But at the same time, if we're going to look at Price as a what if case, shouldn't we do the same thing for Lundqvist? What if Lundqvist plays his whole career in Boston? Would his trophy case not look something similar to what you're saying Price's would? Lundqvist actually was a five-time Vezina finalist. Is it so much of a stretch that he would have won five Vezina's on a loaded team? Do the Bruins not fare better in their Stanley Cup appearances? They're still going to win 2011, but I think they probably win in 2013 as well. Rask had an absolute meltdown in game 6 that year, something that I feel confident Lundqvist wouldn't have done. Could you say the same thing about Price there? Yeah, probably so, so what's the difference. Point is, if you can play the what if game with Price, I think you can play it with Lundqvist as well, and I think you can come to the same verdict.
To that, I'd say that there's a reason most participants would rank Lundqvist ahead of Price anyways, and it's not pure talent. I'm fairly confident that Lundqvist on the Bruins would rank at the exact same place as where I ranked him.
To that, I'd say that there's a reason most participants would rank Lundqvist ahead of Price anyways, and it's not pure talent. I'm fairly confident that Lundqvist on the Bruins would rank at the exact same place as where I ranked him.
I feel like that's the way it should be though, isn't it? If we're doing this right, the team that the players are on shouldn't have a major impact on their rankings. Aren't we supposed to be trying to see through all of the noise and seeing the goalies' careers for what they are?
Am I crazy for thinking Rollins belongs somewhere close to Hextall? You guys know how I feel about awards being the be all, end all, but Rollins' Hart record can't be ignored can it? I think it is somewhat overblown (as in I don't know why he won the Hart in 1954 unless it was a sympathy vote), but he did a pretty danged good job on some really bad Blackhawks teams. Not a lot of longevity, but meanwhile, Hextall had what? His rookie year? He finished 51 on the aggregate list last time, and while he would have been passed by a few, he's still going to be comfortably top 80, which I feel makes Rollins safe too.
Rollins is a good case study for how hard it was to win and keep one of the six goaltending jobs in the NHL of the time. He lost jobs to four Hall of Fame goalies, and Hank Bassen was constantly threatening his job in his final three seasons in Chicago.
Pre NHL - age 18-22
Rollins started playing pro hockey in the PCHL as a teenager and put in three seasons from age 18-20. After he left the Vancouver Sun said Rollins had been the best goalie on the coast, thanks to the coaching of Porky Levine.
Rollins then moved to senior amateur hockey with the Edmonton Flyers, gambling that a run at the Allan Cup would get the attention of NHL teams. His gamble paid off as the Flyers won the Allan Cup, and Chicago signed the 21 year old Rollins. He played a season with their Kansas City minor league affiliate. The plan was that he would move up to the big club the next season, but Rollins was blocked when they signed the veteran Frank Brimsek instead.
NHL - age 23-30
Rollins started the 1949-50 season with the Cleveland Barons of the AHL. Early in the season, Turk Broda was 7 pounds over his playing weight in Toronto, and Conn Smythe made a splash by trading four players and cash "in five figures" for Rollins. However, the dispute between Broda and Smythe was quickly resolved, and Rollins spent the year riding the pine in Toronto and playing for the Leafs' AHL affiliate in Pittsburgh.
Rollins finally got a chance to play in the NHL in 1950-51, at the age of 24. He and Broda split time in the net, under a scheme where Smythe planned for them to play an equal number of games so they could jointly win the Vezina. However, Rollins was clearly outplaying Broda, so late in the season, so the Leafs abandoned the scheme and went with Rollins. Rollins finished with a 1.77 GAA in 40 GP and did indeed win the Vezina trophy, as the Leafs allowed one fewer goal than the Wings.
Rollins played a full season for the Leafs in 1951-52. However, he and the team tailed off late in the season, and Toronto fans started to talk of trading for Harry Lumley. Rollins was hurt going into the playoffs and Turk Broda came out of retirement to split the playoff games with him. It didn't matter who played because the Leafs couldn't score. Smythe traded Rollins and three other players to Chicago for Harry Lumley after the season, and admitted that he had heard the demands from fans to see Lumley in the Leaf goal.
Rollins played very well in the 1952-53 season and got the largest share of credit for the Hawks making the playoffs for the first time in seven seasons. He finished second in Hart voting.
In 1953-54, Chicago was terrible. Rollins actually had pretty good stats through the end of December, and tailed off after that. However, writers admired his never say die attitude and the incredible number of pucks he stopped and narrowly voted him the Hart trophy over Red Kelly. (Kelly was voted the Player of the year by the United Press, suggesting that Rollins' Hart win did not mean he was seen as the best player.)
In 1954-55, Rollins was hurt in the second half of the season and Hank Bassen played very well in replacing him. Bassen challenged Rollins in 1955 training camp, but Rollins kept the job. He briefly lost it in January when he was sent to the minors to get some rest and give Bassen a look. Although Rollins threatened to retire, he reported to Buffalo and later came back to Chicago to finish out the season.
Rollins won the job again over Bassen in the 1956 training camp and played his final season in Chicago. After the 1956-57 season, Chicago traded for Glenn Hall, who took the starting job. Rollins threatened to retire again but ended up accepting the assignment to Calgary of the WHL.
Post NHL - age 31-34
Rollins led Calgary to the WHL finals in 1957-58 and was named Calgary's sportsman of the year. But the Chicago organization replaced him after the season with young Roy Edwards, formerly of the Whitby Dunlops. Rollins had to sue the Chicago organization to get his release, and he signed with the WHL's Winnipeg Warriors for the next two seasons. He got back to the NHL for 10 games in 1959-60 with the cellar-dwelling Rangers, filling in for the injured Gump Worsley.
Rollins subsequently went into coaching, but he briefly returned to the nets at age 35, as the WHL-leading Portland Buckaroos signed him for the stretch run and playoffs. (They lost in the playoffs).
Overall
I would say Rollins had one outstanding full season (1952-53), and two outstanding half seasons (1950-51 and the first half of 1953-54). Other than that, he was a lower end starter for the other five seasons who was constantly at risk of losing his starting job. He was probably one of the best goalies outside the NHL for several seasons as well (arguably 46-47 through 49-50, and 57-58).
Rollins played seven NHL seasons from 1950-51 through 1956-57. And he was under constant competitive pressure from other goalies the whole time, including several Hall of Famers. He missed out on the Chicago job when they signed Brimsek. He had to compete with Broda for playing time in Toronto, and then he was traded away to be replaced by Lumley. In Chicago, he had Bassen challenging for his job the whole time, and he lost his job when they traded for Hall. Then Edwards pushed him out of his minor league spot with Chicago, and he was behind Worsley with the Rangers' organization.
In a larger league, maybe Rollins makes the league in 48-49 at age 22 and has a 10 season career through 1957-58. Maybe longer if he wanted it, it's hard to say because he was a minor league vet who transitioned into coaching, but maybe he would have kept going through his 30s if he had a shot at an NHL salary. He would still have a couple of big seasons but nothing outstanding otherwise.
Rollins, at 6'2", was the tallest goalie in NHL history when he entered the league. Doug Vaughan of the Windsor Star said of Rollins, "There were times when, despite his unorthodox style, he looked like a world-beater. On other occasions he didn't look like a National Leaguer."
I get the perspective- and it's not the first time I've heard this sort of expression- but I think one could pursue this to a fault.
To take examples from one of my previous posts- three Hall of Fame Goaltenders lost significant time to WWII. Two of them lost a couple of years of prime. Johnny Bower languished for how long before being given an NHL-chance? In the former case, these players did not stop being great Goaltenders- they were occupied with something that they judged more important. In the latter case, it doesn't take too much of a leap of imagination to consider that maybe Bower was a perfectly fine Goalie before he was given his NHL chance. To cite an alternative example up-thread, it's not like Shesterkin had limited effectiveness before coming onto the NHL-scene.
Brodeur was one of the (at least) three top-100 players collected in all three NHL work-stoppages of the last 30 years. Now, probably didn't ding his legacy so much to have fewer games in 94-95-- since Goaltenders are judged on rate statistics more than counting statistics. And it certainly didn't ding his legacy to have a reduced season in 2012-2013, as he was an aging below-average G in hang-on mode. However, the complete loss of the 2004-2005 season kind of stings, as he'd won the Vezina, achieved AS-1 status, and was a Hart finalist in the two preceding campaigns.
You can. But the difference is: Lundqvist is a "B" talent and he probably maximized that in an already very goalie-friendly (defensively at least) environment.
I look back at the last 25 years of goalies, maybe more...I think I have 4 or 5 "A" talents. So, that's why they're special to me and they get that extra "what if". I'm not asking anyone to go with me on it, but I'm going to state the case because it was asked and I think it's a valuable piece of the project. This particular project.
Re: Lemieux and Orr. I don't have to wonder about anything. They absolutely dominated. Lemieux was the only 2 point per game player ever AND he did it without playing in War years (including the British/Argentina dust-up, which seemed to have a remarkable effect on the league for some reason). Orr played for 7 years, won 11 Norrises. I'm good.
The "what if" situation just doesn't seem quite as pronounced for skaters because they have a way of emerging. Erik Karlsson was on a dog **** team in 2023 and dropped a 100 point season on us and won the Norris for just being turned loose to do whatever he wanted. No one was up late enough to watch it, he piled up the stats. It's tough to pile up the stats as a goalie when that **** is going on in front of you.
Re: "What actually happened" is that since 2000, this happened:
Best save pct. in a season: Brian Elliott .940, Tim Thomas .938, ... Ben Bishop .934, Dwayne Roloson .933,
The best GAA combined in the last 25 years (min. 200 games) -
1. Roman Cechmanek 2.08
2. Dominik Hasek 2.13
The list starts with a guy who shouldn't have been in the NHL to begin with. Has that ever happened with points or goals for forwards or defensemen? Did P.A. Parenteau ever become a Hart finalist playing with Tavares?
I'm not saying throw out what happened by any means, I'm not saying that the real events are disputed or anything categorically similar.
What I'm saying is, we run some risk of turning this into or - to some degree - including a "best defensive teams" element to this list as opposed to it being just a greatest or best goalies list.
I will make it my duty to rail against that, but try to do so conversationally.
Because "what actually happened" is that absolutely horrible goaltenders lead in our "best" categories over the last 25 years. Not exactly renowned for being Mr. Playoffs - has any goalie in the last 20 years been to more Finals than Marc-Andre Fleury? This is what actually happened.
It's more fragile with goalies. So unless Roman Cechmanek is gonna be on every list because since EXPANSION in '67 no one keeps the puck out of the net on average better than that guy and he gave us four seasons - we're talking about Al Rollins, Thomas was on the list, etc. so four is fine.
If he's not on every single list, then everyone is playing a little bit of "my" game...
Something that I'd be very interested in seeing is a timeline of goalie coaching in the NHL and juniors.
Everyone in the league now has had goalie coaching at every level. That wasn't always the case. Historically, as far as I know, goalies didn't have any technical coaching until some time after expansion (correct me if I'm wrong), beyond just learning from other goalies. When did that change, how gradual was the change, and which goalies benefited from that change?
The falloff in statistical performance isn't linear when expanding from 6 to 30 teams. If you ranked years, you won't get 1->1, 2->2, 3->3, 4->4, 5->5, 6->6, it's more like 1->1, 2->7, 3->13, 4->19, 5->25, 6->30.
Yes you could map the 1960 NHL stats to the 2024 NHL. You could also map stats from any other league with similar scoring levels to the 2024 NHL. Maybe a WHA season, maybe an AHL season, maybe a WHL season. Will stats still map 1->1, 2->7, 3->13, 4->19, 5->25, 6->30? Maybe they will, but so what? The stats may be the same but they don't mean the same thing.
Let me return to the 62-63/19-20 comparison again, because I've found another way to show you comparable seasons. [Reminder: Teams played 70 games in both those years, and scoring was the same in both years, at 208.]
Name
GP
Shots
Saves
Sh/G
GA
GAA
SV%
Min
Sawchuk
48
1328
1211
27.67
117
2.54
0.912
2760
Binnington
50
1430
1304
28.60
126
2.56
0.912
2948
Same games played, same shots faced, same GAA, same save percentage, same minutes played. Sawchuk had an AS-2 finish, Binnington finished 7th in both AS and Vezina voting. St. Louis scored 223 and allowed 190 in 71 games, Detroit scored 200 and allowed 194 in 70 games. They're identical resumes, 60 years apart.
Except that Sawchuk played in a league where the talent was compressed into 6 teams, and Binnington played in a league where the talent was spread over 30 teams, with a salary cap enforcing parity.
The stats are nearly identical but the stats don't mean the same thing. They were generated under different conditions.
2nd place doesn't have any intrinsic meaning. It's obviously harder to finish 2nd in a sample of 500 than it is in a sample of 2. As your sample expands, you're just getting a more accurate range of seasons, whatever you're measuring.
Under the conditions of competitive selection, adding additional seasons does not make the range more accurate. It makes the range different, because you're adding lower quality seasons.
You could add 498 sample seasons to your initial sample of 2, and end up with the top 2 being the same.
[For the sake of the argument, imagine a 100m dash - your initial sample includes Noah Lyles and Kishane Thompson, and their times are 9.784 and 9.789. You can add 498 other times (say of male 100m sprinters from this year) to your sample, those 2 are still going to be 1 and 2 in it.
OK, now we're getting somewhere. I agree. You could have peak Usain Bolt racing against any number of sprinters. It doesn't matter what the "sample size" is, Bolt will win.
Now imagine your initial sample is Noah Lyles and yourself, and then fill in the other 498 slots. Now, given the same parameters, you finish 1st and 500th. 2nd place has no raw value.]
Now let's put it into practice. Which example more accurately describes the Original Six NHL, with Glenn Hall, Jacques Plante, Terry Sawchuk, and Johnny Bower? Are they Noah Lyles and Kishane Thompson? Or are they Noah Lyles and myself?
Hall, Plante, Sawchuk, and Bower were Noah Lyles and Kishane Thompson. The best in the world.
The list starts with a guy who shouldn't have been in the NHL to begin with. Has that ever happened with points or goals for forwards or defensemen? Did P.A. Parenteau ever become a Hart finalist playing with Tavares?
Is RNH 104 points season on the crazy Oilers power play of 2023 that different in terms of impact on is stats than playing for Julien-Chara or Hitchcock for a goaltender ?
How much less points Blair Mcdonald score on the second line of a regular team instead of putting 46 goals and almost 100 points with Gretzky ?
Something that I'd be very interested in seeing is a timeline of goalie coaching in the NHL and juniors.
Everyone in the league now has had goalie coaching at every level. That wasn't always the case. Historically, as far as I know, goalies didn't have any technical coaching until some time after expansion (correct me if I'm wrong), beyond just learning from other goalies. When did that change, how gradual was the change, and which goalies benefited from that change?
Marc-Andre Fleury - you know, the 1st overall selection that started opening night a couple months later as a child - didn't have a full-time goalie coach when he first came into the league. But that's because he drafted by practically an expansion team.
I don't believe Jim Corsi was technically full-time in Buffalo during the Hasek and post-Hasek era.
Jacques Plante was brought in as a goalie coach and goalie development coach in the middle of the 85-86 season for St. Louis. I say goalie development too because they state that he was going to be headed to Windsor to coach the Blues prospects soon after.
So, it probably varies quite a bit. Whether it's FT guys, PT guys, consultants, etc. There's apparently pretty loose records of that, but that's another really interesting part of this. It might be interesting to see if there's a goalie coaching tree that forms.
Is RNH 104 points season on the crazy Oilers power play of 2023 that different in terms of impact on is stats than playing for Julien-Chara or Hitchcock for a goaltender ?
How much less points Blair Mcdonald score on the second line of a regular team instead of putting 46 goals and almost 100 points with Gretzky ?
You're exactly right. The difference is: They didn't win ****. Yeah, they had better stats. We know why. Did Dupuis become a 1-AS or 2-AS? Was RNH a Hart finalist for 100 point season out of nowhere? Not by a long shot.
Why? That is what actually happened? Nuge out-goaled and out-pointed and was better defensively than Artemi Panarin. Panarin was voted the better LW. Why? Folks made an adjustment for the situation.
Also, in a rare double whammy verdict for you guys, I've also proven that reincarnation is real...
Here I am in a previous life...
The Hockey News - Dec 24 said:
Adams also vetoed the other suggestion of the fan, John M. Tweedy, which would have rated netminders on the average number of saves in relation to total shots on goal. His objection to this idea was the difficulty of differentiating between real tough shots on the net and those weak dribblers and rollers which every goalie turns aside with ease.
“If there were some way of taking into consideration a goalie’s saves on good scoring chances by the other team — say a line were drawn halfway between the goal and the blueline, with only shots from inside that line counting as good scoring chances — then the rating might mean something,” he submitted.
Even the Boston chief’s idea of such a line wouldn’t be a fair test, since decisive goals frequently are scored on sizzling drives from the blueline, and other long shots test the goalie’s skill just as much, as weaker efforts from a few feet out.
As THE HOCKEY NEWS sees it, the only logical solution would be to have officials whose duty would be to rule on shots on goal, classifying them as potential scoring shots, or easy saves. That brings in the element of human judgment, which sometimes errs, but one point to remember is that baseball has the same situation in its classification of safe hits and fielding errors, and the system on the whole works out very satisfactorily.
Haha just following this thread for now. I'll send one in, been a busy year on the personal side, just have to update it. Hopefully I'll have time to participate in the project fully.
At some point or another, I'm (we're?) gonna be up on cinderblocks when it comes to goalies from too far back. I also wonder at what point they actually became goaltenders, as opposed to defensemen with bigger pads. I've read some accounts of guys on the Coast skating out with the puck to a degree that sounds a little fishy to me. I think it's in the pre-1950 goalie research thread, but since I'm working backwards from present day, I haven't gotten close to that yet.
The way to deal with those guys is to compare them against what the expectation was for that period at the time.
Goalies absolutely skated around with the puck (a lot) through 1909. I'm talking to the point where a couple guys made rushes on goal, and even scored (Fred Brophy being the guy that really sticks out- I've got him with 2 goals over 17 games in the CAHL and ECAHA, which is a better rate than quite a few defenders of the time).
Were goalies as refined as puck stoppers back then as they are today? Absolutely not. But I don't think we should be holding that against them in this project, since they didn't have access to the pads, training, and tactics that later goalies had/have. We'd be ignoring a sizeable chunk of hockey history otherwise.
The way to deal with those guys is to compare them against what the expectation was for that period at the time.
Goalies absolutely skated around with the puck (a lot) through 1909. I'm talking to the point where a couple guys made rushes on goal, and even scored (Fred Brophy being the guy that really sticks out- I've got him with 2 goals over 17 games in the CAHL and ECAHA, which is a better rate than quite a few defenders of the time).
Were goalies as refined as puck stoppers back then as they are today? Absolutely not. But I don't think we should be holding that against them in this project, since they didn't have access to the pads, training, and tactics that later goalies had/have. We'd be ignoring a sizeable chunk of hockey history otherwise.
I don't want to throw out "sizeable chunk of hockey history" - which is almost entirely in the pre-consolidation era - but I also don't want to spend a lot of time thinking about guys that may or may not have been interchangeable with a spare defenseman. I'm not saying that's necessarily the case, but I'm skeptical of anything that a) I can't see and b) the newspaper accounts paint a skepticism-inducing picture
At some point - and I'm not saying it's 1930 or 1910 or 1895 or whatever - we can say that the talent level was not high enough to include a player.
On the previous list, we have Vezina, Gardiner, Benedict, Worters, Thompson, Hainsworth, Holmes, LeSueur, and Connell as guys that played or started playing before the Great Depression.
We have Billy Smith, Grant Fuhr, Tom Barrasso, and Mike Liut as the guys who primarily played in the 1980s...
Are we sure about that? Seems like we skipped town for lower scoring times. I have my doubts. I have my doubts about who the 80's guys are, I have my doubts about the quantity of 70's and 80's guys vs 1920's guys, I have my doubts about the caliber of hockey in 1918, I have a lot of questions...
For whatever it's worth, I'm going to include Tom Paton on my list.
But one takeaway I got from the pre merger project was how relatively weak goaltending was in early hockey. You get skilled skaters early on and centres/rovers become technical really early. By the early 1890s skaters were just straight up good.
But goaltending has a bit of a lag period. I think we recognized that too. Only 3 goalies in the top 26. Tom Paton was the only goalie who had his prime pre 1905. We had loads of skaters who were in their primes pre 1905.
I am the last guy to diminish early hockey. But I also think we can acknowledge that goaltending made a significant technical jump later than skaters did.
At the same time I think we can also recognize that Vezina being elite for as long as he was in the situation he was in is worthy of a top 10 spot.
Mike Karakas is an interesting guy for me. I'm not far enough into my list yet to know if he's necessarily near my top 80, but he sure has a strange career arc. Wins the Calder in 35-36, then has a heroic performance with a broken foot for the notorious 37-38 Black Hawks, but then things start getting weird.
The way I've seen it framed, he sort of gets buried in the minors for asking for a $500 raise from Blackhawks management. Am I reading that right, or is there a legitimate reason to believe he's not NHL calibre for those "lost" seasons? He goes from playing all 48 games 4 seasons in a row his age 24-28 seasons, to out of the league for 4, and then makes it back.
He also wins a Calder Cup with the Providence Reds during his demotion, then returns with the Black Hawks during the weakened war years. Played his whole career for bad teams too. Big, athletic goal tender for the time, also credited as the inventor of the trapper, as well as being the first US born (and trained goalie). Some meat on the resume for sure but also lots of question marks. He's a guy I'll likely try to make space for late in my aggregate but I'm certainly interested in hearing other takes about him.
What about Hern and Hutton? I think it's pretty clear that Paton is the best of the trio (something I believe the pre-consolidation project got right), but depending on how highly you think of him, is there room on a list for the one or both of them?
Also, unrelated, but someone help me out with Normie Smith. He made 15 of 27 ballots last time around and finished 56th on the aggregate list, but why? I see a good rookie year, but one where he played in less than half of the games, and two years in the mid 30s where he was top notch. But beyond that, what is there? The longest game in history? The cap? I'm looking hard to figure out where I'm wrong, but I simply don't see the case for Smith.
If you don't have at least TWO European non-NHLers and TWO pre-NHLers on your list then how all-time great can your list be? (Have 1 certainly; get two) The NHL has been great but it hasn't been the end all and be all.
Mike Karakas is an interesting guy for me. I'm not far enough into my list yet to know if he's necessarily near my top 80, but he sure has a strange career arc. Wins the Calder in 35-36, then has a heroic performance with a broken foot for the notorious 37-38 Black Hawks, but then things start getting weird.
The way I've seen it framed, he sort of gets buried in the minors for asking for a $500 raise from Blackhawks management. Am I reading that right, or is there a legitimate reason to believe he's not NHL calibre for those "lost" seasons? He goes from playing all 48 games 4 seasons in a row his age 24-28 seasons, to out of the league for 4, and then makes it back.
He also wins a Calder Cup with the Providence Reds during his demotion, then returns with the Black Hawks during the weakened war years. Played his whole career for bad teams too. Big, athletic goal tender for the time, also credited as the inventor of the trapper, as well as being the first US born (and trained goalie). Some meat on the resume for sure but also lots of question marks. He's a guy I'll likely try to make space for late in my aggregate but I'm certainly interested in hearing other takes about him.
Thank you for bringing up, Mike Karakas! He was an incredibly underrated goaltender that will definitely be in my top 80.
A few years back I was in contact with his family and they sent over a 20-page bio that they had done up for a family book. I'm going to check it out now and look for some stories/anecdotes. If you or anyone else is interested in it to do some research on Karakas, feel free to shoot me a PM with your email address.
Yes you could map the 1960 NHL stats to the 2024 NHL. You could also map stats from any other league with similar scoring levels to the 2024 NHL. Maybe a WHA season, maybe an AHL season, maybe a WHL season. Will stats still map 1->1, 2->7, 3->13, 4->19, 5->25, 6->30? Maybe they will, but so what? The stats may be the same but they don't mean the same thing.
Except that Sawchuk played in a league where the talent was compressed into 6 teams, and Binnington played in a league where the talent was spread over 30 teams, with a salary cap enforcing parity.
The stats are nearly identical but the stats don't mean the same thing. They were generated under different conditions.
Under the conditions of competitive selection, adding additional seasons does not make the range more accurate. It makes the range different, because you're adding lower quality seasons.
OK, now we're getting somewhere. I agree. You could have peak Usain Bolt racing against any number of sprinters. It doesn't matter what the "sample size" is, Bolt will win.
Sure. We agree on the theory.
Now let's put it into practice. Which example more accurately describes the Original Six NHL, with Glenn Hall, Jacques Plante, Terry Sawchuk, and Johnny Bower? Are they Noah Lyles and Kishane Thompson? Or are they Noah Lyles and myself?
Hall, Plante, Sawchuk, and Bower were Noah Lyles and Kishane Thompson. The best in the world.
I feel like we're getting closer to speaking the same language, but we're not quite there yet.
To me, I think your biggest block is almost a romanticism associated with their names. That's why I referred to it as a tautology - it's a TERRY SAWCHUK season, therefore it's good because he's TERRY SAWCHUK. It is exasperated by there just being 6 teams, so an indifferent season from Sawchuk is still technically the 4th or 5th best season.
There's almost like two separate dimensions to this discussion - the theoretical world, and the practical world. In the practical world, in the universe of goaltenders that existed during the O6 save percentage era, Terry Sawchuk was somewhere between the 2nd-5th best goaltender of that time. You'd be correct that most of the seasons that would be filled in that fashion would be lower quality than Terry Sawchuk, because those goaltenders were not as good as him. We are in agreement there.
In theory world, however, your sample is whatever you deem it to be. If I want it to be the years 59-60 through 62-63, and just rank those 4 seasons (stats here, minimum games played 20, sample of 31), that 62-63 Terry Sawchuk year is somewhere between the 7th-12th best season. If I want to add the 17-18 through 20-21 seasons, because they're all about the same scoring levels (stats here, minimum games played 20, sample of 220), there's no reason to think that because Sawchuk's season was the 10th best in the first sample, that it will be somewhere between the 10th-20th best season in the combined sample. That's why percentiles are good as a shorthand - in the O6 sample of 31, Sawchuk's season ranked 7th-12th best is somewhere in the 20th to 40th percentile, in a sample of 220, the median 30th percentile is associated with the 65th best season, with a range of 44th to 88th.
Before expounding on that, I'm going to do a quick aside behind a spoiler using skater stats instead of goaltender stats, but sticking with that same 62-63/19-20 comparison.
The reason why I keep saying a player-season is a player-season, whether or not it is against 5 teams or 15 teams or 30 teams, is that while the methods and nature of hockey played has had a ton of changes throughout the years, on a statistical level it just hasn't since basically consolidation. I can randomize the best 6 seasons of goal-scoring wingers, from Conacher to Richard to both Hulls, to Bossy and Ovechkin, and on a percentage basis, you cannot tell the difference between each player. In their peak years, they all scored between 15-25% of their team's goals, and had points on between 30-45%. Like I did with the Sawchuk/Binnington comparison above, here's 4 player-seasons from 62-63 and 19-20.
Name
Year
Team
Games
Goals
Assists
Points
Team GF
LA GF
% LA
G%
P%
2
Andy Bathgate
62-63
NYR
70
35
46
81
211
208
1.014
0.166
0.384
3
Stan Mikita
62-63
CHI
65
31
45
76
194
208
0.933
0.160
0.392
8
Patrick Kane
19-20
CHI
70
33
51
84
208
208
1.000
0.159
0.404
10
Jack Eichel
19-20
BUF
68
36
42
78
193
208
0.928
0.187
0.404
The specific comparisons I want to make are Bathgate to Kane, and Mikita to Eichel. They're separated by nearly 60 years, but look at just how similar the seasons are. On teams that scored at a nearly identical clip, their P% is all just about 40, and while Eichel is a bit above, they're all around 16% of goals for their team. You'll note also that Bathgate/Mikita were 2nd and 3rd in scoring, while Kane and Eichel were 8th/10th, but again, that's the vagaries associated with expanding the league from 6 to 31 teams. Scoring isn't random, but adding teams just gives you more opportunities at having outlier seasons. It's the same with goaltenders as it is for skaters, and why the Sawchuk/Binnington seasonal comparison works for me.
Also, another quick aside, as to why exactly I chose those 4 seasons to look at - I had the league average numbers from 12-13 through 23-24 as well as pre-expansion, and these 4 consecutive seasons all were around the same scoring level each year. I didn't examine the samples in-depth until I had chosen the years.
Year
LA GF
LA GF/G
LA GF/82
59-60
206
2.94
241.3
60-61
210
3.00
246.0
61-62
211
3.01
247.2
62-63
208
2.97
243.7
17-18
240
2.93
240.0
18-19
244
2.98
244.0
19-20
208
2.97
243.7
20-21
162
2.89
237.2
Now back from that aside, let's return to my 2 samples from above, 59-60 through 62-63 and 17-18 through 20-21, and instead of looking at the best seasons, let's look at the worst. Here's a table of 3 of the worst years by GAA from all 8 years:
GP
GAA
SV%
GP
GAA
SV%
GP
GAA
SV%
59-60
Bower
66
2.69
0.919
59-60
Lumley
42
3.48
0.896
59-60
Worsley
39
3.52
0.897
60-61
Sawchuk
37
3.13
0.897
60-61
Worsley
59
3.29
0.913
60-61
Gamble
52
3.71
0.892
61-62
Worsley
60
2.92
0.912
61-62
Sawchuk
43
3.28
0.888
61-62
Head
38
4.16
0.879
62-63
Bower
42
2.60
0.912
62-63
Worsley
67
3.27
0.914
62-63
Johnston
50
3.98
0.893
51.25
2.835
0.910
52.75
3.330
0.903
44.75
3.843
0.890
GP
GAA
SV%
GP
GAA
SV%
GP
GAA
SV%
17-18
Anderson
58
3.32
0.898
17-18
Johnson
36
3.55
0.891
17-18
Greiss
27
3.82
0.892
18-19
Quick
46
3.38
0.888
18-19
Anderson
50
3.51
0.903
18-19
Ward
33
3.67
0.897
19-20
Bobrovsky
50
3.23
0.900
19-20
Dubnyk
30
3.35
0.890
19-20
Howard
27
4.20
0.888
20-21
Jones
34
3.28
0.896
20-21
Korpisalo
33
3.30
0.894
20-21
Hart
27
3.67
0.877
47.00
3.303
0.896
37.25
3.428
0.895
28.50
3.840
0.889
The modern years are a bit more curated, as I took the last place finisher every year, then the next player chosen was one who had played more games than last place, and then the final position was a goalie who played more than half his team's games, if one hadn't been chosen already.. So for 17-18, with 56 qualified goalies, I chose 52 Craig Anderson, 55 Chad Johnson, and 56 Thomas Greiss, for 18-19 with 60 qualified goalies, I chose 57 Jonathan Quick, 59 Craig Anderson, and 60 Cam Ward, for 19-20 with 57 qualified goalies I chose 54 Sergei Bobrovsky, 56 Duvan Dubnyk, and 57 Jimmy Howard, and finally for 20-21 with 47 qualified goalies, I chose 43 Martin Jones, 44 Joonas Korpisalo, and 47 Carter Hart.
I initially said I didn't do much curation for pre-expansion, as the only change I made was to nix Bob Perreault's 22 games in 62-63 for Bower's season instead, but I realized when I went back that had I not done that, I would've been double counting Boston goalies each year except for 60-61. So I went back, and chose the goalie with the most playing time for the 3 worst unique teams. The swaps this engendered - in 59-60 I swapped Simmons 28-3.26-0.900 for Bower's 66 games, both at 4th place, 60-61 was unchanged, 61-62 lost Gamble in last place, 28-4.32-0.879, so Sawchuk and Head moved from 4th/5th to 5th/6th, and Worsley's 60 games dropped to 4th, and in 62-63, my initial removal of Perreault's 22-3.83-0.893 for Bower's 42 games was upheld - otherwise Worsley would've been 4th and Perreault 5th.
I looked at the averages of those 4 years, and then I had a bit of existential crisis, because the "4th place" average was much better than the nominal AS-4 average that I thought I had posted before. What I realized is that from a previous version of this post, instead of looking at the worst goalies, I instead had curated the 4 best goalies from these particular years, and due to wonkiness in the AS voting/games played by tandems, the AS-4 finisher was sometimes 5th or 6th in GAA, whereas this table of three goalies is strictly 4th/5th/6th in GAA. The full 12 year average of AS-4 goalies that I had posted earlier was 54.67-2.79-0.913.
[I had a thought about the differences if I hadn't curated the modern selection, and changed the table around to see how the average changed - last place unchanged, next place goes from 37.25-3.43-0.895 to 34.25-3.52-0.893, and next to that goes from 47-3.30-0.896 to 29-3.37-0.895. Not a huge change, but I'm glad for the bigger games played samples. I also had a thought of trying to create a 1->7->13->19->25->30 chain of modern goaltenders, sorted by GAA, to see how that average compared to the O6 sorted in the same manner, but I'm not going to spend that time in this post now.]
In any case, look how similar last place is to each other, and how much pre-expansion jumps from 4th to 6th. Also, you have 3 Vezina finalists in Quick, Bobrovsky and Dubnyk, with Greiss winning the Jennings the very next year as part of a tandem (speaking to Mike Farkas' emphasis on defensive systems protecting goalies), compared to Bower, Sawchuk, Worsley and Lumley populating the pre-expansion grid.
Anyways, the more time I end up working on each individual post, the more meandering it gets, so despite not being completely satisfied with my narrative, I'm just going to stop editing and hit post.
but I also don't want to spend a lot of time thinking about guys that may or may not have been interchangeable with a spare defenseman. I'm not saying that's necessarily the case, but I'm skeptical of anything that a) I can't see and b) the newspaper accounts paint a skepticism-inducing picture
If the role of goalkeeper at the time was to act as a third defender, (which I agree that it definitely feels like it was at times, particularly pre-1900) then why shouldn't we just acknowledge that and go with it?
Why? It was the best hockey available in the world at that time. Small talent pool (especially pre-1900), sure, but again, they were playing (likely) against the best the world had to offer at the time. And as time progressed, there wasn't some watershed moment where good players suddenly became obsolete- generally speaking, the star players remained competitive and competed with the stars of the next era.
For example, let's look at a guy like Paddy Moran. Moran was (in my opinion, based on my readings of the papers) the best goalie in hockey by according to a poll in 1905. Probably earlier, but that all-star team makes it easy to use that as a mark. He was also being talked about as the best in 1913 (Victoria Daily Times, 22 March 1913), by which time Vezina had 3 years in the NHA. I'm not saying Moran was better than Vezina, but the evidence suggests he was in Vezina's area code. I don't think I need to talk about who else Vezina played with/against, so now I'll work backward from Moran-
Paddy Moran's Senior career began in 1902, and while he was definitely a good goalie that year, I don't think he was the best goalie in the CAHL (let alone hockey-wide; Art Brown may have a claim, or Riley Hern) that year. That was likely Big BIlly Nicholson (who I think deserves a look on a top-80 list) or Bouse Hutton (who I know has some support on this board). Both Nicholson and Hutton started playing goal at the Senior level in 1900. In 1900 you've still got Frank Stocking (severely underrated due to playing for crap Quebec HC teams) playing at a high level. Stocking's career goes back to 1893, which was Tom Paton's last season. Tom Paton goes back to 1885.
Great players are great players because they cross eras. We do this with skaters all the time, and we should extend that line of thinking to goalies.
For the record, I'm not convinced a pre-1900 goalie is a top-80 goalie of all time. But that isn't because I don't think we should be looking at hockey from the time period, but rather because I don't think that a pre-1900 goalie differentiated himself in the way that some skaters of the time period did, or in the same way that later goalies did.
On the previous list, we have Vezina, Gardiner, Benedict, Worters, Thompson, Hainsworth, Holmes, LeSueur, and Connell as guys that played or started playing before the Great Depression.
That's all fair, and ultimately, I think we agree more or less. I'm looking back at some of the instructional booklets and what not from way back and I'm thinking, "how much is scalable?" I'm a nut for adaptability and scalability. Are their scalable traits in Frank Stocking's game that could get me to, say, Clint Benedict? I'm not sure, I'm not sure if I'm ever gonna be sure because I don't think I'm going to get much film...
On the plus side, folks like you have done extensive research from back then and we'll just have to haggle out where these players may belong. I will say, if anything, I'd favor innovators a lot more from this era.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
