less often for defensive zone faceoffs (the two most likely factors, IMO), then he should be credited for that in his stats.

For save percentages in the .925-.930 range, crediting one extra save and shot per 60 minutes adds .003 to a goalie's save percentage, so if you want to see what Brodeur's numbers look like without that adjustment then just subtract .003 from his totals below.

Road Save Percentage Leaders, 2002-03:
1. Marty Turco: 730 SA, .937
2. Roberto Luongo: 1084 SA, .928
3. Olaf Kolzig: 874 SA, .928
4. Ed Belfour: 828 SA, .926
5. Dwayne Roloson: 732 SA, .926
6. Martin Brodeur: 933 SA, .925

Road Save Percentage Leaders, 2003-04:
1. Dwayne Roloson: 609 SA, .939
2. Roberto Luongo: 1129 SA, .935
3. Andrew Raycroft: 716 SA, .930
4. Martin Brodeur: 1025 SA, .930
5. Miikka Kiprusoff: 424 SA, .927

Road Save Percentage Leaders, 2006-07:
1. Martin Brodeur: 1222 SA, .929
2. Roberto Luongo: 1032 SA, .927
3. Tomas Vokoun: 592 SA, .926
4. Ryan Miller: 862 SA, .922
5. Marty Turco: 856 SA, .919

Road Save Percentage Leaders, 2007-08:
1. Dan Ellis: 566 SA, .929
2. Martin Brodeur: 1059 SA, .924
3. J.S. Giguere: 804 SA, .924
4. Pascal Leclaire: 627 SA, .920
5. Carey Price: 689 SA, .920

Brodeur's finishes in save percentage go from 3, 8, 13, 14 in overall save percentage to 1, 2, 4, 6 on the road only, which is a massive improvement. I didn't run the numbers for all goalies, just ones with a lot of games played, so there might be a few in the 20-30 game range who would have snuck in there ahead of him in road save % for a few of those seasons but they wouldn't exactly be Vezina threats anyway.

Take those results, emphasize Brodeur's durability, and that's probably the most pro-Brodeur case that can be made for those four seasons.

I'm not saying he necessarily deserved those Vezinas (I would have gone Turco, Luongo, Luongo, Brodeur, although the 2007 one was pretty much a toss up), just pointing out that he very likely did have more of a legitimate case than the overall save percentages indicate. He also doesn't exactly dominate the field like Roy or Hasek did, although as TDMM points out the depth of the talent pool has been rapidly increasing over the last two decades with the influx of European goalies which is an important consideration.

It should be noted that this does not account for shot quality or special teams factors, which were generally to Brodeur's advantage. In '03, '04 and '07 New Jersey faced 28% fewer PP against than the average team, and under Pat Burns in '03 and '04 Brodeur's backups had a .944 save percentage on 375 shots, an extreme number which suggests possible shot quality effects even though it is a very small sample size (and they did play against weak opposition). I also think the ideal method is to adjust the home sample for overcounting rather than remove it completely. However, this is at least one pretty good argument in support of Brodeur.

So.... Brodeur Is Not a Fraud?

I feel like my whole world is crumbling down!

Even without the extrapolation to home numbers, to conclusions are very clear. During these four seasons Martin Brodeur was a significantly better goalie on the road than at home,

I don't think that's what it says. I think it says that he had shots against at home that weren't being counted.

 

[Ohashi_Jouzu*]

I don't think that's what it says. I think it says that he had shots against at home that weren't being counted.

I think it's quite possibly a little bit of both.

 

[BM67]

Not in dispute at all. Fact remains that goalies that can sustain a SV% > .925 on the road while leading (top 3) in road starts/appearances are far from commonplace.

I'm not worried about convincing people that don't want to be convinced, but rather coming up with a reasonable hand drawn facsimile of just how big a factor it really is.

I see that 7.17 shots per game for both teams and it looks huge, but it pales in comparison to the SA only numbers of Brodeur's H/R split.

Running the numbers for both teams and Brodeur make it look like over 2/3 of the "missing" shots are on Brodeur's net. Are most of the "missing" shots happening at the end of the ice where Brodeur plays two periods of each game? Are the opponents twice as likely to try to pick the corner on Brodeur as the Devils are on their goalie?

 

[pluppe]

I am not only responding to your words but to a specific philosophy articulated in your posts, MasterofDistrict's posts, and the article on consistency that MasterofDistricts linked to.

So why are you responding to me? I have time and time again said that we should be careful about making to sure deductions because random chance plays a role.

To attribute all variation in past performance to random chance when judging past performance is to give goaltenders all the agency of a hockey-playing robot of constant talent level. As the linked article says:

The implication is that Bryzgalov is not responsible for any deviations from that 0.915 save percentage - no matter in what situation those deviations occurred. He has no possibility of changing his level of play.

And then you go on quoting a article I never linked too. And saying that I "attribute all variation in past performance to random chance". And try to claim that it´s not a response to my words but to a philosophy. Like I said, please reread all my posts and then tell me that I am not saying that it is a combination of thing where chance plays a part.

When you post things like this:



and suggest that clutch does not exist, what is the purpose of that post for this debate? Is it not to remove all responsibility from players for their performance in specific, important situations or "sample sizes", because they have zero ability to play any better or worse than their constant ability level? For that is the inevitable outcome of your philosophy.

Thanks for leaving out the following line when quoting "I´m not saying it is true in this case but it´s worth mentioning." I think it is worth mentioning and you can see why in my answer to qpq below. And again: Where is this my philosophy?

I doubt very much that goaltenders themselves experience their career in this way. If they perform poorly in a big spot, you'd better believe they're going to regret that after their career is over. They're not just going to shrug and say "I randomly happened to have a bad game at a bad time. Not my fault."

Of course they won´t. But that does not make it not true. Goalies like all people are affected by attributional biases. So? Or are you claiming that there is no such thing? Is the project not about discussing these things with as much context as possible. If we can´t mention that chance plays a role (which I remind you is all I have done) then I think you are missing out on important information.

(I also think that if we did evaluate them based on the goaltenders own perception, Roy would be the runaway winner)

I'm not sure how Hasek's consistency when he's in an equal or advantageous position in a series is a good counter for an argument that he struggles when faced with adversity (competition at his position, playing with injuries), but sure. Subtracting the trailing numbers from the total numbers gives us this:

Roy (Leading/Tied): 134-67; 66.7%
Roy (Trailing): 35-10; 77.7%
Roy (Difference): +10%

Hasek (Leading/Tied): 64-26; 71.1%
Hasek (Trailing): 15-13; 53.6%
Hasek (Difference): -18%


So, as you can see, Hasek's consistency falls significantly when his team is in a bad position, whereas Roy's consistency when down in a series offers his team the best chance to get back into the series.

I think you are very biased here. Your way is one way of looking at it but couldn´t somebody claim something like "Winning the first game or a game 7 is the most important thing in the playoffs so performing when tying is the best quality in a goaltender. That Hasek performed worse when trailing probably has to do with that Buffalo as a team realized they were not good enough to overcome the situation, gave up and did not support their goalie."

And if you believe that Roy actually lifted his performance +10% when trailing you have to answer why he did not do this in the other playoff games? Do you really want a goalie who lacks the motivation to give it his best when tying a series in the most important time of the year.

Now I want to point out that I don´t believe in these statements. I´m just showing that it is very easy to choose arguments that suits your agenda.

In this case I am not surprised that the numbers vary so much when it is such a small sample. I find it highly unrealistic that these numbers shows such clear clutchness when larger studies are having problems finding it at all. That is a hot topic and I am not actually sure what I believe but I am sure that if clutchness exists you are overrestimating it in this example because I find it very hard to believe that a goalie that has an amazing "cluthgap" of a staggering +24% when trailing suddenly are -4% in all other games in the playoffs where every game is a high pressure game. A 28% clutch swing. That would be the most selective clutch ability I have ever seen. In this case random variance as part of the explanation is a much better hypothesis in my eyes.

I would like to say though that I would guess there are some truth to the numbers. I think Roy might have had a little better resilience when down. I think Hasek might have been a little more at risk of breaking down and I would guess that on the whole these differences are small and that they are very close in the playoffs with an edge to Roy because of his slightly better stats over a larger sample that resulted in a phenomenal trophy case. This is what the numbers also tell me. Just not to the simplified extent your very selective analysis claims.

 

[Dennis Bonvie]

It's not a question of whether you buy it or not. It's been exhibited statistically.

You're free to not buy it.

So why does this unbelievable phenomenon occur?

Was the guy counting shots the same person over 13 years? Or was it a massive conspiracy?

Did he have a bad seat (apparently the only guy in his job title in the NHL to have one)?

Could it be the NJ counts are closer to correct and everyone else overcounts?

Did he hate Brodeur? Or was he a gambler winning sucker bets on shots on goal?

Is it even a possibility that the consistently great defensive team with a World Class goaltender usually get a lead at home and then strangle the life out of their opponents? Certainly its easier to cash it in on the road when a comeback looks hopeless. I know I've seen a lot of 3-1, 3-0, 2-0 third periods in NJ where the puck does not go near either net for long stretches.

I would say the numbers are unusual. But I would also say the Devils have been an unusual team compared to the norm for all those years. Remember, that 6 shot shutout in game 6 of the playoffs was in Toronto, not New Jersey.