HOH Top 60 Goaltenders of All Time (2024 Edition) - Preliminary Discussion Thread

overpass · Aug 20, 2024

Postseason All-Star Teams Voting Shares (goalie)

When we're speaking about post-season all-star teams for goalies, we have most or all of the actual votes for all seasons. We don't have to treat all first team selections equally or all second team selections equally.

I've calculated vote shares for goalie all-star voting running through 1949-50. This season serves as a clear breaking point for eras. None of Bill Durnan, Frank Brimsek, or Turk Broda were listed in all star voting after 1949-50. And Terry Sawchuk began dominating the all-star voting in the 1950-51 season.

Vote share=player's votes for the season / the maximum available voting points available for a player. For example,
- In 1934-35, Lorne Chabot was the first team all-star goalie and received 34 of 99 possible points. His vote share for that season is 0.34.
- In 1934-35, Tiny Thompson was the second team all star goalie and received 29 of 99 possible points. His vote share for that season is 0.29.
- In 1935-36, Tiny Thompson was the first team all star goalie and received 90 of 93 possible points. His vote share for that season is 0.97.
- In 1935-36, Wilf Cude was the second team all star goalie and received 18 of 93 possible points. His vote share for that season is 0.19.

All star voting shares from 1930-31 through 1949-50

Goalie	Seasons w/ votes	AS Shares
Bill Durnan	7	5.99
Frank Brimsek	8	4.39
Tiny Thompson	10	3.23
Turk Broda	8	2.91
Charlie Gardiner	4	2.53
Dave Kerr	8	1.93
Chuck Rayner	4	1.54
Roy Worters	6	1.05
Johnny Mowers	2	0.84
John Ross Roach	3	0.83
Wilf Cude	5	0.72
Harry Lumley	4	0.70
Normie Smith	2	0.47
George Hainsworth	5	0.46
Lorne Chabot	5	0.45
Earl Robertson	3	0.42
Mike Karakas	4	0.40
Paul Bibeault	1	0.23
Alec Connell	2	0.20

Bill Durnan is the leader in AS Shares pre-Sawchuk. Brimsek, Thompson, Broda, and Gardiner fill out the top 5.

Although Brimsek, Thompson, Broda, and Kerr all received votes in more seasons, Durnan was the landslide voting winner for all six of his first-team all-star spots. 6 of the 11 goalie seasons receiving more than 80% of the available votes belonged to Durnan.

Highest single season AS Shares, 1930-31 through 1949-50

Season	Goalie	Points	Max	Share
1943-44	Bill Durnan	756	756	1.00
1944-45	Bill Durnan	378	378	1.00
1948-49	Bill Durnan	25	25	1.00
1935-36	Tiny Thompson	90	93	0.97
1945-46	Bill Durnan	47	51	0.92
1939-40	Dave Kerr	77	84	0.92
1941-42	Frank Brimsek	78	90	0.87
1946-47	Bill Durnan	13	15	0.87
1933-34	Charlie Gardiner	89	105	0.85
1949-50	Bill Durnan	21	25	0.84

Next, I'll add in the four seasons when coaches voted for unofficial all-star positions, from 1926-27 through 1928-29.

While we don't have actual point totals for 1926-27, I've estimated point totals based on a description of the votes (Hainsworth and Roach tied for first, Hainsworth had more first place votes, Worters finished third, Benedict and Chabot also received votes.)

All star voting shares from 1926-27 through 1949-50

Goalie	Seasons w/ votes	AS Shares
Bill Durnan	5.99	7
Frank Brimsek	4.39	8
Tiny Thompson	4.22	12
Roy Worters	3.05	10
Turk Broda	2.91	8
Charlie Gardiner	2.53	4
Dave Kerr	1.93	8
John Ross Roach	1.63	5
George Hainsworth	1.58	8
Chuck Rayner	1.54	4
Johnny Mowers	0.84	2
Wilf Cude	0.72	5
Harry Lumley	0.70	4
Lorne Chabot	0.55	6
Normie Smith	0.47	2
Earl Robertson	0.42	3
Alec Connell	0.40	3
Mike Karakas	0.40	4
Paul Bibeault	0.23	1

After adding these four post-consolidation 1920s seasons in, Tiny Thompson has nearly caught Frank Brimsek for second. Roy Worters moved up past Broda, Gardiner, Kerr, and Rayner to place fourth in voting shares, and second in seasons receiving votes. John Ross Roach and George Hainsworth both moved into the top 10 for this time period.

What if we replace the writers' voting with the coaches' voting? The table above includes the writers' voting from 1930-31 through 1945-46. I'll include all voting totals we have from the coaches. And for any goalies who we know received votes from the coaches but not how many, I'll include their vote share from the writers.

So it's 90-95% coaches voting, with a handful of writers votes filling in the gaps (including two full seasons in 1930-31 and 1931-32 where we have no votes from the coaches).

Coaches' all-star voting from 1926-27 through 1949-50

Goalie	Seasons w/ votes	AS Shares
Bill Durnan	7	5.68
Frank Brimsek	8	4.89
Tiny Thompson	11	4.76
Roy Worters	9	3.61
Turk Broda	6	3.11
Charlie Gardiner	4	2.96
Dave Kerr	4	1.58
Chuck Rayner	4	1.54
Wilf Cude	6	1.51
George Hainsworth	7	1.44
Harry Lumley	4	1.12
John Ross Roach	4	0.94
Earl Robertson	2	0.93
Johnny Mowers	2	0.82
Lorne Chabot	5	0.51
Normie Smith	1	0.42
Alec Connell	4	0.40
Mike Karakas	4	0.40

The coaches' voting results are pretty similar to the writers, overall. No changes to the order of the top 7.

Brimsek and Thompson are still in a dead heat for #2, and are closer to #1 Durnan than in the writers' results.

Per the coaches results, #6 Charlie Gardiner has more separation from the #7-10 finishers than he does in the writers results, finishing close behind #5 Broda.

Wilf Cude is in the top 10 per the coaches, and the writers have John Ross Roach. Otherwise it's the same 9 names (Durnan, Brimsek, Thompson, Worters, Broda, Gardiner, Kerr, Hainsworth, Rayner).

Chuck Rayner is the only goaltender listed here for whom we have all-star votes after 1949-50. He was the second team all-star goalie in 1950-51, receiving 40 of 90 points (0.44 share). If we add his 1950-51 share to his pre-1950 totals, he sits at 1.98 shares, ahead of Dave Kerr for #7 on both the writers' and coaches' list.

NHL goalie leaders from 1926-27 through 1949-50

Lorne Chabot, Alec Connell, Mike Karakas, and Bill Beveridge were goalies with significant NHL careers in this time who received relatively little support in all-star voting.

While Chabot was voted the first-team goalie for 1934-35 by the writers, the votes were split among several goalies.

Season	Goalie	Points	Max	Share	Rank
1934-35	Lorne Chabot	34	99	0.34	1
1934-35	Tiny Thompson	29	99	0.29	2
1934-35	Roy Worters	25	99	0.25	3
1934-35	Alec Connell	19	99	0.19	4
1934-35	George Hainsworth	14	99	0.14	5
1934-35	Dave Kerr	7	99	0.07	6

Coaches picked Tiny Thompson (14 votes of max 27) and Roy Worters (11 votes of max 27) over Chabot.

Chabot's 1934-35 is the lowest vote share of any official first team all star from 1930-31 through 1949-50. 16 second or third place seasons have higher vote shares.

Season	Goalie	Points	Max	Share	Rank
1934-35	Lorne Chabot	34	99	0.34	1
1936-37	Normie Smith	32	69	0.46	1
1938-39	Frank Brimsek	54	102	0.53	1
1930-31	Charlie Gardiner	61	111	0.55	1
1940-41	Turk Broda	52	87	0.60	1

Michael Farkas · Aug 20, 2024

All time GAA leader Alec Connell doesn't fair very well here. That's in line with my research of him 10 or so years ago. Connell is missing one season of note here, but nothing more probably. That said...

I found an article in the Brooklyn Daily Eagle - April 4, 1926 by George Currie "(Former Canadian Hockey Authority)"

Eagle's All-National League Hockey Teams
Position: 1st team | 2nd team
Goal: Connell (Ott) | Benedict (Mtm)
RD: Clancy (Ott) | Conacher (Pit)
LD: Cleghorn (Bos) | Langlois (NY)
C: Nighbor (Ott) | Milks (Pit)
RW: Morenz (Mtl) | Siebert (Mtl)
LW: Stewart (Mtl) | H.Smith (Ott)

Spares: Dye (Tor), Denenny (Ott), McCurry (Pit), Herberts (Bos), Burch (NY), Noble (Mtm), Worters (Pit)

overpass · Aug 20, 2024

Postseason All-Star Teams Voting Shares for 1950-51 through 1970-71
These seasons are selected to capture the careers of Terry Sawchuk, Glenn Hall, and Jacques Plante.

Many of the seasons in the 1950s only include AS voting points for the top 2 or 3 finishers, so these shares are not complete.

1950-51 through 1970-71 AS voting

Goalie	Seasons w/ votes	AS Shares
Glenn Hall	14	7.09
Terry Sawchuk	15	5.83
Jacques Plante	12	5.50
Ed Giacomin	5	3.06
Johnny Bower	8	1.77
Harry Lumley	2	1.55
Gump Worsley	9	1.54
Tony Esposito	2	1.42
Roger Crozier	2	1.31
Charlie Hodge	5	1.20
Al Rollins	3	0.79
Gerry McNeil	2	0.71

For the 1953-54 season through the 1967-68 season, awards voting was conducted in two parts. One vote at midseason, and one vote at the end of the season. The awards were based on the sum of first half voting and end of season voting.

This awards voting scheme is unusual and may have double counted first half results, depending on whether voters considered the full season in their end of season voting, or just the second half.

It's possible that excluding the first half voting entirely might better reflect the season, if voters considered the full season when voting after the season.

Here are the first half and second half award shares for the 1953-54 through 1967-68 seasons.

Goalie	AS Shares	1st half AS Shares	2nd half AS Shares
Glenn Hall	6.45	5.88	7.02
Jacques Plante	4.39	4.12	4.66
Terry Sawchuk	3.28	4.57	1.99
Johnny Bower	1.77	2.28	1.27
Harry Lumley	1.55	1.52	1.58
Gump Worsley	1.40	1.54	1.26
Roger Crozier	1.31	1.34	1.28
Charlie Hodge	1.20	1.49	0.90
Ed Giacomin	1.17	1.08	1.26
Denis DeJordy	0.43	0.25	0.61
Al Rollins	0.39	0.79	0.00

The most significant result is that Terry Sawchuk fared far better in first half voting than in end of season voting. He frequently got off to a strong start and then tailed off toward the end.

Glenn Hall and Jacques Plante both performed better in the end of season voting.

Finally, if we accept the conjecture that it's better to exclude first half voting, here are the results from 1950-51 through 1970-71, using end of season votes only for the 1953-54 through 1967-68 seasons.

1950-51 through 1970-71 AS voting, excluding first half voting

Goalie	Seasons w/ votes	AS Shares
Glenn Hall	14	7.66
Jacques Plante	12	5.77
Terry Sawchuk	15	4.53
Ed Giacomin	5	3.14
Harry Lumley	2	1.58
Tony Esposito	2	1.42
Gump Worsley	9	1.39
Roger Crozier	2	1.28
Johnny Bower	8	1.27
Charlie Hodge	5	0.90
Gerry McNeil	2	0.71
Al Rollins	3	0.40

ChiTownPhilly · Aug 20, 2024

Prominent Hall of Fame Goaltenders who lost significant time due to World War II

Frank Brimsek
Seasons lost: 1943-44, 1944-45, part of 1945-46.
Playing age during lost seasons- 28-29-(30)

Turk Broda
Seasons lost: 1943-44, 1944-45, most of 1945-46
Playing age during lost seasons- 29-30-(31)

I thought these two had the greatest time-loss, but then there's also...

Chuck Rayner
Seasons lost: 1942-43, 1943-44, and (all of) 1944-45
Playing age during lost seasons- 22-23-24

[Non-HoF Goalie Dave Kerr is a special case. If one looked at his stat-line, one might speculate that he served and never returned to the league- but he actually exited Pro Hockey in a contract dispute.]

So during the WWII years of 1943-44 and 1944-45....
The Boston Bruins HoF starting Goaltender was in the service
The Toronto Maple Leafs HoF starting Goaltender was in the service
The former Brooklyn Americans and future New York Rangers HoF starting Goaltender was in the service
Even the Detroit Red Wings non-HoF starting Goaltender {Johnny Mowers} was in the service
The former New York Rangers non-HoF starting Goaltender [Kerr] was out of the league over a year, and the non-HoF Goalie who replaced him [Jim Henry] just turned drinking-age before the start of the 1941-42 season, and campaigned on for that entire earlier playing year, until he too changed out of his Pro Hockey uniform and into his Army-issue garments, joining his colleagues in the military.

Vilica · Aug 20, 2024

Those unanimous seasons by Durnan are a bit of an anomaly. They're legitimate seasons, but look at his competition for best goalie in both years.

43-44

44-45

The 43-44 season is most egregious, as McAuley gave up nearly triple the amount of goals in the same amount of games, while Gardiner gave up nearly double in 9 less games. Bibeault finished 2nd basically by default, despite not even playing 30 games. It's the same the next year, as Karakas finished 2nd despite giving up nearly 4 goals a game. The only team that wasn't decimated by WW2 enlistments carried their goals against average through the war, while the other teams ballooned.

The 50-51 through 70-71 deep dive is a good look at those goalies, but again I think it misses one important point - being 3rd/4th best in a 6 team league is closer to league average than it is to top of the league. If we take say 30 games as the minimum games played for a season, you end up with 184 player-seasons [if you choose 35, 160 player-seasons, 40 games, 134 player-seasons - I'm just not sure which cutoff is the best one]. Count up the seasons from those 12 goalies, you end up with 79 player-seasons. So at the low end, they make up 43% of the seasons, in the middle 50% of the seasons, at the high end 59% of the seasons. They can't all be top 10% seasons, or even top 20% ones. It isn't like now, where the 3 Vezina finalists represent the best of the ~30 starter seasons, or top 10%.

I set my cutoffs at 66-67, rather than 70-71, but the pool of pre-expansion seasons just isn't that big. There are 361 player-seasons of 20 games or more, 268 of 35 games or more, and 108 of 50 or more. The 10 goalies Broda Brimsek Rayner Durnan Lumley Sawchuk Plante Worsley Hall and Bower represent 75 of those 108 seasons, all of them having at least 5. It is a small enough pool that you could probably rank all 108 of them. The ones outside the top 30 just aren't that significant in an all-time sense. Somebody posted earlier a bunch of lists, including most times finishing top 10 in save percentage, and Hall Plante Sawchuk and Worsley made up 4 of the top 5, and it's like well yeah, the goalies that played in the save percentage era in a 6 team league are going to finish in the top 10 of save percentage.

overpass · Aug 20, 2024

Vilica said:
The 50-51 through 70-71 deep dive is a good look at those goalies, but again I think it misses one important point - being 3rd/4th best in a 6 team league is closer to league average than it is to top of the league.

Why is this an important point? Can you spell it out?

Let's take the 1959-60 season. 82% of the minutes were played by HHOF goaltenders Hall, Plante, Bower, Sawchuk, Lumley, and Worsley. 47% of the minutes were played by top 10 goalies of all time in Hall, Plante, and Sawchuk.

1959-60 NHL Goalie Statistics | Hockey-Reference.com

Check out the Goalie Statistics for the 1959-60 NHL season on Hockey-Reference.com

www.hockey-reference.com

It's a pretty amazing league when the median goalie is Terry Sawchuk or Johnny Bower! In fact, it says more about the league than it says about Sawchuk or Bower.

Vilica said:
The 10 goalies Broda Brimsek Rayner Durnan Lumley Sawchuk Plante Worsley Hall and Bower represent 75 of those 108 seasons, all of them having at least 5.

Again, surely the fact that ten Hall of Fame goaltenders make up 75 of 108 full seasons says a lot about the strength of the league.

Vilica said:
If we take say 30 games as the minimum games played for a season, you end up with 184 player-seasons [if you choose 35, 160 player-seasons, 40 games, 134 player-seasons - I'm just not sure which cutoff is the best one]. Count up the seasons from those 12 goalies, you end up with 79 player-seasons. So at the low end, they make up 43% of the seasons, in the middle 50% of the seasons, at the high end 59% of the seasons. They can't all be top 10% seasons, or even top 20% ones. It isn't like now, where the 3 Vezina finalists represent the best of the ~30 starter seasons, or top 10%.

Are top 20% seasons a meaningful metric that can be compared across different leagues?

Let's take the 1979-80 season and compare it to the already-mentioned 59-60 season. In 1979-80 your 80th percentile goalie is Pete Peeters or Gilles Meloche or someone like that. In 1959-60 it's Jacques Plante or Glenn Hall.

1979-80 NHL Goalie Statistics | Hockey-Reference.com

Check out the Goalie Statistics for the 1979-80 NHL season on Hockey-Reference.com

www.hockey-reference.com

When you get a result like that, don't you have to question the value of the percentile, or the value of the league average, rather than questioning the value of the goaltender?

MadLuke · Aug 20, 2024

Vilica said:
the goalies that played in the save percentage era in a 6 team league

And could all (like goaltender during an Olympic tourney or MLB shortstops defensively) be very good at the same time i.e. the median goaltender here can be a top 3 in the world, making how good versus their peers talk a bit hard to do and for a part of it; no too almost no video.... Contemporary testimony can become a lot of what people rely on.

Think of a tournament like the 2016 world cup, Halak-Price-Bobrovksy-Lundqvist-Rask-Quick-Rinne-Markstrom.... what would being just average in a 50 games tournament with those teams would mean for someone in that group, would finishing in the bottom 2 mean you are bad ?

Michael Farkas · Aug 20, 2024

While I'm dabbling on the scouting time machine...if anyone has any goalies that you thought might have looked good but the team wasn't or the numbers don't jive with what you saw...feel free to drop that name in here, I'm happy to check him or them out as time allows...

rmartin65 · Aug 20, 2024

Michael Farkas said:
While I'm dabbling on the scouting time machine...if anyone has any goalies that you thought might have looked good but the team wasn't or the numbers don't jive with what you saw...feel free to drop that name in here, I'm happy to check him or them out as time allows...

I’m always curious about the dynasty players. Like, yeah, good teams have good players, but are all of those players as good as we make them out to be?

I realize that’s a broad question, I can try to provide names of the guys I’m most curious about over the next couple days.

jigglysquishy · Aug 20, 2024

Michael Farkas said:
While I'm dabbling on the scouting time machine...if anyone has any goalies that you thought might have looked good but the team wasn't or the numbers don't jive with what you saw...feel free to drop that name in here, I'm happy to check him or them out as time allows...

I think Kiprusoff was a top 3 goalie from ~2003-2008. He got Vezina recognition, but he played on a weak defensive team and his stats were always suppressed.

Give me Dryden. He's the hardest top goalie to figure.

Give me Tretiak and Holocek.

I would love a real Sawchuk/Plante/Hall scouting comparison. They will all end up top 10 and likely 2 of them top 5. They're modern enough that we should be able to pull actual tape. I think this is probably the single most valuable thing we could pull.

Michael Farkas · Aug 20, 2024

rmartin65 said:
I’m always curious about the dynasty players. Like, yeah, good teams have good players, but are all of those players as good as we make them out to be?

I realize that’s a broad question, I can try to provide names of the guys I’m most curious about over the next couple days.

I'm going to take a long look at Dryden and Fuhr. I did a longer look at Billy Smith in the previous iteration of this and came away surprisingly impressed (as the Islanders are not only a dynasty, but a defensively responsible one...it was a perfect storm, but I remember liking him, but I'll check back there too).

To answer your question though -- no, certainly not.

And with goalies in particular, I'm thinking it's a lot harder to stand out statistically in a poor situation. And a lot easier to be insulated by a great situation than it is for forwards or defensemen...we have "bad team scorer", we have guys like Erik Karlsson in San Jose who can just be turned loose to do whatever they want and score 100 points on a team that scored 101 goals all year...but really, if you're a team that can't clear rebounds or gives up even 2, 3, 4 extra 'A' scoring chances a game, your goalie is toast.

The toughest thing for me is just having the time to do a deep dive on enough guys over the....entire...history...of the...................game. *fake cough*

So, I'm going to cut a corner or three somewhere...

Michael Farkas · Aug 20, 2024

jigglysquishy said:
I think Kiprusoff was a top 3 goalie from ~2003-2008. He got Vezina recognition, but he played on a weak defensive team and his stats were always suppressed.

He almost certainly was.

jigglysquishy said:
Give me Dryden. He's the hardest top goalie to figure.

Top on my list to figure out, I'm working backwards to make sure I carry the (de-)evolution of the position as best I can, while picking up (or cutting) goalies along the way. Dryden is my #1 "target".

jigglysquishy said:
Give me Tretiak and Holocek.

These are top 10 targets as well. I think Tretiak is going to hold up well, I'm fairly familiar with his game already.

jigglysquishy said:
I would love a real Sawchuk/Plante/Hall scouting comparison. They will all end up top 10 and likely 2 of them top 5. They're modern enough that we should be able to pull actual tape. I think this is probably the single most valuable thing we could pull.

Yeah, that will be very interesting. Camera work gets a little tough, replays become fewer and farther (further?) between as we go back...it starts to become difficult. But I'm on the case. Your initial thoughts here are quite similar to mine.

overpass · Aug 20, 2024

I'll just drop one thing about Mike Liut here, because he's one of the names that always comes up for the 1980s.

Over the past 10 years, every NHL team has realized the analytics say you shouldn't start goalies in back to back games. It's just not done anymore.

And yet, when I take a look at nhl.com's Days of Rest report for goalies, most goalies in history have done just fine with 0 days of rest. Maybe the penalty for playing on 0 days of rest is a modern pro-fly thing? Anyway I'm not here to figure out the modern game. I'll just say that two goalies in history stood out to me who had difficulty with 0 days of rest, both from the 1980s.

One of them, Greg Millen, may not be mentioned again in this project. The other is Mike Liut.

From 1979-80 through 1987-88, Liut carried a heavier workload than any other goaltender, averaging 59 GP per season at a time when most teams split the work more evenly. During these years, he had an 0.884 SV% overall...and an 0.867 SV% in 81 games played with 0 days of rest.

So Liut would probably have been better off with fewer regular season GP and better management of his workload. But that shouldn't have been an issue in the playoffs, right? There aren't so many back to back games in the playoffs.

Apparently there are. Mike Liut played more playoff games on 0 days of rest than any other goalie in NHL history. 19 of his 67 career playoff games, in fact 19 of his 56 playoff games during his 1979-80 through 1987-88 prime, or 34% of his playoff games in his prime. His SV% in those games was 0.879, which isn't terrible, but a little quick math says his SV% in the other 37 playoff games was around 0.897. Which is pretty good for the 80s! Right around Grant Fuhr.

(It's Mike Liut, so...the 1981 Canada Cup final game was on 1 day of rest.)

jigglysquishy · Aug 20, 2024

Anyone want to make an argument for anyone other than Hasek or Roy at #1? They've owned that spot on our lists. Does anyone think that's wrong?

Michael Farkas · Aug 20, 2024

jigglysquishy said:
Anyone want to make an argument for anyone other than Hasek or Roy at #1? They've owned that spot on our lists. Does anyone think that's wrong?

I think it might be. I'm not convinced I'll get there because it's so tertiary to my cause, plus it's virtually "un-winnable", for lack of a better term.

But will anyone throw out my list if, say, Bill Ranford was #1? You know I'm genuine in my beliefs, if nothing else.

(Obviously not serious, I was just watching the 1990 Stanley Cup Final while working out this afternoon - Ranford was just so lights out in the series...)

tony d · Aug 20, 2024

Craig Anderson obviously won't get #1 but was he good enough to sneak in to the top 60? Could be my Ottawa fan days but he had some good numbers for the Senators and is being given some consideration for my top 80.

Michael Farkas · Aug 20, 2024

tony d said:
Craig Anderson obviously won't get #1 but was he good enough to sneak in to the top 60? Could be my Ottawa fan days but he had some good numbers for the Senators and is being given some consideration for my top 80.

I don't think that's likely...he ran really hot and cold.

I actually referenced Anderson specifically in my goalie evolution video here:

He was absolutely unbeatable in games...somehow. And then would come back and lay an egg the next time out.

The one legit season where he got any Vezina consideration he was 3rd in the NHL in shutouts. He also gave up 4+ in 17 games. More than all the other goalies tied or ahead of him in shutouts that year by at least three games.

Hell, in that year's playoffs - game 2 vs. San Jose, gives up 6...game 3 -- 51 save shutout.

Professor What · Aug 20, 2024

tony d said:
Craig Anderson obviously won't get #1 but was he good enough to sneak in to the top 60? Could be my Ottawa fan days but he had some good numbers for the Senators and is being given some consideration for my top 80.

With Anderson I think it depends on how much you value consistency. Good Anderson was really good, but he always seemed to have trouble stringing together multiple quality seasons. It always seemed like after he had a good season, a bad one was guaranteed. As @Michael Farkas said, there was also a game to game question where he was concerned too.

Could he make a top 80 list? Yeah, I think he could. He's just outside the top 80 list in my working rankings, but top 60? Eh. I don't know. It's not like I'd reject a list that had him top 60, but for me personally, it would take a lot of selling.

Vilica · Aug 21, 2024

overpass said:
Why is this an important point? Can you spell it out?

Let's take the 1959-60 season. 82% of the minutes were played by HHOF goaltenders Hall, Plante, Bower, Sawchuk, Lumley, and Worsley. 47% of the minutes were played by top 10 goalies of all time in Hall, Plante, and Sawchuk.

1959-60 NHL Goalie Statistics | Hockey-Reference.com

Check out the Goalie Statistics for the 1959-60 NHL season on Hockey-Reference.com

www.hockey-reference.com

It's a pretty amazing league when the median goalie is Terry Sawchuk or Johnny Bower! In fact, it says more about the league than it says about Sawchuk or Bower.

Again, surely the fact that ten Hall of Fame goaltenders make up 75 of 108 full seasons says a lot about the strength of the league.

Are top 20% seasons a meaningful metric that can be compared across different leagues?

Let's take the 1979-80 season and compare it to the already-mentioned 59-60 season. In 1979-80 your 80th percentile goalie is Pete Peeters or Gilles Meloche or someone like that. In 1959-60 it's Jacques Plante or Glenn Hall.

1979-80 NHL Goalie Statistics | Hockey-Reference.com

Check out the Goalie Statistics for the 1979-80 NHL season on Hockey-Reference.com

www.hockey-reference.com

When you get a result like that, don't you have to question the value of the percentile, or the value of the league average, rather than questioning the value of the goaltender?

Using that 59-60 season, look at the actual stats (here) - you have 10 goalies of significance, with Riggin Rollins and perhaps Paille being too small a sample. Is that Sawchuk season anything to write home about? He's basically league average that year. He's still 4th in save percentage and 3rd in GAA, because again, 6 teams. He received a bunch of 1st half votes for the all-star team because he had an great October, but it's 10 games, and he has 5 competitors. The same goes for Lumley and Worsley, neither of whom had good years. They both still finished in the top 10 in save percentage though.

Lumley and Worsley are guys I'm really skeptical of - they both had 2 good years, and not much else. I see them somewhere on the spectrum between Varlamov and Quick, to use modern equivalents. Nobody gives those two guys much credit for being the 15th best goaltender in the league in some of their more average years. Sawchuk less so, because of his early peak, but during the save percentage era, he basically spent a decade being league average. If Terry Sawchuk retired after 54-55, and Serry Tawchuk played in the NHL from 55-56 through 69-70, Serry's not making it in the Hall of Fame based on his performance during those years. In fact, Serry Tawchuk's performance in those ~600 games, depending on what years you cut off at the end due to partial playing time, is basically equivalent to Meloche or Peeters in their careers, roughly league average by GA%-, over about the same number of games.

Your premise almost becomes a tautology - the goaltending is great because all these goaltenders are Hall of Famers - Hall of Famers are obviously great goaltenders.

It just ignores the fact that a) making the Hall of Fame isn't reflective of your performance every year, and b) Hall of Famers can have bad years too. (Also c) the Hall of Fame's track record with player election isn't the greatest.)

With regards to percentiles, you'd expect in 2 separate samples of similar sizes, that seasons in each percentile would be roughly equivalent to each other. So if you have say 200 goaltender seasons, and divide them in half randomly, you'd expect that the top 10 seasons in each sample would be the same in both samples, seasons 11-20 would be the same, and so on. There's a fudge factor at the top and bottom of each percentile, but you're not going to get a result where the 50th best season in one sample is better than the 20th best season in the other.

That's basically what you're trying to say, that because all these Hall of Famers make up the majority of this 108 player-season sample, that the 50th best season is better than the 20th best season of a different 100 player-season sample, whether that be in the 80s, 90s, or today. The other aspect of that, is because we have 30+ teams now, the sample fills up much quicker. Here's a comparison just by GAA, of goaltenders playing 50+ games, the first sample being 1917-18 through 66-67, the second sample being 2005-06 through 23-24.

GAA Under	66-67	23-24
2	9	4
2.1	12	13
2.2	20	26
2.3	24	61
2.4	30	108
2.5	42	153
2.6	50	191
2.7	61	225
2.8	65	255
2.9	71	283
3	79	308
3.1	84	324
3.2	90	332
3.3	95	338
3.4	99	344
3.5	100	345
Total	108	349

There are 108 total seasons in the first sample, there are 349 total seasons in the second sample. The top end runs close together, there's 41 player-seasons 2.2 and under pre-expansion, 43 post-lockout, so you're going to end up with a near 50/50 balance if you were to rank the best 100 or so seasons, but look at how post-lockout fills in. You add another 35 player-seasons between 2.2 and 2.3, while pre-expansion needs to expand out to 2.7 before it picks up another 35 player-seasons, and then post-lockout adds another 47 seasons to match the 108 of the pre-expansion total. You rank those 216 seasons, and the last 50-70 seasons are all going to be pre-expansion. Cut it down to the top 125, and you're going to have 30-40 pre-expansion, and 85-95 post-lockout.

Also, this shows the power of controlling the sample size. Setting it at 50 excludes everything prior to WW2, including all the low scoring 20s and 30s goalies, who played 44 or 48 games. You have 108 player-seasons of 50 games, 162 of 48 games, 214 of 44 games, and 235 of 40 games. The equivalent for post-lockout is 349 player-seasons of 50, 381 of 48, 444 of 44, and 533 of 40. The GAA sample changes from 9-4 under 2, to 49-11 under 2. The low-scoring era means that of the 235 player-seasons in the pre-expansion sample, 107 were under 2.4, akin to the 108 the post-lockout sample had. However, reducing the games played to 40 there boosts it to 149 player-seasons, with 211 being under 2.5, and 270 under 2.6, while a similar 92-87 breakdown exists for player-seasons under 2.3. So without normalizing for scoring, the top 200 or so seasons breakdown 50/50ish, but go out to 500, and again the last 90 or so are all going to be pre-expansion, because all 270 of the post-lockout sample is under 2.6, while the last 90 of the pre-expansion sample are above 2.6.

No matter what you do, you're going to run out of great seasons pre-expansion before you run out of great seasons post-lockout.

[I also have some breakdowns on that initial sample of 108 post-lockout player-seasons - they cover every non-shortened year (no 12-13, 19-20, or 20-21), involve 28 of the 32 teams, and 50 different goaltenders.]

Vilica · Aug 21, 2024

I was rereading my 2nd to last paragraph about sample sizes, and it's a whole lot of numbers that an updated table would explain easier, so here it is:

GAA Under	66-67	23-24
2	49	11
2.1	62	26
2.2	76	44
2.3	92	87
2.4	107	149
2.5	127	211
2.6	145	270
2.7	159	324
2.8	168	371
2.9	181	416
3	192	453
3.1	198	480
3.2	206	497
3.3	212	507
3.4	216	519
3.5	221	526
Total	235	533

I feel like we should probably do something about normalizing pre-WW2 goalie-seasons to account for just how low-scoring that era was, because of how lopsided the samples get. But again, it ends up being like 50 goalie-seasons, because the NHL is still just 6-8 teams over a decade, and they get swallowed up eventually.

Also, here's a combined table with the teams of those initial 108 post-lockout seasons, plus the years:

ANA	ANA	ANA	ANA					05-06	4
ARI	ARI	ARI						06-07	7
BOS	BOS	BOS	BOS	BOS	BOS	BOS		07-08	9
BUF								08-09	5
CAR	CAR							09-10	6
CBJ	CBJ	CBJ	CBJ					10-11	9
CGY	CGY	CGY	CGY	CGY				11-12	9
CHI	CHI	CHI	CHI					12-13
COL								13-14	10
DAL	DAL	DAL	DAL					14-15	16
DET	DET	DET	DET	DET				15-16	10
EDM								16-17	9
FLA	FLA	FLA						17-18	4
LAK	LAK	LAK	LAK	LAK				18-19	2
MIN	MIN	MIN	MIN	MIN				19-20
MTL	MTL	MTL	MTL					20-21
NJD	NJD	NJD	NJD	NJD				21-22	4
NSH	NSH	NSH	NSH	NSH				22-23	2
NYI	NYI							23-24	2
NYR	NYR	NYR	NYR	NYR	NYR	NYR	NYR
PHL								MTL/WSH
PIT	PIT	PIT	PIT	PIT				ARI/MIN
SJS	SJS	SJS	SJS	SJS	SJS			LAK/TBL
STL								STL/WSH
TBL	TBL	TBL	TBL					VAN/FLA
VAN	VAN	VAN	VAN
WPG	WPG	WPG
WSH	WSH	WSH

If you count the teams, it's actually 110, because both Colorado and St. Louis had seasons that just missed - Varlamov's 13-14 was a 2.41 for Colorado, and the same for Legace's 07-08 for St. Louis. I added them to the teams to get to 28 uniques instead of 26, but did not add them to the years. The teams that missed were Seattle and Vegas as expansion teams, and Ottawa and Toronto as the other teams. Here's the quick summation of their misses, by games played and by GAA.

Team	Year	Games	GAA	Sv%	Goalie
VGK	20-21	36	1.98	0.928	Fleury
	17-18	46	2.24	0.927	Fleury
	18-19	61	2.51	0.913	Fleury
SEA	23-24	50	2.46	0.916	Daccord
TOR	22-23	42	2.33	0.919	Samsonov
	21-22	49	2.64	0.914	Campbell
	16-17	66	2.67	0.918	Andersen
OTT	05-06	43	2.09	0.925	Hasek
	06-07	58	2.47	0.918	Emery

It runs the gamut from hall of famers to one hit wonders. Ottawa also has the 24 game runs by Craig Anderson in 12-13 as well as Andrew Hammond in 14-15, where they both put up .941 save percentages. You can let the sample creep down from 50 to 40 to 20 games played to try and include more pre-expansion seasons, but there's always seasons like those lurking in the post-lockout sample.

Michael Farkas · Aug 21, 2024

No need for any of that. There's six teams, there's basically six goalies because they tried to play all the games. The goal distribution is such where just because you finished third doesn't mean you were "average" - that's a really silly application of mathematics haha

Instead trying to cut between a 2.66 and 2.69 GAA, we just need to do the work on evaluating them. Why was Sawchuk "only" diesel for a small chunk of his career? Was Plante a product of the defensive teams that he played for or would it have worked anywhere? How effective was Hall with his style - did he ever cost his team with that?

Re-arranging numbers this tight without adding anything of substance to the equation isn't likely to yield anything actionable...

Maybe if you went through the game logs and found blown leads and goals surrendered after another a goal or things like that, maybe we'd get a better picture. But season and career long averages are only gonna be able to be hammered in so many ways...

overpass · Aug 21, 2024

Vilica said:
Using that 59-60 season, look at the actual stats (here) - you have 10 goalies of significance, with Riggin Rollins and perhaps Paille being too small a sample. Is that Sawchuk season anything to write home about? He's basically league average that year. He's still 4th in save percentage and 3rd in GAA, because again, 6 teams. He received a bunch of 1st half votes for the all-star team because he had an great October, but it's 10 games, and he has 5 competitors. The same goes for Lumley and Worsley, neither of whom had good years. They both still finished in the top 10 in save percentage though.

You say Sawchuk was league average and suggest that's nothing to write home about. But league average is meaningful in a league where the average goaltender is a hall of famer in mid-career.

Worsley and Lumley both had better years, yes. But they were playing on weaker teams against some incredibly strong teams. I don't think we can say they didn't have good seasons based on looking at the stats.

What is probably more meaningful is that Lumley lost his NHL job after the season. And the Rangers brought in Jack McCartan to challenge Worsley, and Worsley fought him off to keep his job. But again, that's the nature of a six team league loaded with talent. There are only six starting goalie positions, and the goalies in the bottom half frequently had to hold off challengers for their positions.

Vilica said:
Lumley and Worsley are guys I'm really skeptical of - they both had 2 good years, and not much else. I see them somewhere on the spectrum between Varlamov and Quick, to use modern equivalents. Nobody gives those two guys much credit for being the 15th best goaltender in the league in some of their more average years. Sawchuk less so, because of his early peak, but during the save percentage era, he basically spent a decade being league average. If Terry Sawchuk retired after 54-55, and Serry Tawchuk played in the NHL from 55-56 through 69-70, Serry's not making it in the Hall of Fame based on his performance during those years. In fact, Serry Tawchuk's performance in those ~600 games, depending on what years you cut off at the end due to partial playing time, is basically equivalent to Meloche or Peeters in their careers, roughly league average by GA%-, over about the same number of games.

Again, league average in Sawchuk's league was not the same as league average in Meloche and Peeters' league.

Vilica said:
Your premise almost becomes a tautology - the goaltending is great because all these goaltenders are Hall of Famers - Hall of Famers are obviously great goaltenders.

It just ignores the fact that a) making the Hall of Fame isn't reflective of your performance every year, and b) Hall of Famers can have bad years too. (Also c) the Hall of Fame's track record with player election isn't the greatest.)

It's actually the opposite of a tautology. I'm introducing outside expert opinion to support my argument. Sawchuk, Hall, Bower, Plante, Worsley, and Lumley were all selected for the Hall of Fame between 1971 and 1980. Only Lumley had to wait significant time after he retired (1960 to 1980), the others were all selected soon after retirement.

Why were they selected? Partly for their outstanding quality of play. Partly for their outstanding quantity of NHL play. They all played in the 6 team NHL for over a decade, and every year in camp they fought off the best minor league goals in the world who were trying to take their jobs. Every year they had to maintain their quality of play or risk losing their starting job to Charlie Hodge, or Denis DeJordy. In fact, Bower and Worsley spent years competing for the Rangers starting job, leaving the loser in the minor leagues.

My argument doesn't rely on the HHOF making perfect selections. Maybe the selection of Lumley was a mistake. But Lumley was probably the 20th percentile or so goalie for that season. My point is that the average, 50th percentile, 80th percentile, 90th percentile, etc for goaltending performance in the 1959-60 season were much higher than for later post-expansion seasons. Such as, for example, 1979-80.

Vilica said:
With regards to percentiles, you'd expect in 2 separate samples of similar sizes, that seasons in each percentile would be roughly equivalent to each other.

The mean (average) and distribution of the two samples would have to be the same. It seems like you are assuming their are. I am saying there is no reason to believe they are the same, and for much of the Original Six there is reason to believe the mean was much higher.

Vilica said:
So if you have say 200 goaltender seasons, and divide them in half randomly, you'd expect that the top 10 seasons in each sample would be the same in both samples, seasons 11-20 would be the same, and so on. There's a fudge factor at the top and bottom of each percentile, but you're not going to get a result where the 50th best season in one sample is better than the 20th best season in the other.

Sure. Where does the random division come in?

Vilica said:
That's basically what you're trying to say, that because all these Hall of Famers make up the majority of this 108 player-season sample, that the 50th best season is better than the 20th best season of a different 100 player-season sample, whether that be in the 80s, 90s, or today.

Yes. Because the 108 player sample is from a heavily selected and truncated high end of the talent distribution. The different 100 season post-expansion sample includes goaltenders who would have been in the minor leagues in 1960 and would not have been in the 108 player sample.

Vilica said:
The other aspect of that, is because we have 30+ teams now, the sample fills up much quicker. Here's a comparison just by GAA, of goaltenders playing 50+ games, the first sample being 1917-18 through 66-67, the second sample being 2005-06 through 23-24.

Vilica said:
GAA Under 66-67 23-24
2 9 4
2.1 12 13
2.2 20 26
2.3 24 61
2.4 30 108
2.5 42 153
2.6 50 191
2.7 61 225
2.8 65 255
2.9 71 283
3 79 308
3.1 84 324
3.2 90 332
3.3 95 338
3.4 99 344
3.5 100 345
Total 108 349

There are 108 total seasons in the first sample, there are 349 total seasons in the second sample. The top end runs close together, there's 41 player-seasons 2.2 and under pre-expansion, 43 post-lockout, so you're going to end up with a near 50/50 balance if you were to rank the best 100 or so seasons, but look at how post-lockout fills in. You add another 35 player-seasons between 2.2 and 2.3, while pre-expansion needs to expand out to 2.7 before it picks up another 35 player-seasons, and then post-lockout adds another 47 seasons to match the 108 of the pre-expansion total. You rank those 216 seasons, and the last 50-70 seasons are all going to be pre-expansion. Cut it down to the top 125, and you're going to have 30-40 pre-expansion, and 85-95 post-lockout.

Also, this shows the power of controlling the sample size. Setting it at 50 excludes everything prior to WW2, including all the low scoring 20s and 30s goalies, who played 44 or 48 games. You have 108 player-seasons of 50 games, 162 of 48 games, 214 of 44 games, and 235 of 40 games. The equivalent for post-lockout is 349 player-seasons of 50, 381 of 48, 444 of 44, and 533 of 40. The GAA sample changes from 9-4 under 2, to 49-11 under 2. The low-scoring era means that of the 235 player-seasons in the pre-expansion sample, 107 were under 2.4, akin to the 108 the post-lockout sample had. However, reducing the games played to 40 there boosts it to 149 player-seasons, with 211 being under 2.5, and 270 under 2.6, while a similar 92-87 breakdown exists for player-seasons under 2.3. So without normalizing for scoring, the top 200 or so seasons breakdown 50/50ish, but go out to 500, and again the last 90 or so are all going to be pre-expansion, because all 270 of the post-lockout sample is under 2.6, while the last 90 of the pre-expansion sample are above 2.6.

No matter what you do, you're going to run out of great seasons pre-expansion before you run out of great seasons post-lockout.

[I also have some breakdowns on that initial sample of 108 post-lockout player-seasons - they cover every non-shortened year (no 12-13, 19-20, or 20-21), involve 28 of the 32 teams, and 50 different goaltenders.]

Not sure I understand what these paragraphs to do with ranking goaltenders.

My core point is that you're talking about how few player seasons there were in the six team NHL, but you're failing to recognize that these were the best goalies in the world who were selected to play in NHL games through a highly competitive process from the beginning of training camp to the end of the playoffs. Not a 100% efficient or accurate process, as we see from Johnny Bower's career when he was likely one of the best in the world while in the AHL. But close enough.

You made the point about Bill Durnan's weak competition in the two war years. It's a good point! The NHL goalie seasons for those two years were basically 17% Hall of Fame (Durnan), and 83% fringe NHLer/minor leagues. We know this not just by looking at the statistics from the seasons in question, but also at the full careers of the players involved.

So if you recognize that the mean level of goaltender performance was lower in 43-44 and 44-45, why cant you recognize that it was higher in other seasons? And also that the number of teams inevitably affects the mean level of performance, because NHL goalies are drawn from the far right tail of the distribution and it makes a difference where you draw the line in that tail.

Professor What · Aug 21, 2024

Here's what I guess I don't get in all of this: If third/fourth place is league median (better term than average here, I think), why is it so hard to believe that league median could be elite? I'll gladly grant you that when expansion finally took place, it took place at a rate that outstripped the talent pool, but the NHL was so slow to expand that I don't think it's hard to believe that the talent pool would have outstripped the rosters. When everybody is playing all, or nearly all the games too, spots are at a premium. And you're talking being league median if a whole two guys are better than you one year. You're worst in the league if a whole five guys are better than you.

So, I'll grant you that the talent pool is deeper now, but just for reference, I decided to look up the sixth-best goalie in recent years, based on all-star voting.

Year	Goalie	GAA	Sv%
2024	Jeremy Swayman	2.53	0.916
2023	Alexandar Georgiev	2.53	0.918
2022	Andrei Vasilevskiy	2.49	0.916
2021	Semyon Varlamov	2.04	0.929
2020	Darcy Kuemper	2.22	0.928
2019	Darcy Kuemper	2.23	0.925
2018	John Gibson	2.43	0.926
2017	John Gibson	2.22	0.924
2016	Cory Schnieder	2.15	0.924
2015	Steve Mason	2.25	0.928
2014	Henrik Lundqvist	2.36	0.920
2013	Corey Crawford	1.94	0.926
2012	Marc-Andre Fleury	2.36	0.913
2011	Ilya Bryzgalov	2.28	0.921
2010	Miikka Kipprusoff	2.31	0.920
2009	Miikka Kipprusoff	2.84	0.903
2008	Miikka Kipprusoff	2.69	0.906
2007	Ryan Miller	2.73	0.911
2006	Manny Legace	2.19	0.915

There are a couple of things I'll note here, again, the talent pool is deeper overall, but I think it's generally agreed that the post-lockout era isn't exactly the strongest for goaltending, so those two points should kind of balance out to an extent.

I averaged out the stats (just averaged the numbers, didn't recalculate everything with games played or minutes or anything), and it came out to a 2.36 goals against average and a .919 save percentage. Those would be pretty solid career averages. An example of a 2.36 GAA career is Bill Durnan. An example of a .919 SV% career is Roberto Luongo. You'll find lower rated goalies with the same or similar stats too, of course, but that's the kind of player you can find with those exact stats. That GAA would be 17th all-time, and that SV% would be 7th all-time (keep in mind that save percentage doesn't have the same length of history). If a guy carved out that kind of career today with any level of consistency, he'd be hailed as an all-time great, and he'd be going fairly early in our project.

I guess what I'm saying is that if a guy could consistently stay in the top six goalies year in and year out, he'd be something special. I don't see what's so hard to accept that the top 4 or 5 guys who are competing for an entire slate of six slots that have to hold on to the jobs year after year could be greats.

Michael Farkas · Aug 21, 2024

Right. Averaging an average is mixing vanilla with vanilla and expecting to get strawberry.

When you're dealing with such finite numbers, anything is possible. Reasonably possible, at that. There isn't an even distribution of talent. It's very possible that the 5th best goalie in 1961 was better than the 2nd best goalie in 1981.

Ya know...Truman Capote and Harper Lee were neighbors. On any other block in America, the second best writer is probably not as good haha

Bear of Bad News · Aug 21, 2024

Reminds me of Flight of the Conchords' "The Most Beautiful Girl in the Room". Relevant lyrics:

Looking 'round the room, I can tell that you
Are the most beautiful girl in the room
In the whole wide room
And when you're on the street
Depending on the street
I bet you are definitely in the top three
Good looking girls on the street
Depending on the street

(Last line bolded by me).

Anyhow, Mike's right. Here you go.

HOH Top 60 Goaltenders of All Time (2024 Edition) - Preliminary Discussion Thread

How many goalies should make the final list?

Final list of 60, Round 1 list submission of 80

Final list of 80, Round 1 list submission of 100

Registered User

Celebrate 68

Registered User

Not Too Soft

Registered User

Registered User

Registered User

Celebrate 68

Registered User

Registered User

Celebrate 68

Celebrate 68

Registered User

Registered User

Celebrate 68

New poll series coming from me in June

Celebrate 68

Registered User

Registered User

Registered User

Celebrate 68

Registered User

Registered User

Celebrate 68

"The Worst Guy on the Site" - user feedback

Ad

Ad

Ad