Round 2, Vote 1 (HOH Top Goaltenders)

[Nalyd Psycho]

Then may I ask how Brodeur has caught your interest enough outside of 1997 (3rd) and 2007 (3rd) for you to have cited the head-to-head Vezina voting record?

I was just looking at it. I've always heard that Brodeur was always behind both Hasek and Roy, but that isn't the case. I honestly have no idea how I'm voting this round.

 

[vadim sharifijanov]

i don't disagree with what TDDM says above. and really i don't know that i'm advocating for dryden so much as asking questions because i don't know how to place his career.

but if you'll all humour me while i got into a "what if," what if dryden hadn't retired? he's gone the same year he retires, lemaire, bowman, and cournoyer also all leave. cournoyer was on his last legs anyway, but obviously lemaire and bowman would have made an impact had they stayed.

if dryden keeps playing until, say, roy takes over in '86, that's six extra seasons. and those habs teams were still very good teams, but you get the rapid decline of lafleur after 1980, the ends of the road for savard and lapointe. slowly, those guys are replaced by the pillars of the '86 team, minus the amazing rookie class of '86 (roy, claude lemieux, lalor, skrudland, richer, and dahlin).

being that dryden got the lion's share of habs hart votes before lafleur's emergence, then became a hart afterthought even though he put up four first all-star teams in four years to end his career. with lafleur declining, robinson no longer a top three norris guy after 1981, and the rest of the big three gone, i would think dryden gets a lot more credit for the habs' success.

now i don't think any of those habs teams from '80 to '85 dethrones the isles/oilers dynasties with dryden instead of the plumbers they replaced him with, but i do think they go deep in the playoffs more often than not. as it stands, that team lost in the seventh game of the second round twice, and made a conference finals. with dryden playing behind an increasingly defensive team, led by scoring stars named naslund and smith instead of lafleur and shutt, i see him not only adding significant rep to his resume, but also those years would have made us reevaluate his importance to the habs dynasty. similar to marty brodeur; his vezinas post-stevens/niedermayer went a long way to legitimizing his place in the top six, and last year's playoff run showed that he could get it done in the playoffs too.

which is all to say, i think, that for some dryden actually gets penalized and not rewarded for his playoff record. i think if he'd been around for the post-dynasty years, he probably would have gotten more credit than maybe he would have deserved for leading an excellent defensive team without an art ross talent up front. but i think in balance it would have evened out to make his career less polarizing.

 

[quoipourquoi]

Just a quick note on Ken Dryden (I'm about to head-out with my lady-friend for the next two days, but I'll try to log-in when she's asleep ), if he had stuck around at near the same level (no Sawchuk-esque drop), I'd have him in the mix with Plante and Roy and Hasek. I think very highly of him, and hope that we have a knock-down-drag-out fight with him, Hall, Sawchuk, and Brodeur.

 

[tarheelhockey]

Just a quick note on Ken Dryden (I'm about to head-out with my lady-friend for the next two days, but I'll try to log-in when she's asleep ), if he had stuck around at near the same level (no Sawchuk-esque drop), I'd have him in the mix with Plante and Roy and Hasek. I think very highly of him, and hope that we have a knock-down-drag-out fight with him, Hall, Sawchuk, and Brodeur.

 

[TheDevilMadeMe]

I was just looking at it. I've always heard that Brodeur was always behind both Hasek and Roy, but that isn't the case. I honestly have no idea how I'm voting this round.

Brodeur had his ups and downs for the first half of his career, but I think that during the time his career overlapped with Roy's, Brodeur was overall the better regular season goalie by a little bit. Both were well behind Hasek in the regular season over the whole time frame (though Brodeur was awfully close to Hasek in 1996-97, that is just one season).

Of course, this again ignores the fact that Roy's regular season peak was from about 1988-1992.

And in the playoffs, I think Roy was generally the best of the three all through, though Hasek and Brodeur each had their moments, as well.

 

[seventieslord]

You know what I want to see?

A save percentage system for Roy/Brodeur/Hasek that takes into account:

- league averages (obviously)
- shot undercounting
- playing disproportionately higher numbers of home or road games
- the effect of facing the fewest PPs in the league
- shots prevented by superior puckhandling counted as saves (I assume just Brodeur gets this benefit)

and all mathematically sound, of course.

 

[VanIslander]

Ken Dryden was often the TOP star on a talent-laden Habs team of all-time greats. THAT is significant.

Martin Brodeur was often the 3rd to 5th best Devil during the span of the franchise's cup years (but yeah, he's got compiler stats that are better and he faced lesser competition in his later years as clearly goaltending in the 2000s wasn't as strong as the 1990s in terms of elite level players to compete for honours).

Having seen both of their careers, I still think Dryden a greater goalie than Brodeur.

 

[TheDevilMadeMe]

You know what I want to see?

A save percentage system for Roy/Brodeur/Hasek that takes into account:

- league averages (obviously)
- shot undercounting
- playing disproportionately higher numbers of home or road games
- the effect of facing the fewest PPs in the league
- shots prevented by superior puckhandling counted as saves (I assume just Brodeur gets this benefit)

and all mathematically sound, of course.

Agreed. Add a couple of other noteworthy modern goalies too.

The first four should be relatively easy to quantify if one wants to take the time. IMO, the different ways of counting shots at arenas is probably the #1 source of error* in officially recorded save percentages. Quantifying arena effects should be relatively straightforward, though time consuming if you want to be complete and look at all arenas. I agree with TCG that looking at road save percentage only is a good rough estimate if you don't want to do all the work.

Shots prevented by superior puckhandling will be much more difficult to quantify, as it not possible to determine with precision when a shot would have been taken but for a goalie breaking up the forecheck. Perhaps some form of Corsi comparison between a goalie and his backup would be applicable? The problem is finding a large enough sample size of backup games, though. I would be satisfied with not quantifying puck handling and just giving goalies such as Brodeur and Belfour some "mental bonus points" for puckhandling.

*I don't consider ES/PP/PK splits to be a source of "error," so much as variables that need to be taken into account and are easy to do so.

 

[BM67]

Ken Dryden was often the TOP star on a talent-laden Habs team of all-time greats. THAT is significant.

Martin Brodeur was often the 3rd to 5th best Devil during the span of the franchise's cup years (but yeah, he's got compiler stats that are better and he faced lesser competition in his later years as clearly goaltending in the 2000s wasn't as strong as the 1990s in terms of elite level players to compete for honours).

Having seen both of their careers, I still think Dryden a greater goalie than Brodeur.

So why did his teammates vote Brodeur team MVP year after year?

 

[TheDevilMadeMe]

Ken Dryden was often the TOP star on a talent-laden Habs team of all-time greats. THAT is significant.

Martin Brodeur was often the 3rd to 5th best Devil during the span of the franchise's cup years (but yeah, he's got compiler stats that are better and he faced lesser competition in his later years as clearly goaltending in the 2000s wasn't as strong as the 1990s in terms of elite level players to compete for honours).

Having seen both of their careers, I still think Dryden a greater goalie than Brodeur.

You seriously think Brodeur was behind more All-time Greats on the 90s/00s Devils than Dryden was on the 1970s Canadiens?????

Guy Lafleur, Larry Robinson, Serge Savard, Guy Lapointe, Jacques Lemaires, Steve Shutt vs. Scott Stevens, Scott Niedermayer... Patrik Elias?

Brodeur's teammates voted him the Most Valuable Devil 10 times (1996, 1997, 1998, 1999, 2000, 2001*, 2002, 2003, 2007, 2008).

*Tied with Elias

Source = http://en.wikipedia.org/wiki/List_of_New_Jersey_Devils_award_winners

IMO, Brodeur was the clearcut #2 behind Scott Stevens in the playoffs, but he was easily the MVP of the team in the regular season.

Dryden does beat Brodeur 1-0 in Conn Smythes.

 

[quoipourquoi]

the different ways of counting shots at arenas is probably the #1 source of error in officially recorded save percentages.

Personally, I find that the Luongo-esque rhythm goaltending is just as big of an issue. The difference between making 26 of 28 saves and 28 of 30 saves because the latter goaltender wants his team to let him feel the puck at the beginning of a game and the former goaltender is good-to-go from minute-one carries some heavy weight in cumulative save percentages. If a goaltender needs his team to let him see some blockable shots just to get his head in the game, then are the saves on those blockable shots really a positive personal contribution?


EDIT: @ tarheel's picture

 

[Canadiens1958]

Brodeur / Luongo Irony

Little sidebar that should bring a smile or two. Martin Brodeur and Roberto Luongo are both from St. Leonard, Quebec, both learned the craft in the same youth hockey association a few years apart, same basic goalie training template and path to the NHL. Both are somewhat unique and distinct NHL goalies.

The two St. Leonard - now merged into Montreal, arenas are named after each. Ironically the bigger better one named after Martin Brodeur.

 

[MXD]

1. Longevity. See Post 104. Dryden dominated the NHL for 7 seasons, but every other goalie up for consideration this round played at a fairly high level between 15-19 seasons. Count me among those who thinks Terry Sawchuk wasn't playing at an all-time great level after his first 5 seasons, but something is to be said for just keeping an NHL starter's job in a 6 team league.

2. Dryden was not as good against the European East/West game as he was against the traditional North American game.

In the Summit Series:


http://www.1972summitseries.com/dryden.html

#1 makes sense (although one could claim that Dryden made the difference more than Sawchuck ever did).

#2... well, it seems kindof unfair to "blame" Dryden for something that Hall and Sawchuck (and, to a certain extent, Plante) could never experiment to begin with. They might have fared worst than Dryden.

To be honest, to me, Dryden is either 6th or 7th, but that point will not be taken into consideration.

 

[Czech Your Math]

Shots prevented by superior puckhandling will be much more difficult to quantify, as it not possible to determine with precision when a shot would have been taken but for a goalie breaking up the forecheck. Perhaps some form of Corsi comparison between a goalie and his backup would be applicable? The problem is finding a large enough sample size of backup games, though. I would be satisfied with not quantifying puck handling and just giving goalies such as Brodeur and Belfour some "mental bonus points" for puckhandling.

If you look at Brodeur's career totals vs. a composite of his back-ups in each season, his back-ups faced ~4-5% more shots against/60 min. than did Brodeur. Increasing Brodeur's career SA/60 to the level of his back-ups would change his career SV% from ~.913 to ~.917. This seems like a reasonable adjustment, but the only concern is how consistent was that variation in SA/60 between Brodeur and his back-ups?

If one weights the back-ups SA/60 by Brodeur's GP, the difference between the two becomes negligible. If one uses a simple average of each season ('94-'12), the difference in SA/60 is only ~1% (instead of ~4-5%). In the seasons where the back-ups played very sparingly, their SA/60 (as a whole over multiple seasons) was actually less than for Brodeur. In the seasons when the back-ups had more significant playing time (800+ minutes), they consistently had a higher SA/60.

Seasons in which goalies other than Brodeur played > 800 minutes:

YEAR | MBSA60 | OtherMin | OtherSA60
1994 | 28.3 | 2,470 | 29.1
1997 | 25.5 | 1,152 | 27.1
1998 | 22.8 | 855 | 26.0
2009 | 28.8 | 3,151 | 29.3
2011 | 25.3 | 1,835 | 27.2
2012 | 26.0 | 1,591 | 27.1
6 Yr Total | 25.7 | 11,054 | 28.1
Other Yrs | 25.6 | 6,738 | 24.9
TOTAL | 25.6 | 17,792 | 26.9

It could be random variation, but it seems likely that there could be another cause that explains it to a large degree. For instance, when the back-up played sparingly, the team may have tried to play even more defensively to protect that goalie, while this may not have been the case when the back-up was playing more regularly (and the team was more comfortable with him in net). It still seems open to interpretation, rather than provide the more concrete solution one would prefer.

 

[pluppe]

I think this Hasek stat from another thread is one of the most amazing stats in hockey:

On the flip side, here's career losing percentage top 25:

# Player Losing %
1 Kenneth Dryden 14.4%
2 Gerry Cheevers 24.4%
3 Manny Legace 27.1%
4 Michel Larocque 28.5%
5 Chris Osgood 29.2%
6 Bill Durnan 29.2%
7 Andy Moog 29.3%
8 Jacques Plante 29.4%
9 Bob Froese 29.8%
10 Niklas Backstrom 29.8%
11 Dominik Hasek 30.3%
12 Marty Turco 30.4%
13 Martin Brodeur 30.6%
14 Patrick Roy 30.6%
15 Vesa Toskala 30.8%
16 George Hainsworth 31.2%
17 Wayne Stephenson 31.4%
18 Evgeni Nabokov 31.6%
19 Pete Peeters 31.7%
20 Ryan Miller 31.7%
21 Rejean Lemelin 32.0%
22 Richard Wamsley 32.2%
23 Bernie Parent 32.6%
24 Cristobal Huet 33.1%

Again, Dryden with an other wordly mark here -- he lost less than 3 out of every 20 games played. Wow!

One powerhouse after another represented and then Hasek in the middle. For example beating both Roy and Brodeur whom I think all agree played for much better teams on average.

And you can´t blame Dryden for not taking advantage of a very good situation.

(this is from 2010 so Brodeurs numbers are probably a bit off)

 

[BM67]

Here's a chart of home/road splits through 2011-12, and including Roy's entire career.

Player|H Min%|H SV%|R SV%|H SOG/60|R SOG/60|SOG Dif|SOG Dif%
Belfour|53.69%|.904|.909|25.33|28.21|2.88|11.37%
Brodeur|50.90%|.911|.915|24.28|27.03|2.75|11.33%
Burke|50.22%|.905|.898|28.73|31.47|2.74|9.54%
Hasek|52.20%|.925|.919|27.81|28.88|1.07|3.85%
Joseph|52.80%|.909|.903|29.08|30.48|1.40|4.81%
Lundqvist|53.93%|.920|.920|26.94|30.04|3.10|11.51%
Luongo|53.78%|.921|.917|30.35|32.18|1.93|6.32%
Osgood|51.70%|.904|.907|25.40|27.19|1.79|7.05%
Roy|55.50%|.911|.909|27.02|29.77|2.75|10.18%
Vokoun|51.84%|.923|.910|30.79|30.64|-0.15|-0.49%

The higher the Dif the more likely, and stronger the home under-counting influence, and the lower the more likely, and stronger the home over-counting influence. There is an "era" variance in SOG/60, H/R difference, and SV% which makes the career values less than ideal benchmarks, but it stands as an illustration that there is something to investigate further.

Brodeur's Dif% has dropped slightly over the last two seasons, while Lundqvist's has jumped from average-ish to the top of the list.

Using the H/R split data from HockeyReference here is the league average SV% for
1988-89: Total: .879 Home: .880 Road: .878 Dif: .002
1989-90: Total: .881 Home: .884 Road: .878 Dif: .006
1990-91: Total: .886 Home: .891 Road: .881 Dif: .010
2006-07: Total: .905 Home: .907 Road: .903 Dif: .004
2007-08: Total: .909 Home: .910 Road: .908 Dif: .002
2008-09: Total: .908 Home: .911 Road: .907 Dif: .004
2009-10: Total: .911 Home: .914 Road: .908 Dif: .006
2010-11: Total: .913 Home: .914 Road: .912 Dif: .002
2011-12: Total: .914 Home: .916 Road: .911 Dif: .005

 

[TheDevilMadeMe]

Here's a chart of home/road splits through 2011-12, and including Roy's entire career.

Player|H Min%|H SV%|R SV%|H SOG/60|R SOG/60|SOG Dif|SOG Dif%
Belfour|53.69%|.904|.909|25.33|28.21|2.88|11.37%
Brodeur|50.90%|.911|.915|24.28|27.03|2.75|11.33%
Burke|50.22%|.905|.898|28.73|31.47|2.74|9.54%
Hasek|52.20%|.925|.919|27.81|28.88|1.07|3.85%
Joseph|52.80%|.909|.903|29.08|30.48|1.40|4.81%
Lundqvist|53.93%|.920|.920|26.94|30.04|3.10|11.51%
Luongo|53.78%|.921|.917|30.35|32.18|1.93|6.32%
Osgood|51.70%|.904|.907|25.40|27.19|1.79|7.05%
Roy|55.50%|.911|.909|27.02|29.77|2.75|10.18%
Vokoun|51.84%|.923|.910|30.79|30.64|-0.15|-0.49%

The higher the Dif the more likely, and stronger the home under-counting influence, and the lower the more likely, and stronger the home over-counting influence. There is an "era" variance in SOG/60, H/R difference, and SV% which makes the career values less than ideal benchmarks, but it stands as an illustration that there is something to investigate further.

Brodeur's Dif% has dropped slightly over the last two seasons, while Lundqvist's has jumped from average-ish to the top of the list.

Using the H/R split data from HockeyReference here is the league average SV% for
1988-89: Total: .879 Home: .880 Road: .878 Dif: .002
1989-90: Total: .881 Home: .884 Road: .878 Dif: .006
1990-91: Total: .886 Home: .891 Road: .881 Dif: .010
2006-07: Total: .905 Home: .907 Road: .903 Dif: .004
2007-08: Total: .909 Home: .910 Road: .908 Dif: .002
2008-09: Total: .908 Home: .911 Road: .907 Dif: .004
2009-10: Total: .911 Home: .914 Road: .908 Dif: .006
2010-11: Total: .913 Home: .914 Road: .912 Dif: .002
2011-12: Total: .914 Home: .916 Road: .911 Dif: .005

I'm not sure how to interpret this. Based on SOG Diff%, Roy seems to have almost as much of an undercounter as Brodeur (something I've never heard of before). But his home/road save% splits look normal.

Hasek appears to have had a somewhat generous shot counter in Buffalo (which doesn't surprise me), but nothing compared to Tomas Vokoun! The difference between Vokoun's recorded home and road save percentages is laughable.

 

[quoipourquoi]

I think this Hasek stat from another thread is one of the most amazing stats in hockey:



One powerhouse after another represented and then Hasek in the middle. For example beating both Roy and Brodeur whom I think all agree played for much better teams on average.

And you can´t blame Dryden for not taking advantage of a very good situation.

(this is from 2010 so Brodeurs numbers are probably a bit off)

Flawed statistic. WilliamRanford is simply taking Losses divided by Games Played - making no consideration for Games Decided, or the extra possibility to lose a game following the introduction of overtime and the removal of that possibility following 2005.

Under Pre-1983 conditions in which a regulation tie is exactly that this is the statline as of March 17th, 2009:

507-283-207 (.612): Roy
470-273-228 (.601): Brodeur
348-205-154 (.601): Hasek

Losing percentages would be 28.4% for Roy, 28.1% for Brodeur, and 29.0% for Hasek. Obviously, Brodeur has changed since then.

 

[BM67]

I'm not sure how to interpret this. Based on SOG Diff%, Roy seems to have almost as much of an undercounter as Brodeur (something I've never heard of before). But his home/road save% splits look normal.

Hasek appears to have had a somewhat generous shot counter in Buffalo (which doesn't surprise me), but nothing compared to Tomas Vokoun!

Roy has played at home with both under and over counting. Plus Roy's career started earlier than the others, so a larger "era" adjustment is going to be required. His baseline of normal probably isn't in line with the others.

Using the H/R split data from HockeyReference here is the league average SOGA/60 for
1988-89: Home: 28.42 Road: 31.60 Dif: 3.18 or 11.19%
1989-90: Home: 28.52 Road: 31.26 Dif: 2.74 or 9.61%.
1990-91: Home: 28.11 Road: 30.74 Dif: 2.63 or 9.36%.
2006-07: Home: 28.22 Road: 30.24 Dif: 2.02 or 7.16%.
2007-08: Home: 27.54 Road: 29.92 Dif: 2.38 or 8.64%.
2008-09: Home: 28.81 Road: 30.90 Dif: 2.09 or 7.25%.
2009-10: Home: 28.98 Road: 30.87 Dif: 1.89 or 6.52%.
2010-11: Home: 29.15 Road: 30.88 Dif: 1.73 or 5.93%.
2011-12: Home: 28.37 Road: 30.33 Dif: 1.96 or 6.91%.

 

[seventieslord]

I think this Hasek stat from another thread is one of the most amazing stats in hockey:



One powerhouse after another represented and then Hasek in the middle. For example beating both Roy and Brodeur whom I think all agree played for much better teams on average.

And you can´t blame Dryden for not taking advantage of a very good situation.

(this is from 2010 so Brodeurs numbers are probably a bit off)

IIRC though, it was a garbage stat. it was losses divided by GP, not even taking decisions into consideration.

 

[pluppe]

Flawed statistic. WilliamRanford is simply taking Losses divided by Games Played - making no consideration for Games Decided, or the extra possibility to lose a game following the introduction of overtime and the removal of that possibility following 2005.

Under Pre-1983 conditions in which a regulation tie is exactly that this is the statline as of March 17th, 2009:

507-283-207 (.612): Roy
470-273-228 (.601): Brodeur
348-205-154 (.601): Hasek

Losing percentages would be 28.4% for Roy, 28.1% for Brodeur, and 29.0% for Hasek. Obviously, Brodeur has changed since then.

IIRC though, it was a garbage stat. it was losses divided by GP, not even taking decisions into consideration.

Thanks for enhancing the value of the stat. Now tell me that it´s still not amazing that it is that close.

 

[TheDevilMadeMe]

Thanks for enhancing the value of the stat. Now tell me that it´s still not amazing that it is that close and that it says something about Hasek.

I think career winning and losing percentage is a worthless stat.

Would Brodeur be a better goalie if he retired 3 years ago before he started to decline? Because his career winning and losing percentages would have been better if he did.

Hasek is helped by the fact that he didn't come over to the NHL until his late 20s, so his growing pains are factored into his NHL winning percentages. He also spent his declining years playing for an absolutely stacked Red Wings team. And yes, he is hurt by playing for a weaker team during his prime.

There are reasons to like Hasek here, but I don't think career winning or losing percentage is meaningful.

 

[Canadiens1958]

1956 Playoffs - Jacques Plante

Detailed look at Jacques Plante's 1956 playoff performance. Data culled from the HSP project and reference to BM67 post #123 this thread.

Semi Final vs NY Rangers

Game 1 - Home, MTL 7 NYR 1, SOG 4/7/11, First goal 26:45 (3-1)
Game 2 - Home, MTL 2 NYR 4, SOG 10/10/7, First goal 3:45 (0-1)
Game 3 - Away, MTL 3 NYR 1, SOG 6/0/4, First goal, 16:02 (1-1)
Game 4 . Away, MTL 5 NYR 3, SOG 10/5/6, First goal,16:25 (2-1)
Game 5 - Home, MTL 7 NYR 0, SOG 7/12/10 Shut Out.

Notes. 3 PPG allowed 2-1st period, 1-3rd period. Game 2 was the Gordie Bell game - Rangers rallied around a last minute substitute goalie.

SV%
1st period .865
2nd period .971
3rd period .921
Home .,936
Away .871
Series .917

Final vs Detroit Red Wings

Game 1 - Home, MTL 6 DET 4, SOG 9/13/5, First goal 8:17 (0 - 1)
Game 2 - Home, MTL 5 DET 1, SOG 7/7/8, First goal 40:13 (3 - 1)
Game 3 - Away, MTL 1 DET 3, SOG 5/8/12, First goal 14:27 (0 - 1)
Game 4 . Away, MTL 3 DET 0, SOG 6/9/9, Shut Out
Game 5 - Home, MTL 3 DET 1, SOG 7/10/9 First goal 40:35 (3 - 1)

Notes. 3 PPG allowed 2-1st period, 1-2nd period.

SV%
1st period .941
2nd period .936
3rd period .907
Home .,920
Away .938
Series .927

Overall, 6 of 18 goals allowed were PP goals. Never allowed an opponent to comeback and win.

 

[pluppe]

I think career winning and losing percentage is a worthless stat.

Would Brodeur be a better goalie if he retired 3 years ago before he started to decline? Because his career winning and losing percentages would have been better if he did.

Hasek is helped by the fact that he didn't come over to the NHL until his late 20s, so his growing pains are factored into his NHL winning percentages. He also spent his declining years playing for an absolutely stacked Red Wings team. And yes, he is hurt by playing for a weaker team during his prime.

There are reasons to like Hasek here, but I don't think career winning or losing percentage is meaningful.

Well I saw you argue that Hasek probably started to peak around 21 and I would guess that goalies have better stats early than late so I am not so sure that it would change a whole lot but that is of course another what if discussion. And I don´t see how it matters that much which years you spend on a powerhouse. The important point is Hasek spent less than the others.

When you do the numbers from 27 for all 3 they all get around 66% in a simple wins + T/O from hockey-reference.

Here is an old post by overpass.

Goalie stats compared to backups
Goalie stats can be tricky to compare across teams. Team defence, goalie puckhandling ability, shot quality recording biases, etc, can all affect the goalie's stats.

Comparing goalies to the performance of their teammates is one way to implicitly adjust for many of these factors. Here is the goal prevention record of Curtis Joseph and his comtemporaries who have been drafted, as compared to their backup goalies. Lower is better. For example, Dominik Hasek allowed 0.77 goals for every goal his backups allowed on a per-game basis.

Goalie Vs Backups
Dominik Hasek 0.77
Curtis Joseph 0.88
Roberto Luongo 0.89
Patrick Roy 0.90
Martin Brodeur 0.91
Tom Barrasso 0.92
Ron Hextall 0.96
Ed Belfour 0.96
Mike Vernon 0.99
Grant Fuhr 1.03

Methodology notes: I have removed seasons in which the goaltender played less than 30 games. I have adjusted for scoring level, strength of opposition faced, and home/road games, but not back-to-backs. Some goalies had stronger backups than others - Ed Belfour and Grant Fuhr come to mind.

Again mind-boggling separation. When the stat is in line with most other stats that hint that Hasek helped his team significantly more than others I do think it says something.

 

[TheDevilMadeMe]

Well I saw you argue that Hasek probably started to peak around 21

Mischaracterizing my argument will get you nowhere. I said that I thought Hasek was capable of being a solid NHL starter at 21, a far cry from his peak. In my personal opinion, peak Hasek would have easily beaten out Jimmy Waite in Chicago, and he would have dominated the minor leagues when he was sent down.

Again mind-boggling separation. When the stat is in line with most other stats that hint that Hasek helped his team significantly more than others I do think it says something.

Hasek's GAA vs his backups is one of the strongest arguments in his favor, IMO. It's what finally convinced me (quite a few years back) that he had the best regular season peak of the modern era (I've always been a save % skeptic, though I've come around somewhat on that recently). But the quality of the backups does matter (Roy had great backups in Montreal), as do the matchups (most seasons, Brodeur only sat out against the very worst teams in the league).