Dominik Hasek greatest goalie ever?

TheDevilMadeMe

Registered User
Aug 28, 2006
52,271
6,980
Brooklyn
You cannot simply erase 1994 due to small sample size. Why? Because when goal support gets so low, it is impossible for the sample size to get any larger.

Let's not pretend that 1994 was characteristic of the goal support Hasek got throughout his time in Buffalo. Compare Hasek's goal support to Brodeur's for the rest of the 1990s.

In 1995, Brodeur received excellent goal support (3.35 per game) as the Devils won the Cup, while Buffalo gave Hasek a modest 2.60 per game. However, Hasek played so poorly in 1995 that goal support likely did not matter.

Hasek's line in the 1995 playoffs: 1-4, .863 Save %, 3.50 GAA. Hasek's save percentage in 1995 was lower than any playoff Roy or Brodeur had as a starter, despite the fact that Roy got his start in an era when save percentages were typically lower. Brodeur's lowest save % = .881 in 2010. Roy's lowest playoff save % = .873 in 1987. (Brodeur played 32 minutes in the 1992 playoffs as a rookie and had a .800 %, but that wasn't even a full game's worth of work). Sure, 1995 is just one year, but it shouldn't count any less than 1994 right?

In 1996, NJ and Buffalo both missed the playoffs. NJ missed because they couldn't score - They had the second best GA in the league but the second worst GF in the league. Buffalo was 14/26 in GF, 16/26 in GA. Buffalo appears to have missed the playoffs because Hasek was peppered with shots, not lack of goal support.

In the late 1990s, Hasek got more goal support than Brodeur did in the playoffs.
To quote myself from the main board thread:

Sabres lacked star power up front, but they knew how to score goals as a team.

In the 1997-98 regular season, they scored 211 goals (2.57 per game) barely below the league average of 216 goals. In the playoffs, their offense stepped up, scoring 46 goals in 15 games (3.07 goals per game).

In the 1998-99 regular season, the Sabres scored 207 goals (2.52 gpg), not that far behind the league average which was again 216.
Once again, they stepped up big time in the playoffs, scoring 59 goals in 21 games (2.81 GPG).

Buffalo goal support
1997: 27 in 12 (Steve Shields started 9 of 12 games)
1998: 46 in 15
1999: 59 in 21

Total: 132 in 48 = 2.75 gpg

NJ goal support
1997: 27 in 10
1998: 12 in 6
1999: 18 in 7

Total: 57 in 23 = 2.48 gpg
 

Rhiessan71

Just a Fool
Feb 17, 2003
11,618
24
Guelph, Ont
Visit site
Why is Hasek's goal support even being used as an excuse here?

I mean, is it not realised that when Dallas beat them in '99, Belfour had even less support than Hasek did.
Dallas scored 64 goals in 23 games, 2.78 GpG
Buffalo scored 59 goals in 21 games, 2.81 GpG
And leading up to the finals Buffalo scored 3.33 GpG while Dallas only scored 3.00 GpG.

In 1986, Roy received even less goal support than Hasek did in '99, 56 goals in 20 games,2.80 GpG.
Meanwhile, their opponent Calgary scored 81 goals in 22 games, 3.68 GpG.
Again, leading up to the finals, Habs scored 2.73 Gpg and Calgary 4.00 Gpg.

So Roy received less goal support in the middle of the freewheeling 80's than Hasek did in the middle of the dead puck era.

Can we put this BS excuse to bed already?
 

Dennis Bonvie

Registered User
Dec 29, 2007
29,206
17,561
Connecticut
Let's not pretend that 1994 was characteristic of the goal support Hasek got throughout his time in Buffalo. Compare Hasek's goal support to Brodeur's for the rest of the 1990s.

In 1995, Brodeur received excellent goal support (3.35 per game) as the Devils won the Cup, while Buffalo gave Hasek a modest 2.60 per game. However, Hasek played so poorly in 1995 that goal support likely did not matter.

Hasek's line in the 1995 playoffs: 1-4, .863 Save %, 3.50 GAA. Hasek's save percentage in 1995 was lower than any playoff Roy or Brodeur had as a starter, despite the fact that Roy got his start in an era when save percentages were typically lower. Brodeur's lowest save % = .881 in 2010. Roy's lowest playoff save % = .873 in 1987. (Brodeur played 32 minutes in the 1992 playoffs as a rookie and had a .800 %, but that wasn't even a full game's worth of work). Sure, 1995 is just one year, but it shouldn't count any less than 1994 right?

In 1996, NJ and Buffalo both missed the playoffs. NJ missed because they couldn't score - They had the second best GA in the league but the second worst GF in the league. Buffalo was 14/26 in GF, 16/26 in GA. Buffalo appears to have missed the playoffs because Hasek was peppered with shots, not lack of goal support.

In the late 1990s, Hasek got more goal support than Brodeur did in the playoffs. To quote myself from the main board thread:

But that's only in one series.

In the finals against Colorado and Roy in 2001, Brodeur faced 146 shots and allowed 19 goals, a save percentage of .768. And that's in 7 games. Consider the Devils were on the verge of winning that series getting goalkeeping like that.
 

TheDevilMadeMe

Registered User
Aug 28, 2006
52,271
6,980
Brooklyn
But that's only in one series.

In the finals against Colorado and Roy in 2001, Brodeur faced 146 shots and allowed 19 goals, a save percentage of .768. And that's in 7 games. Consider the Devils were on the verge of winning that series getting goalkeeping like that.

Brodeur was not good in the 2001 finals, but saving 127 of 146 shots would give him a save percentage of .870.
 

TheDevilMadeMe

Registered User
Aug 28, 2006
52,271
6,980
Brooklyn
Why is Hasek's goal support even being used as an excuse here?

I mean, is it not realised that when Dallas beat them in '99, Belfour had even less support than Hasek did.
Dallas scored 64 goals in 23 games, 2.78 GpG
Buffalo scored 59 goals in 21 games, 2.81 GpG
And leading up to the finals Buffalo scored 3.33 GpG while Dallas only scored 3.00 GpG.

In 1986, Roy received even less goal support than Hasek did in '99, 56 goals in 20 games,2.80 GpG.
Meanwhile, their opponent Calgary scored 81 goals in 22 games, 3.68 GpG.
Again, leading up to the finals, Habs scored 2.73 Gpg and Calgary 4.00 Gpg.

So Roy received less goal support in the middle of the freewheeling 80's than Hasek did in the middle of the dead puck era.

Can we put this BS excuse to bed already?

Right. As others have said, those Buffalo teams just keep getting worse and worse as the years go on...

Roy (and Brodeur) did have better defensemen in front of them though.
 

GuineaPig

Registered User
Jul 11, 2011
2,425
206
Montréal
"Goal support" can be a somewhat misleading way of representing the strength of a team. Shot prevention is just as important.

If goalie A and goalie B both have 0.920 save percentages (assuming equal shot quality), Team A preventing 5 more shots a game than Team B is equivalent to boosting goal support by 0.4 goals/game (a bit more, really, because of empty net goals).
 

TheDevilMadeMe

Registered User
Aug 28, 2006
52,271
6,980
Brooklyn
"Goal support" can be a somewhat misleading way of representing the strength of a team. Shot prevention is just as important.

If goalie A and goalie B both have 0.920 save percentages (assuming equal shot quality), Team A preventing 5 more shots a game than Team B is equivalent to boosting goal support by 0.4 goals/game (a bit more, really, because of empty net goals).

Right. I don't think anyone is claiming that Buffalo was as as good defensively as Montreal or NJ.

But it's getting ridiculous how "Buffalo was a worse overall team than Montreal or NJ" (true) has been getting stretched into statements like "Buffalo was worse than Montreal and NJ in every way but in goal" (false) or "How did you expect Hasek to do anything with such crappy goal support" (misleading because it insinuates his goal support was worse than what Brodeur and Montreal Roy regularly got).
 

GuineaPig

Registered User
Jul 11, 2011
2,425
206
Montréal
Just picking two examples to further illustrate my point, Brodeur's '00 Devils and Hasek's '99 Sabres (playoffs only):

Devils
2.61 G/game
23.3 SA/game
Win threshold: 0.888

Sabres
2.67 G/game
31.3 SA/game
Win threshold: 0.914

Even though the Sabres' goal support was slightly higher, it was still much tougher to win with that squad than with the Devils (even accounting for shot prevention/undercounting).
 

TheDevilMadeMe

Registered User
Aug 28, 2006
52,271
6,980
Brooklyn
Just picking two examples to further illustrate my point, Brodeur's '00 Devils and Hasek's '99 Sabres (playoffs only):

Devils
2.61 G/game
23.3 SA/game
Win threshold: 0.888

Sabres
2.67 G/game
31.3 SA/game
Win threshold: 0.914

Even though the Sabres' goal support was slightly higher, it was still much tougher to win with that squad than with the Devils (even accounting for shot prevention/undercounting).

The 2002 Red Wings would be a much fairer comparison to the 2000 Devils. I stopped the comparison after 1999, because Brodeur's team was so obviously stacked in 2000 and 2001 and Hasek's was even more stacked in 2002.

I mean, telling us the 2000 Devils were a much better team than the 1999 Sabres isn't exactly breaking new ground. :)
 

Rhiessan71

Just a Fool
Feb 17, 2003
11,618
24
Guelph, Ont
Visit site
"Goal support" can be a somewhat misleading way of representing the strength of a team. Shot prevention is just as important.

If goalie A and goalie B both have 0.920 save percentages (assuming equal shot quality), Team A preventing 5 more shots a game than Team B is equivalent to boosting goal support by 0.4 goals/game (a bit more, really, because of empty net goals).

Then why is it being used to promote Hasek in the first damned place???
 

GuineaPig

Registered User
Jul 11, 2011
2,425
206
Montréal
Then why is it being used to promote Hasek in the first damned place???

I would assume because Hasek's goal support was often lower than Roy's Avs teams.

The 2002 Red Wings would be a much fairer comparison to the 2000 Devils. I stopped the comparison after 1999, because Brodeur's team was so obviously stacked in 2000 and 2001 and Hasek's was even more stacked in 2002.

I mean, telling us the 2000 Devils were a much better team than the 1999 Sabres isn't exactly breaking new ground. :)

True. Those Devils teams were obviously superior, and that was kind of the point I was making (in that they were obviously superior, but still had "inferior" goal support).

Just for kicks, here's Roy's '86 Habs (I assumed they scored two empty netters that playoff, because I can't find the actual stats for those playoffs).

'86 Habs
2.7 G/game
25.3 SA/game
Win threshold: 0.893

Pretty much reinforces what's already been said. They were significantly worse than Hasek's '99 Sabres.
 

struckbyaparkedcar

Guilty of Being Right
Mar 1, 2008
18,243
1,847
Upstate NY
How are they significantly worse if they scored more (obviously not adjusted for era but still), allowed less shots and had a lower win threshold?

What am I missing here?
 

GuineaPig

Registered User
Jul 11, 2011
2,425
206
Montréal
How are they significantly worse if they scored more (obviously not adjusted for era but still), allowed less shots and had a lower win threshold?

What am I missing here?

In '99, the average save percentage was 0.908. The '99 Sabres only needed slightly above-average goaltending (or average starting goaltending, essentially) in order to have an even goal differential in the '99 playoffs (obviously that wouldn't be the result, but we're speaking hypothetically).

In '86, the average save percentage was 0.874. The Habs would've needed much better goaltending than average to achieve an even goal differential.

Quick run through the calculator indicates that a goalie would have to be 6% better (than average) on the Sabres and 18% better on the Habs to achieve equal goal differential. A significant difference, even accounting for the inferior talent pool.
 

quoipourquoi

Goaltender
Jan 26, 2009
10,123
4,126
Hockeytown, MI
Just for kicks, here's Roy's '86 Habs (I assumed they scored two empty netters that playoff, because I can't find the actual stats for those playoffs).

'86 Habs
2.7 G/game
25.3 SA/game
Win threshold: 0.893

Pretty much reinforces what's already been said. They were significantly worse than Hasek's '99 Sabres.

And with the 1993 Montreal Canadiens:

3.02 goals per game
30.0 shots allowed per game
Win Threshold: .899
Avg. Save Percentage: .885
 

ricky0034

Registered User
Jun 8, 2010
14,862
6,937
In '99, the average save percentage was 0.908. The '99 Sabres only needed slightly above-average goaltending (or average starting goaltending, essentially) in order to have an even goal differential in the '99 playoffs (obviously that wouldn't be the result, but we're speaking hypothetically).

In '86, the average save percentage was 0.874. The Habs would've needed much better goaltending than average to achieve an even goal differential.

Quick run through the calculator indicates that a goalie would have to be 6% better (than average) on the Sabres and 18% better on the Habs to achieve equal goal differential. A significant difference, even accounting for the inferior talent pool.

there's a pretty huge problem with this though

your win threshold thing is based on Playoff stats while those average save percentage stats you just used are from the regular season

if you use Playoff stats for the average save percentage too then it's .915 for 99 and .895 for 86

so really they both needed very slightly above average Goaltending

And with the 1993 Montreal Canadiens:

3.02 goals per game
30.0 shots allowed per game
Win Threshold: .899
Avg. Save Percentage: .885

you're also using Playoff stats for the threshold and regular season stats for the average save percentage

the average save percentage for 93 in the Playoffs was .896
 

quoipourquoi

Goaltender
Jan 26, 2009
10,123
4,126
Hockeytown, MI
there's a pretty huge problem with this though

your win threshold thing is based on Playoff stats while those average save percentage stats you just used are based on stats from the regular season

if you use Playoff stats for the average save percentage too then it's .915 for 99 and .895 for 86

so really they both needed very slightly above average Goaltending



you're also using Playoff stats for the threshold and regular season stats for the average save percentage

the average save percentage for 93 in the Playoffs was .896

Using the average for the playoffs specifically would be a mistake for several reasons:

1. It's not representative of a league average, since many league goalies are not a part of the playoffs.

2. When top-end goaltenders have an unusually bad playoff - like, say, Dominik Hasek in 1995 - it majorly skews where the average save percentage in the playoffs ends up with in comparison to the average save percentage in the regular season.

3. The average save percentage in the playoffs is predominately going to be made up of the high-end goalies you are looking at, because they play the most games: 1986 is four parts Patrick Roy, four parts Mike Vernon, three parts John Vanbiesbrouck, two parts Mike Liut, but only a dash of the goaltenders who played poorly and subsequently got swept in three games.

4. It's a much smaller sample size of games reflecting what goaltenders in a given year are actually like.


When you look at league averages, use the league average. Simple.
 

ricky0034

Registered User
Jun 8, 2010
14,862
6,937
Using the average for the playoffs specifically would be a mistake for several reasons:

1. It's not representative of a league average, since many league goalies are not a part of the playoffs.

2. When top-end goaltenders have an unusually bad playoff - like, say, Dominik Hasek in 1995 - it majorly skews where the average save percentage in the playoffs ends up with in comparison to the average save percentage in the regular season.

3. The average save percentage in the playoffs is predominately going to be made up of the high-end goalies you are looking at, because they play the most games: 1986 is four parts Patrick Roy, four parts Mike Vernon, three parts John Vanbiesbrouck, two parts Mike Liut, but only a dash of the goaltenders who played poorly and subsequently got swept in three games.

4. It's a much smaller sample size of games reflecting what goaltenders in a given year are actually like.


When you look at league averages, use the league average. Simple.



the massive difference between the numbers for the regular season and Playoffs for 86 and the tiny difference between them for 99 suggests that there was a way more defensive playstyle in the 86 Playoffs compared to the 86 regular season while there was comparatively very little difference between the two in 99

further supporting this is the goals per game stats

7.937 in the 86 regular season and 6.500 in the 86 Playoffs

5.266 in the 99 regular season and 5.116 in the 99 Playoffs

it's really stupid to just ignore that and mix numbers from the regular season and Playoffs
 
Last edited:

quoipourquoi

Goaltender
Jan 26, 2009
10,123
4,126
Hockeytown, MI
the massive difference between the numbers for the regular season and Playoffs for 86 and the tiny difference between them for 99 suggests that there was a massively more defensive playstyle in the 86 Playoffs compared to the 86 regular season while there was comparatively very little difference between the two in 99

Actually, it suggests that Patrick Roy played 1218 minutes at .049 above what the league average goalie was expected to contribute, thereby making the average save percentage in the playoff significantly higher.

Leaders in Minutes - 1986 Playoffs
1. Vernon, 1229
2. Roy, 1218
3. Vanbiesbrouck, 899
4. Wregget, 607
5. Millen, 586

Total Minutes Played for All Goalies: 8738

So, Patrick Roy directly factors into 14% of that playoff average you would like to use instead, as opposed to the regular season leaders in minutes factoring into just 3.5% of the regular season average. Now do you see why it makes no sense to use to use the playoff average? You're comparing Patrick Roy to an average that is 14% Patrick Roy.
 

ricky0034

Registered User
Jun 8, 2010
14,862
6,937
Actually, it suggests that Patrick Roy played 1218 minutes at .049 above what the league average goalie was expected to contribute, thereby making the average save percentage in the playoff significantly higher.

Leaders in Minutes - 1986 Playoffs
1. Vernon, 1229
2. Roy, 1218
3. Vanbiesbrouck, 899
4. Wregget, 607
5. Millen, 586

Total Minutes Played for All Goalies: 8738

So, Patrick Roy directly factors into 14% of that playoff average you would like to use instead, as opposed to the regular season leaders in minutes factoring into just 3.5% of the regular season average. Now do you see why it makes no sense to use to use the playoff average? You're comparing Patrick Roy to an average that is 14% Patrick Roy.

and Hasek was 11% of 99's

but fine let's remove Roy and Hasek

it becomes:

.892 for 86
.912 for 99

99 just decreased as much as 86 did

....and there's still an absolutely massive difference between the 86 regular season and Playoff numbers
 

quoipourquoi

Goaltender
Jan 26, 2009
10,123
4,126
Hockeytown, MI
and Hasek was 11% of 99's

but fine let's remove Roy and Hasek

it becomes:

.892 for 86
.912 for 99

99 just decreased as much as 86 did

....and there's still an absolutely massive difference between the 86 regular season and Playoff numbers

Because Mike Vernon was worth another 14% of the 1986 number you want to use... that's two goalies who were their teams' Conn Smythe favorites comprising 28% of a number you want to call the "league average." And then that number is also mostly derived from the goalies that competed for the Division Title. Eight teams make up 76% of the statistic. That's not what a league average is.

Listen, if you don't get it, you don't get it. Just don't massage the numbers to the point where playoff goaltenders have to chase their own tails just because the below-average goaltenders don't play as much especially when there is a best-of-five round. League average goaltenders cannot be derived from a statistic that is 76% made up by the top eight teams.
 

Ad

Upcoming events

Ad

Ad

-->