Top-100 Hockey Players of All-Time - Round 2, Vote 2

[Sentinel]

Of course you can believe what you want to believe, but you should also back it up with an argument. Why does Hašek's head-to-head record vs Roy indicate who the better goaltender is, but Hašek's head-to-head record vs Hedberg does not indicate who the better goaltender is?

For reasons stated above. I made an argument. You and others are free to disagree.

Between goaltenders?

Between teams. Come on, now.

 

[K Fleur]

Ohhh, gimme dem Brodeur h2h numbers. He's the only goalie I have in my top 20 besides Roy.

Fight me irl

They only went “head to head” in one playoff series(1994). Brodeur won so he takes the World Cup of Soccer tiebreaker.

 

[Sentinel]

Especially when Hasek has *really good arguments that are more substantive*. But really this is just another way to pump a RW.

Yeah, that must be it... Especially given how clearly post-prime he was in Detroit. I guess next I will be "pumping" Alfredson and Modano.

*shrug*

 

[Sentinel]

Has a single person brought forward "wins" as a way to rank goaltenders in this? It's widely regarded as a ****ty stat for goalies that few people take seriously.

You have somehow found a worse metric than wins, so congrats?

I don't really care if like this metric or not. Done with you.

 

[The Macho King]

Ya know what, as long as you're consistent and use it if Sawchuk and Plante are in the same vote, go for it.

 

[Captain Bowie]

Yes we should. Neither Roy nor Hasek are worthy of top 5 All Time goalie status let alone top 5 overall.

Both required managed RS starts and took longer than most to claim #1 status with an NHL team.

Did they require managed RS starts? Or was the just the era they played in? I've seen you talk about the "era of tandem goalies" several times, but somehow you hold it against these two specifically for not doing something that wasn't done at the time?

From 50-67 when the schedule was 70 games, there were 29 times a goalie played all 70 games. Almost 2 each season (about 1/3 of the league)

From 86-02 the season has been 80-84 games, and from 21-30 teams. 146 seasons of 60+ games (8.5 per season, about 1/3 of the league). So playing all 70 games in the O6 era is about the same as playing at least 60/82 games in the late 80's + 09's.

Patrick Roy had 10 of those 60+ seasons.

So if you're argument against Roy is durability/lack of endurance, it's a poor one. But like usual, I'm not exactly sure what it is you are trying to argue.

 

[Theokritos]

Between teams. Come on, now.

Well, that's a pretty significant difference, no? You're not ranking teams here, not players. No-one would rank soccer goalies based on their head-to-head record.

 

[24 others]

I don't get the "Roy's playoffs" argument in the Roy vs. Hasek debate.

"So, Joe Thornton is a better regular season player, Justin Williams is a better playoff performer, therefore they both have a valid claim for being a better player"?

You have a 82 GP/year sample size which is comparable across teams and players and statistically robust as much as NHL data can be.

And then you have a crapshoot sample of a few games each year, which is generally unpredictable (see for instance the Art Ross or regular season PPG consistency versus the wild inconsistency of playoff scoring races and playoff PPG). All interpretations can only be made post-hoc, through heroic narratives and fandom-produced myths of greatness and "clutchness" (and if you dare speak otherwise then "hockey is about winning" is a reply which beats everything).

Yet, somehow these two sample are treated as equal when it comes to possible inference. Makes no sense to me.

 

[Captain Bowie]

"Wins" is an accepted metric for goalies, even though they play with twenty other people. This one is no less useless.

At any rate, I made my point, I like this metric, it means more to me than a bunch of games against meaningless opponents in RS (although, truth be told, in Buffalo's case, every game mattered), and I fully intend to stick with it.

That's all that really matters to me. People can use whatever metrics they want, as long as they are consistent throughout the process.

 

[The Macho King]

I don't get the "Roy's playoffs" argument in the Roy vs. Hasek debate.

"So, Joe Thornton is a better regular season player, Justin Williams is a better playoff performer, therefore they both have a valid claim for being a better player"?

You have a 82 GP/year sample size which is comparable across teams and players and statistically robust as much as NHL data can be.

And then you have a crapshoot sample of a few games each year, which is generally unpredictable (see for instance the Art Ross or regular season PPG consistency versus the wild inconsistency of playoff scoring races and playoff PPG). All interpretations can only be made post-hoc, through heroic narratives and fandom-produced myths of greatness and "clutchness" (and if you dare speak otherwise then "hockey is about winning" is a reply which beats everything).

Yet, somehow these two sample are treated as equal when it comes to possible inference. Makes no sense to me.

Joe and Williams have a colossal spread in the regular season, so the comparison doesn't make sense. I'd rather use something like maybe... I don't know, Thornton and Marty St. Louis. Thornton (seemingly) has the more impressive regular season resume (but it's fairly close), but factor in playoffs and MSL looks a whole lot better than Thornton.

Roy and Hasek aren't far apart in the regular season. There is separation, but even factoring in *only* regular season, Roy would be pretty damn close to Hasek. So the playoffs aren't the only comparison point between two disparate players, it's a weight on two fairly similar players.

How you rank the two is going to depend on mainly how strongly you rate Roy's regular seasons. Factoring in for era I think they're pretty impressive.

The other thing with Hasek ... and this is killing me when trying to place him... the dude quit on his team. Numerous times. I think if you ask the Sabres, they'd prefer Hasek at 90% over whoever his backup was (Fuhr? Roloson?) at 100%. But Hasek wouldn't play if he wasn't 100%. That... damn that f***ing bugs me. I can't imagine any other player on this vote who wouldn't play on one leg if they thought it would give their team a better chance to win except for him.

 

[Captain Bowie]

I don't get the "Roy's playoffs" argument in the Roy vs. Hasek debate.

"So, Joe Thornton is a better regular season player, Justin Williams is a better playoff performer, therefore they both have a valid claim for being a better player"?

You have a 82 GP/year sample size which is comparable across teams and players and statistically robust as much as NHL data can be.

And then you have a crapshoot sample of a few games each year, which is generally unpredictable (see for instance the Art Ross or regular season PPG consistency versus the wild inconsistency of playoff scoring races and playoff PPG). All interpretations can only be made post-hoc, through heroic narratives and fandom-produced myths of greatness and "clutchness" (and if you dare speak otherwise then "hockey is about winning" is a reply which beats everything).

Yet, somehow these two sample are treated as equal when it comes to possible inference. Makes no sense to me.

Do you surmise that the gap between Thornton's and Williams' regular season resume is the same size as the gap between Roy's and Hasek's?

 

[Theokritos]

For reasons stated above. I made an argument.

I didn't see an argument why one head-to-head team record should matter in a comparison between individual players while the other shouldn't.

 

[DannyGallivan]

The other thing with Hasek ... and this is killing me when trying to place him... the dude quit on his team. Numerous times. I think if you ask the Sabres, they'd prefer Hasek at 90% over whoever his backup was (Fuhr? Roloson?) at 100%. But Hasek wouldn't play if he wasn't 100%. That... damn that ****ing bugs me. I can't imagine any other player on this vote who wouldn't play on one leg if they thought it would give their team a better chance to win except for him.

Of course, one could say the epitome of "quitting on your team" was this...

 

[The Macho King]

Of course, one could say the epitome of "quitting on your team" was this...

Lol - touche.

 

[ChiTownPhilly]

I don't get the "Roy's playoffs" argument in the Roy vs. Hasek debate. [...]
You have a 82 GP/year sample size which is comparable across teams and players and statistically robust as much as NHL data can be.

And then you have a crapshoot sample of a few games each year, which is generally unpredictable (see for instance the Art Ross or regular season PPG consistency versus the wild inconsistency of playoff scoring races and playoff PPG).

Say, WHAT?!?!

Patrick Roy played over 240 post-season games. It's scarcely a "few-game-crapshoot-sample." He won three Smythes- pretty much earned, as part of that three-plus seasons quantity-equivalent of game samples. He did not achieve this by happenstance. Happenstance. Three times... Now, how does that Ian Fleming quote go???

"Once may be happenstance; twice, coincidence- but three times is Enemy Action."

 

[Canadiens1958]

Why, is Jacques Plante not eligible for this list or something?



Games Played 1994-2002:
1. Brodeur, 588
2. Roy, 548
3. Joseph, 533
4. Hasek, 528
5. Belfour, 515

I mean, if your standard is everything below Martin Brodeur is relatively low, then fine. But are you sure you are applying the same standard to all goalies and you aren't over-correcting for Hasek's allegedly low games played numbers? Patrick Roy was 3rd in GP from 1988-1992 during his statistical peak at a time when there weren't any Martin Brodeurs in the league, I don't really see why Hasek finishing fourth over a nine season stretch is any less remarkable.

Hasek's career games played numbers are low because he has 53 GP before his age 29 season, not because he played a low number of games when he was a starting goalie.



I'm not sure at all that Roy required managed RS starts, but he did mostly have managed starts throughout his career with the exception of 1992-1995 and 1998 (65 GP in an Olympic year). Roy finished in the top 6 in minutes played only three times in his career, once during a lockout-shortened season and once when had 54 GP.

Dominik Hasek, notably, did not have managed starts at all during his prime, which makes sense because his teams needed all the help they could get. He did miss time due to injury, but nobody plans injuries. When Hasek was available to play, he played in 89% of his team's games from when he took over the starting job in 1994 to his first retirement in 2002. For the sake of comparison, in Martin Brodeur's best nine season stretch he played in 90% of his team's games. If you take the best nine seasons from Patrick Roy's entire career, he played in 79% of his team's games. I think it was absolutely smart of Roy's teams to manage his starts, keeping him healthy and rested for the playoffs, but not every goalie had that luxury.

Further evidence that you are ignoring recently released archival data.

The issue with Hasek and Roy is not the total number of games played but the nature of the games they did not play.

Key is the "Days of Rest" between games. Toughest sequence for a goalie playing games on consecutive days/nights during the RS. Playoff scheduling precludes such problems.

0 Days of Rest (= career GAA)

Martin Brodeur. 180 such starts, 93-70-9/7. 2.28 GAA. (2.24)

Dominik Hasek. 103 such starts, 44-45-9/0. 2.44 GAA.(2.20)

Patrick Roy. 91 such starts, 43-31-14, 2.76 GAA.(2.54)

Brodeur was only 0.04 GAA below his career level. Given that his main back-ups ranged between 2.75 and 3.07, this provided a major team advantage that neither Roy or Hasek offered. This is before looking at older or one-goalie system goalies.

 

[Canadiens1958]

I count six.

Grant Fuhr played the majority in the 1993 victory over Boston. Steve Shields had to be their playoff goalie in 1997 and took the series over Ottawa. Then two series in 1998, three series in 1999, and one series in 2001.

He was within the franchise radius restriction for televised games.

 

[tony d]

Will vote soon but here are my thoughts on the players up this week for ranking:

Bobby Hull- Still the best left winger ever, IMO. Also the best goal scorer ever. He should be high up in the Vote 2 rankings.
Dominik Hasek- The 2nd best goalie ever, the work he did with those Sabres teams in the 90s was awesome.
Doug Harvey- 2nd best defenseman ever, should be the highest ranking defenseman on this list.
Eddie Shore- Not as big on him as others are. Guy had a very good career but for some reason I'm not that big on him compared to others on this list.
Howie Morenz- It'll be interesting to see the ranking of Morenz, Crosby and Beliveau, which guy is going to be ranked 1st?
Jean Beliveau- My #1 on this list. When you ask me who my top 5 players ever are it's: Gretzky, Orr, Lemieux, Howe and Belieau
Maurice Richard- The Rocket and one of the first true goal scorers in the NHL, will he go before Harvey or after I wonder?
Patrick Roy- Best goalie ever, Hasek certainly had his moments throughout his career but when it comes to goaltending noone did it better than Patrick Roy.
Ray Bourque- As time has gone on Bourque has seemingly settled into the 3rd best defenseman ever. Guy really could do it all out thre.
Sidney Crosby- I really like him as a player but I don't see him as a top 5 or top 10 player of all time yet, I think he'll end up finishing closer to 14 on this list than he will 5.

 

[MXD]

I'm not sure at all that Roy required managed RS starts, but he did mostly have managed starts throughout his career with the exception of 1992-1995 and 1998 (65 GP in an Olympic year). Roy finished in the top 6 in minutes played only three times in his career, once during a lockout-shortened season and once when had 54 GP.

Roy finished in the Top-8 for minutes played 8 times, which is actually more than Hasek.

 

[MXD]

I don't get the "Roy's playoffs" argument in the Roy vs. Hasek debate.

"So, Joe Thornton is a better regular season player, Justin Williams is a better playoff performer, therefore they both have a valid claim for being a better player"?

You have a 82 GP/year sample size which is comparable across teams and players and statistically robust as much as NHL data can be.

And then you have a crapshoot sample of a few games each year, which is generally unpredictable (see for instance the Art Ross or regular season PPG consistency versus the wild inconsistency of playoff scoring races and playoff PPG). All interpretations can only be made post-hoc, through heroic narratives and fandom-produced myths of greatness and "clutchness" (and if you dare speak otherwise then "hockey is about winning" is a reply which beats everything).

Yet, somehow these two sample are treated as equal when it comes to possible inference. Makes no sense to me.

... The thing with arguments like this one is, it's so nonsensical that it comes up bolstering the player argued AGAINST.

Unless your whole point was to make Sentinel's argument about H2H look incredibly smart by comparison.

And the Williams vs. Thornton applied to Roy vs. Hasek sounds a lot like : "Dave Brown has more career PIMs than Mario Lemieux, and Lemieux is a better player, ergo, PIMs are bad; thus, Bobby Hull is a better player than Maurice Richard, because he has less PIMs".

Hull might, or not, be a better player than Richard. It's arguable. PIMs are a terrible angle to judge this.

 

[quoipourquoi]

Six Vezina Trophies (twice as many as Roy and the most since the Vezina actually went to the "best" goalie).

Worth a note: two of Hasek’s Vezina Trophies came with just 8 out of 27 1st-place vote and 9 out of 30 1st-place votes - so less than one-third of the voters thought he should win.

They kinda stick out relative to Hasek and Roy’s other Vezina years where Roy took 71.4%, 71.4%, and 77.3% of the 1st-place votes and Hasek took 57.7%, 65.4%, 84.6%, and 92.3% of the 1st-place votes. Even Roy’s 2001-02 where he was up against Theodore saw him take 40.0% of the available 1st-place votes.

Of course, the media didn’t respond as negatively to Hasek’s month-long absence as the GMs did, but I think people have a habit of remembering the award but not always what went into it. If Hasek was up against a 1991 Ed Belfour or a 2002 Jose Theodore in those years when he received less-than-1/3rd of the top votes instead of 1999 Curtis Joseph and 2001 Roman Cechmanek, the 6-3 Vezina disparity (that is a 6-0 vs. 4-2 All-Star gap) probably tightens up without changing anything that Hasek himself did or did not do.

 

[ContrarianGoaltender]

Further evidence that you are ignoring recently released archival data.

The issue with Hasek and Roy is not the total number of games played but the nature of the games they did not play.

Key is the "Days of Rest" between games. Toughest sequence for a goalie playing games on consecutive days/nights during the RS. Playoff scheduling precludes such problems.

0 Days of Rest (= career GAA)

Martin Brodeur. 180 such starts, 93-70-9/7. 2.28 GAA. (2.24)

Dominik Hasek. 103 such starts, 44-45-9/0. 2.44 GAA.(2.20)

Patrick Roy. 91 such starts, 43-31-14, 2.76 GAA.(2.54)

Brodeur was only 0.04 GAA below his career level. Given that his main back-ups ranged between 2.75 and 3.07, this provided a major team advantage that neither Roy or Hasek offered. This is before looking at older or one-goalie system goalies.

I'm not ignoring it at all, you just showed exactly why I was correct that Hasek and Roy are not the same in terms of managed starts! Hasek actually has the same % of career GP that come from back-to-back starts as Brodeur:

Brodeur: 180 B2B starts in 1266 career GP, 14%
Hasek: 103 B2B starts in 735 career GP, 14%
Roy: 91 B2B starts in 1029 career GP, 9%

Secondly, GAA doesn't tell us which goalie is better, it merely tells us which goalie and team combination is better. Also, whether goalies are playing at home or on the road matters a lot for their GAAs.

But even assuming that Brodeur is better in back-to-backs, that doesn't make up for the fact that he is significantly worse in games involving all other kinds of rest. You seem to have a tendency to fixate on certain stats (zero days rest, first goal, etc.). Those things are important, but they are only a portion of the entire total, I'm not sure making ranking decisions based on a single microstat is advisable.

 

[24 others]

Say, WHAT?!?!

Patrick Roy played over 240 post-season games. It's scarcely a "few-game-crapshoot-sample." He won three Smythes- pretty much earned, as part of that three-plus seasons quantity-equivalent of game samples. He did not achieve this by happenstance. Happenstance. Three times... Now, how does that Ian Fleming quote go???

"Once may be happenstance; twice, coincidence- but three times is Enemy Action."

240 or 300, the comparability potential of PO games vs. RS games is vastly different.

For any statistical purposes RS is simply much better population than PO.

And since we are talking about Roy and Hasek, the "players don't giver their best in RS" argument does not apply. They were both psychos who would kill a man to win a meaningless RS game.

 

[Nick Hansen]

Brodeur's durability and longevity always stuns me. Incredible. I'd have him behind a handful of goalies, I think, but it'll be interesting to see his argument against some other all-time greats. He's very hard to pinpoint where he'll end up on this list.

 

[quoipourquoi]

Of course, one could say the epitome of "quitting on your team" was this...

Not to take the wind out of the sails, but Montreal tried to trade him a few weeks earlier than this and he won the Stanley Cup that year with a .974 in the Finals.