What's with the disrespect of Maurice Richard?

[tarheelhockey]

Why would removal of a number of players from the NHL automatically lead to higher scoring? And/or why does higher scoring automatically lead one to conclude a weaker era? There seems to be a bit of a "cause and effect" dynamic here.

That’s true, but there is an intuitive connection. Say you have a league where everyone is moving at 8-10 speed. 10s are your Richards, 8s are your “plodding” NHL depth defensemen.

Now a war comes along, and a whole bunch of guys leave. 8s, 9s, even some 10s disappear and are replaced by 7s and the occasional 6. Now you’ve got a league where every so often, a 10 comes roaring down the wing against a 6. It makes sense that scoring would go up.

Now, factor in that you’ve done the same thing with the goalies... a bunch of guys who naturally save in the .910-.930 range have been replaced with .890-900 types. Scoring will necessarily go up, without question.

Of course this can’t be proven to a strict standard because real players aren’t rated EA-style. But it does make sense, and it would tend to predict both inflated overall scoring and certain top-level scorers and goalies having otherwise unattainable seasons.

Not trying to validate my claim here, which for the record is that there is no statistical reason to think Richard's level of dominance was an anomaly in 44/45, just not quite getting this angle especially when the early 80s is also pointed to as being a weaker era.

The statistical reason, even just in the abstract numerical sense, is that the entire 44/45 league was an anomaly. Not just Richard but everything else about it was badly comparable to anything that came before or after it. That’s just what pure numbers show, narrative-free.

I completely acknowledge that a rise in scoring levels opens the door to more extreme gaps between offensive talents. I believe Wayne, Mario and Orr all benefit from this but the counter argument that every player had the chance to score 50 goals or 200 points also applies.

A nuance to this: higher scoring levels open the door to larger numerical gaps, but not to larger proportionate gaps. If anything, higher scoring ensures that fluke margins of proportionate victory are unlikely.

The most extreme example happens in Game 1 of each season, when some random player ends up leading the league in scoring with 2 goals to anybody else’s 1. At that moment, this player has a Mario-like 50% lead on everybody... and if he scores a hat trick, he’s in uncharted territory! But the margin shrinks rapidly as more goals are scored. The more goals scored by the end of the season, the less likely we are to see a flukishly huge margin of victory.

(The real-life mechanical reason for this: you can stumble into a hat trick with a legitimate goal, a puck off the butt, and an ENG. But you’re not going to score 80 goals on the basis of fluke events... the proportionate significance of random events and hot/cold streaks fades as the numbers climb higher.)

Total goal scoring being a function of (goals/game)*(number of games), we should look at both lower-scoring environments and shorter seasons with guarded suspicion of fluke outcomes. This leaves us with a bit of a paradox — lower scoring rates tend to be correlated with a higher overall quality of play, but also with a higher likelihood of randomly inflated achievements.

 

[The Panther]

Maybe rather than "disrespect of Maurice Richard", I should have said "over-emphasis on two War-era seasons early in his career when judging the totality of his accomplishments".

Let's focus on his career 1946 to 1958, since a lot of people can't get past beating him down for excelling for two seasons when the League was slightly weakened. (Somehow I never hear this argument used against Bobby Orr for basically his entire prime, yet it's always used against Maurice Richard for two seasons very early in his career.)

 

[The Panther]

I met Richard around 1990 or so. He still had those eyes. Those piercing eyes. He was rough around the edges, heck, he scared me a bit. He was probably 70 years old at this time and I wouldn't have wanted to anger him.

Wow, cool story. What did you say to Richard?

He was certainly a bear of a man. A powerful dude.

 

[JackSlater]

Maybe rather than "disrespect of Maurice Richard", I should have said "over-emphasis on two War-era seasons early in his career when judging the totality of his accomplishments".

Let's focus on his career 1946 to 1958, since a lot of people can't get past beating him down for excelling for two seasons when the League was slightly weakened. (Somehow I never hear this argument used against Bobby Orr for basically his entire prime, yet it's always used against Maurice Richard for two seasons very early in his career.)

Who is doing these things? You have repeatedly posted in this thread to whine about people pointing out the reality of the seasons affected by WW2 and grossly exaggerated what has been said. Who has over emphasized Richard's performance in the very weak NHL of his early years when judging his whole career? Who has beaten Richard down for excelling when the league was very, very weak? I'd love to see the posts that you keep whining about where this is happening.

I also remain amazed that you seem intent on embracing ignorance about the WW2 seasons. In no way was the NHL merely "slightly" weakened for two years in which Richard excelled, nor was it slightly weakened the year earlier either for what it's worth. It's also laughable to try and draw a parallel between the WW2 years and Bobby Orr's prime years due to the WHA but that might be worthy of a thread of its own.

 

[The Panther]

It's also laughable to try and draw a parallel between the WW2 years and Bobby Orr's prime years due to the WHA but that might be worthy of a thread of its own.

I was referring more to expansion (incl. WHA). From Orr's first season to his ninth (his entire Boston career, save 10 games) pro-hockey in North America went from 6 teams to 35. I personally don't care about this as any sort of defining factor in evaluating Orr's accomplishments, but I keep hearing Maurice Richard being downgraded because of two seasons in his career.

 

[JackSlater]

I was referring more to expansion (incl. WHA). From Orr's first season to his ninth (his entire Boston career, save 10 games) pro-hockey in North America went from 6 teams to 35. I personally don't care about this as any sort of defining factor in evaluating Orr's accomplishments, but I keep hearing Maurice Richard being downgraded because of two seasons in his career.

Expansion that didn't take players away from the NHL isn't particularly relevant. The WHA is relevant, but it didn't take away the percentage of the league or the quality of the players that WW2 did. I've also seen it suggested plenty of times, pretty reasonably, that scoring from Orr down to everyone else was inflated in that era due to the WHA.

I'll ask again. Who is downgrading Richard because of two seasons in his career? Where are these posts?

 

[Staniowski]

That’s true, but there is an intuitive connection. Say you have a league where everyone is moving at 8-10 speed. 10s are your Richards, 8s are your “plodding” NHL depth defensemen.

Now a war comes along, and a whole bunch of guys leave. 8s, 9s, even some 10s disappear and are replaced by 7s and the occasional 6. Now you’ve got a league where every so often, a 10 comes roaring down the wing against a 6. It makes sense that scoring would go up.

Now, factor in that you’ve done the same thing with the goalies... a bunch of guys who naturally save in the .910-.930 range have been replaced with .890-900 types. Scoring will necessarily go up, without question.

Of course this can’t be proven to a strict standard because real players aren’t rated EA-style. But it does make sense, and it would tend to predict both inflated overall scoring and certain top-level scorers and goalies having otherwise unattainable seasons.



The statistical reason, even just in the abstract numerical sense, is that the entire 44/45 league was an anomaly. Not just Richard but everything else about it was badly comparable to anything that came before or after it. That’s just what pure numbers show, narrative-free.



A nuance to this: higher scoring levels open the door to larger numerical gaps, but not to larger proportionate gaps. If anything, higher scoring ensures that fluke margins of proportionate victory are unlikely.

The most extreme example happens in Game 1 of each season, when some random player ends up leading the league in scoring with 2 goals to anybody else’s 1. At that moment, this player has a Mario-like 50% lead on everybody... and if he scores a hat trick, he’s in uncharted territory! But the margin shrinks rapidly as more goals are scored. The more goals scored by the end of the season, the less likely we are to see a flukishly huge margin of victory.

(The real-life mechanical reason for this: you can stumble into a hat trick with a legitimate goal, a puck off the butt, and an ENG. But you’re not going to score 80 goals on the basis of fluke events... the proportionate significance of random events and hot/cold streaks fades as the numbers climb higher.)

Total goal scoring being a function of (goals/game)*(number of games), we should look at both lower-scoring environments and shorter seasons with guarded suspicion of fluke outcomes. This leaves us with a bit of a paradox — lower scoring rates tend to be correlated with a higher overall quality of play, but also with a higher likelihood of randomly inflated achievements.

What accounts for the very big increases in scoring from '40-'41 to '41-'42 and from '41-'42 to '42-'43?

Were there many guys off to war in '42-'43? I know there were some, like Milt Schmidt.

 

[The Macho King]

I was referring more to expansion (incl. WHA). From Orr's first season to his ninth (his entire Boston career, save 10 games) pro-hockey in North America went from 6 teams to 35. I personally don't care about this as any sort of defining factor in evaluating Orr's accomplishments, but I keep hearing Maurice Richard being downgraded because of two seasons in his career.

Expansion watered down the league, but the best hockey players in the world (well - North America) were still playing here. Hull still had to compete with the Mikitas, Hulls, Parks, etc.

WHA is a bit of a different story - some talent did drain due to that. Still - for the most part the best players were still playing outside of a slightly older Hull and a now mid-40s Howe, and in fact you had a new generation of players coming up in Potvin, Robinson, Lafleur, et al. And Orr was still the best.

But yeah - Orr played in a weak era. That's pretty well established, and probably got one Norris he wouldn't have in a stronger era - doesn't change the fact he was still *by far* the best player in the league. Still the GOAT Dman.

 

[daver]

Expansion watered down the league, but the best hockey players in the world (well - North America) were still playing here. Hull still had to compete with the Mikitas, Hulls, Parks, etc.

WHA is a bit of a different story - some talent did drain due to that. Still - for the most part the best players were still playing outside of a slightly older Hull and a now mid-40s Howe, and in fact you had a new generation of players coming up in Potvin, Robinson, Lafleur, et al. And Orr was still the best.

But yeah - Orr played in a weak era. That's pretty well established, and probably got one Norris he wouldn't have in a stronger era - doesn't change the fact he was still *by far* the best player in the league. Still the GOAT Dman.

Substitute Richard for Orr in this narrative and I surely cannot see that much of a difference. At least one where Richard's 50 in 50 season is consistently questioned in narrative vs. any of Orr's seasons/career. Perhaps if there was some statistical evidence, like never approaching that level of dominance again, to back up the premise that Richard couldn't have had as dominant a 44/45 season if the war never happened then I could see that as reasonable claim.

Instead, I am left wondering: what else could Richard have done to eliminate any doubt as to the strength of his 44/45 season?

 

[daver]

A nuance to this: higher scoring levels open the door to larger numerical gaps, but not to larger proportionate gaps. If anything, higher scoring ensures that fluke margins of proportionate victory are unlikely.

The most extreme example happens in Game 1 of each season, when some random player ends up leading the league in scoring with 2 goals to anybody else’s 1. At that moment, this player has a Mario-like 50% lead on everybody... and if he scores a hat trick, he’s in uncharted territory! But the margin shrinks rapidly as more goals are scored. The more goals scored by the end of the season, the less likely we are to see a flukishly huge margin of victory.

(The real-life mechanical reason for this: you can stumble into a hat trick with a legitimate goal, a puck off the butt, and an ENG. But you’re not going to score 80 goals on the basis of fluke events... the proportionate significance of random events and hot/cold streaks fades as the numbers climb higher.)

Total goal scoring being a function of (goals/game)*(number of games), we should look at both lower-scoring environments and shorter seasons with guarded suspicion of fluke outcomes. This leaves us with a bit of a paradox — lower scoring rates tend to be correlated with a higher overall quality of play, but also with a higher likelihood of randomly inflated achievements.

I don't see the argument you are making, at least as it applies to higher scoring = more opportunities for generational talent to widen the gap between themselves and the pack.

No argument that more games played will bring individual scoring back to the mean but I am referring to the level of parity that lower scoring brings in. Tighter checking hockey i.e. playoff type hockey always lowers overall scoring, lowers the PPGs of the best offensive players, and lessens the % gap between the best offensive talents.

 

[daver]

The statistical reason, even just in the abstract numerical sense, is that the entire 44/45 league was an anomaly. Not just Richard but everything else about it was badly comparable to anything that came before or after it. That’s just what pure numbers show, narrative-free.

That 44/45 was an anomaly, and I am not denying it wasn't, and that Richard's raw total and GPG was anomaly, and I am not denying it wasn't, doesn't automatically make his performance, as measured by relative dominance, an anomaly. And by relative dominance, I mean Richard's total/GPG vs. a decent sample of his peers e.g. the other Top Ten goalscorers.

If you are going to use league GPG and individual goals/game stats, in a strict statistical sense, to declare something an anomaly, then in order to be consistent, one has to look at Richard's other seasons to declare his relative level of dominance in 44/45 an anomaly. I see overwhelming evidence that his 44/45 was not, in a strict statistical sense, an anomaly.

That one or two players could have reached a higher total than 32 goals in 44/45 or that one or two players could have gotten closer %-wise to Richard if one believes that scoring overall would not have ballooned in the war years, would make a marginal difference in the strength of that season. More than likely the players who reached career highs in terms of goalscoring placement e.g. Cain in 2nd place, fall back. So maybe you have a player or two higher than 32 but you also likely have a player or two lower than what they actually scored.

My whole issue with the "what if" scenario type of argument is that bias can be easily introduced to back up one's subjective narrative.

 

[JackSlater]

What accounts for the very big increases in scoring from '40-'41 to '41-'42 and from '41-'42 to '42-'43?

Were there many guys off to war in '42-'43? I know there were some, like Milt Schmidt.

By my count there were at least 43 players who were regulars in 1942 (I defined regular as having played half of a team's games in the 1942 season) who didn't play in 1943 and were involved in the war effort in some way that season. There are also players who left before the 1942 season like Muzz Patrick, but I didn't go back to count them. From 1942 to 1943 Brooklyn folded, which should have strengthened the remaining teams, but obviously 43+ players disappearing is equivalent to more than two whole teams disappearing in that era, and over a third of the league. Some guys who left were fringe regulars, some were well entrenched NHLers, 13 were players who either had made or would make a post season all star team, and out of those players 8 eventually got into the HHOF. Three of the players who left were starting goaltenders in 1942.

I'd say that this must be the biggest factor in the increase in scoring, and that's despite the NHL getting rid of overtime, and thus overtime goals, due to war time restrictions and dropping the highest GA team from the previous season in Brooklyn.

Instead, I am left wondering: what else could Richard have done to eliminate any doubt as to the strength of his 44/45 season?

Richard could only control what Richard did. He had nothing to do with the weakness of the NHL, and if Richard had been accepted into the military the NHL would have been all the weaker for it. He torched a weak league, which is the reality of the situation and a lot better than not torching a weak league. He then went on to torch the league when it was much stronger too, proving beyond any doubt that he was a great player. There were people at the time who thought that Richard was just a product of the war weakened league, most famously Smythe. Those people were obviously wrong. There is a large gap between "I'm going to take these accomplishments at face value" and "I'm going to dismiss everything that season because the NHL was extremely weak" and Richard's season very easily fits within that gap. It's possible to have a great season that is inflated by situations beyond the control of the player.

 

[Big Phil]

Wow, cool story. What did you say to Richard?

He was certainly a bear of a man. A powerful dude.

I am paraphrasing, but when he asked me if the paper he was holding was mine for his autograph I just basically said "Yes sir." That was it. haha.

 

[Big Phil]

I was referring more to expansion (incl. WHA). From Orr's first season to his ninth (his entire Boston career, save 10 games) pro-hockey in North America went from 6 teams to 35. I personally don't care about this as any sort of defining factor in evaluating Orr's accomplishments, but I keep hearing Maurice Richard being downgraded because of two seasons in his career.

The people who don't know hockey will use that against him, I find.

Look, Herb Cain had a great year in 1944. He won the scoring title that year with 82 points and was never near those point totals again. We all tend to recognize that Cain is a player who benefitted from the war years. He still carved out a long career in the NHL, he played 13 seasons and won two Cups, but I think we all agree he wasn't a star outside of those years and he only was in 1944 because of a depleted NHL. It is like, oh, I don't know, William Nylander winning a scoring title, probably along those lines.

Flash Hollett is probably not in the HHOF because of this too although there is a lot more of a case to say he was still pretty good outside of the war years. But people look at his big year in 1945 and say, well, it was 1945.

Richard didn't just do well in the war years. He we won the Hart in 1947, he led the NHL in goals 5 times, he obliterated the all-time goals record and he is probably the most clutch player in NHL history outside of Gretzky. Not to mention, Montreal lost in 1955 when he wasn't there, no coincidence I don't think.

Lastly, I'll refer to 1945. Sure there were guys missing. But there was also Richard, Blake, Lach, Kennedy, Cowley, Mosienko, Syd Howe. I think we forget there were still NHL stars in the NHL in those years. The league was missing the Kraut Line, the Bentley Brothers, Syl Apps, Roy Conacher and even goalies like Turk Broda were gone.

This is where the league was hurt the most, the goaltending was poor. Durnan and Lumley were still there but you had guys like Tubby MacAuley and Frank McCool in the league, so a lot of goals were allowed.

Richard had 50 goals and the next best was 32 in 1945. All the players were still shooting on the same goalies. But I have always felt 1947 was probably his best season and an even more impressive year than 1945 and that was his 45 goal year where he won the Hart. The next best player had 30, and no one was fighting a war, they were all playing.

 

[tarheelhockey]

I don't see the argument you are making, at least as it applies to higher scoring = more opportunities for generational talent to widen the gap between themselves and the pack.

No argument that more games played will bring individual scoring back to the mean but I am referring to the level of parity that lower scoring brings in. Tighter checking hockey i.e. playoff type hockey always lowers overall scoring, lowers the PPGs of the best offensive players, and lessens the % gap between the best offensive talents.

Lower scoring does everything you said except for lower the % gap for the best offensive talents.

Let's think about it this way. Imagine a ridiculously low-scoring and evenly-balanced hockey league where they play a 10-game schedule and every single game is 2-1.

• The best team in the league wins 60% of its games (6/10... this is a very evenly balanced league!) and therefore scores 16 total goals. The best player on that team scores an extremely impressive 25% of his team's goals, leading the league with 4 goals.
• The next-best team in that league wins 5/10 games, scoring 15 goals. Their top player is almost as good as the other guy, scoring 3 goals to lead the team.
• If we look at that season from a % basis, the league's leading goal scorer had a season for the ages -- he scored a quarter of his team's goals and won the league race by 25%! That's the kind of thing we'd compare to the very best of a Bobby Hull or Mario Lemieux. This guy is going down as a legend because of how he completely crushed the competition.
• But if we change lenses and look at the absolute totals, the same player appears to have won by the most negligible margin possible. In fact, we could take it a step farther and say one of those goals was a fluke own-goal by the opponent... in which case his victory is completely meaningless, effectively identical to a tie.

How can we square that circle? By recognizing that the lower-scoring a league is, the more it forces meaningless extremes in margins of victory. Not only does a larger scoring-total necessarily contain a smaller proportion of random flukes (for the same reason that 3 coin flips always produces a 100%+ margin of victory, while 3 million flips will effectively always produce nothing higher than a 10% margin) but the decreasing significance of each goal forces lower margins of victory unless some god-like player consistently and truly crushes the competition on a nightly basis.

The bolded rule means that we can predict that ludicrous 50%-type margins are likely to be tied to at least one of three conditions:

• Low-scoring seasons where a player benefitted from random variation
• Player benefitting from external circumstances (e.g. other elite players injured)
• The apotheosis of a god-like player who simply cannot be stopped by conventional means

A review of NHL history shows that a 49% goals margin has been achieved in the following seasons:

Brett Hull 1991 - 69%
Wayne Gretzky 1984 - 55%
Phil Esposito 1971 - 49%
Bobby Hull 1967 - 49%
Bobby Hull 1966 - 69%
Bobby Hull 1962 - 52%
Gordie Howe 1953 - 53%
Gordie Howe 1952 - 52%
Maurice Richard 1947 - 50%
Maurice Richard 1945 - 57%

Now, it's tempting to look at that list and say the "apotheosis" cases are easy to identify. But remember: the total number of goals scored doubled after 1967 and had nearly tripled by Esposito's 1971 season, due in part to expansion/dilution and in part to a longer schedule. From an abstracted standpoint, we would expect to see more huge-margin victories prior to 1967, and that's exactly what we see. It is not necessarily the case that pre-1967 scoring titles were driven by talent alone -- they may also be helped along by the margin-of-victory curve inherent in a smaller number of goals.

Of course, to validate that conclusion we would have to focus on specific circumstances (NHL seasons are not lab experiments). In certain cases (Richard, Esposito) we know that the numbers are partially driven by wildly imbalanced competitive dynamics (condition #2 above). Howe's margins are driven in part by historically low GPG during the early 50s, and Hull's in part by a curious lack of goal-scoring competition during those seasons (#2 finishing in the low-30s rather than the low-40s). That leaves a spotlight on Gretzky (an apotheosis situation) and Brett Hull, the latter season being particularly interesting IMO.

If you are going to use league GPG and individual goals/game stats, in a strict statistical sense, to declare something an anomaly, then in order to be consistent, one has to look at Richard's other seasons to declare his relative level of dominance in 44/45 an anomaly. I see overwhelming evidence that his 44/45 was not, in a strict statistical sense, an anomaly.

Hopefully the list above helps illustrate why people view it as an anomaly. Richard only put up massive margins of victory twice, and they were under extenuating circumstances.

I do agree, though, that this kind of thing can be painful if in fact the player really was that good coincident with all the external factors happening at the same time. There's really no way to get around the possibility of a player being downgraded unfairly for happening to have hit his peak at the wrong time. I suppose one way to potentially make an argument for him would be to show that he didn't benefit from running up scores on weaker teams, though it still is a hard mountain to climb even after having done the homework.

 

[daver]

Hopefully the list above helps illustrate why people view it as an anomaly. Richard only put up massive margins of victory twice, and they were under extenuating circumstances.

50/51 is on par with 46/47 (and not much less than 44/45) when you exclude Howe as an outlier. His GPG is just as dominant.

If we are going to add context to his 44/45 and 46/47 seasons, then we need to add context to his 50/51 season. It was clearly another dominant season by Richard after the war years had ended.

 

[JackSlater]

50/51 is on par with 46/47 (and not much less than 44/45) when you exclude Howe as an outlier. His GPG is just as dominant.

If we are going to add context to his 44/45 and 46/47 seasons, then we need to add context to his 50/51 season. It was clearly another dominant season by Richard after the war years had ended.

This makes no sense. Richard led the league in goals in 1945 by 56%. He led the league in goals in 1947 by 50%. He led the league minus Howe in goals in 1951 by 35%. That's not the same, and even if it were it doesn't mean that all totals should be treated the same and we should ignore that one season of the three is part of a massive outlier period in league history.

 

[tarheelhockey]

50/51 is on par with 46/47 (and not much less than 44/45) when you exclude Howe as an outlier. His GPG is just as dominant.

If we are going to add context to his 44/45 and 46/47 seasons, then we need to add context to his 50/51 season. It was clearly another dominant season by Richard after the war years had ended.

Richard was clearly dominant after the war, I don't think anyone disputes that. He and Howe were by far and away the dominant goal scorers of the early 1950s, with nobody else coming anywhere near them. He is acknowledged for that in being ranked the #9 player -- the #6 forward -- and as of a few years ago, the #3 winger of all time around these parts.

That being said, nobody is an "outlier" when they have peers neck-and-neck with them. Deleting Howe from the picture doesn't turn 1951 into an outlier season parallel to 1945 or 1947, seasons when Richard actually was an outlier.

The fundamental issue here is that Richard happened to enter his scoring peak at a time when a player of his caliber would necessarily have been an outlier due to the circumstances of the league in 1945-47. It's asking an awful lot of people to suggest they should take those outlier numbers at face-value, knowing that they were produced in what was probably the least-competitively balanced environment in NHL history. Scoring 50-in-50 against standard NHL competition is simply not the same achievement as doing the same against minor leaguers, right?

 

[daver]

This makes no sense. Richard led the league in goals in 1945 by 56%. He led the league in goals in 1947 by 50%. He led the league minus Howe in goals in 1951 by 35%. That's not the same, and even if it were it doesn't mean that all totals should be treated the same and we should ignore that one season of the three is part of a massive outlier period in league history.

His GPG in 50/51 was as dominant vs. the average GPGs of the other Top Ten goalscorers (62%) as it was vs. his 46/47 season (60%) and slightly less vs. his 44/45 (67%) season.

I hope you aren't going to deny this statistical reality.

 

[daver]

That being said, nobody is an "outlier" when they have peers neck-and-neck with them. Deleting Howe from the picture doesn't turn 1951 into an outlier season parallel to 1945 or 1947, seasons when Richard actually was an outlier.

The fundamental issue here is that Richard happened to enter his scoring peak at a time when a player of his caliber would necessarily have been an outlier due to the circumstances of the league in 1945-47. It's asking an awful lot of people to suggest they should take those outlier numbers at face-value, knowing that they were produced in what was probably the least-competitively balanced environment in NHL history. Scoring 50-in-50 against standard NHL competition is simply not the same achievement as doing the same against minor leaguers, right?

You can't continue to call out 44/45 as needing context and then not apply context to 50/51. Do I really have to cite the seasons where Mario and Wayne were both outliers to point out how disingenuous this is.

What else needs to be said when Richard is going head to head with the "Wayne Gretzky" of the "50s as statistical evidence that his war year totals do not automatically need to be devalued.

"The fundamental issue here is that Richard happened to enter his scoring peak at a time when a player of his caliber would necessarily have been an outlier due to the circumstances of the league in 1945-47."

You are right, it is an issue because Richard gave us every statistical reason to believe he could have been an outlier in any circumstance. My issue is that this season was "inflated" is being treated as fact, not speculation, let alone reasonable statistical speculation.

 

[tarheelhockey]

You can't continue to call out 44/45 as needing context and then not apply context to 50/51. Do I really have to cite the seasons where Mario and Wayne were both outliers to point out how disingenuous this is.

What else needs to be said when Richard is going head to head with the "Wayne Gretzky" of the "50s as statistical evidence that his war year totals do not automatically need to be devalued.

I think you're starting to abuse the concept of an 'outlier' here.

An outlier is not a Very Good Player. An outlier is not winning a scoring title. An outlier is a statistical mark that is ridiculously far ahead of (or behind) every other point in a data set. Brett Hull scoring 86 goals is an outlier. Mario Lemieux scoring 85 goals, not really. Maurice Richard scoring 45 goals in 1947 is an outlier... Maurice Richard scoring 42 in 1951, not at all.

The point of identifying outliers is to acknowledge that sometimes, weird things happen statistically. Above, I identified 3 major reasons why a weird thing might happen... they reduce to this:
1) Maybe there's just a mathematical quirk going on
2) Maybe something weird is happening off-ice
3) Maybe someone is just that good at hockey, to a point that he's completely wrecking the curve by which all other players are measured

Looking closely at outlier seasons allows us to determine where the "curve" exists for the rest of the league. Fortunately, almost all of them fall into categories #1 and #2, allowing us to maintain a fairly universal standard of measure for what hockey scoring looks like.

In the case of the 1980s, we had two players capable of potentially wrecking the curve to the extent of a silly 50% margin. Only one of them (Gretzky) actually did that... the other is forever consigned to "on pace for" type arguments which are statistically sticky (and for good reason). So no, there are not two 80s outliers comparable to 1940s Richard.

Even early-80s Gretzky has to be taken into a reasonable amount of context. While he did set the statistical bar by which all players are measured -- and we know he wasn't benefitting from statistical flukes, nor was his NHL some expansion-diluted or war-wracked environment -- that still doesn't mean we just completely ignore that the elite Soviets were absent, or that Gretzky was running up massive assist totals while playing on a firewagon-hockey dynasty team. It's not being "disrespectful" of Gretzky to say he was really more of a 30% type of outlier by nature, and the 50% comes into play from environmental circumstances. The crazy thing about Gretzky is, you can go ahead and make that discount and he's still dominant because of the consistency and scale of his dominance. So even taking his numbers in a very cynical light, there's no question that he really was that much better than the rest of the pack.

You are right, it is an issue because Richard gave us every statistical reason to believe he could have been an outlier in any circumstance. My issue is that this season was "inflated" is being treated as fact, not speculation, let alone reasonable statistical speculation.

We cannot know for a fact that Richard would have fallen short of 50-in-50 against standard NHL competition.

We can know for a fact that the competition during his 50-in-50 season was, in a big way, substandard.

Anything beyond that is inference. What you are asking us, is to infer that Richard would have done exactly the same thing against any group of players, no matter how good or bad they were. Most people infer that he would have done better against substandard players, because in general that is simply how competition works.

The only reasonable way that I can see anyone believing that Richard's 50-in-50 would have happened against standard competition, is to demonstrate that his numbers against really bad competition (e.g. the Rangers with Ken McAuley in net and Ab Demarco leading the offense) actually resembled what he would have done against standard bad NHL teams. Basically, demonstrate that he held back and chose not to pad his numbers against those teams, but that he completely dominated a relatively ordinary Wings team, and you have the bones of an argument that the 50-in-50 is real gold.

 

[Sentinel]

If anything, Richard gets TOO MUCH RESPECT. He is #9 at the All-Time List. He has ONE Hart and ZERO Art Rosses. Next time we see a player with regular season resume this weak is probably #36 Mike Bossy.

 

[daver]

I think you're starting to abuse the concept of an 'outlier' here.

An outlier is not a Very Good Player. An outlier is not winning a scoring title. An outlier is a statistical mark that is ridiculously far ahead of (or behind) every other point in a data set. Brett Hull scoring 86 goals is an outlier. Mario Lemieux scoring 85 goals, not really. Maurice Richard scoring 45 goals in 1947 is an outlier... Maurice Richard scoring 42 in 1951, not at all..

Serious question: Are you just trolling right now? I have stated that his goals per game is the metric used to show Richard's 50/51 is on par with his 46/47 in terms of relative dominance.

 

[daver]

Anything beyond that is inference. What you are asking us, is to infer that Richard would have done exactly the same thing against any group of players, no matter how good or bad they were. Most people infer that he would have done better against substandard players, because in general that is simply how competition works.

In terms of scoring 50 goals, nope, not asking that.

In terms of dominance, yep. You know why? Because there is statistical proof it wasn't an outlier season for him unless you want to f*** around with the numbers. He did again in 46/47 and in 50/51. Noone has put up any statistical evidence to back up their assertion that he isn't as dominant in 44/45 under different league circumstances.

 

[tarheelhockey]

The only reasonable way that I can see anyone believing that Richard's 50-in-50 would have happened against standard competition, is to demonstrate that his numbers against really bad competition (e.g. the Rangers with Ken McAuley in net and Ab Demarco leading the offense) actually resembled what he would have done against standard bad NHL teams. Basically, demonstrate that he held back and chose not to pad his numbers against those teams, but that he completely dominated a relatively ordinary Wings team, and you have the bones of an argument that the 50-in-50 is real gold.

I'm willing to put some skin in this game, if it helps resolve the argument. The Habs played each opponent 10 times in 1944-45, and thankfully Richard played all 50 games.

1944-45 Maurice Richard, goals by opponent (listed in order of final league standings)

Detroit Red Wings - 16
Toronto Maple Leafs - 8
Boston Bruins - 10
Chicago Black Hawks - 8
New York Rangers - 8

OK, so this ends up being a little more interesting than expected. The Wings were young, but full of NHL talent at that time. They finished 2nd in the league, yet Richard torched them. Basically, he scored the 8 goals he put up against every other team and then added a hat trick and a 5-goal game to boot.

It's not as though the Habs held back against the lower-ranked teams. Here are their team results that season:

1944-45 Montreal Canadiens, GPG by opponent (listed in order of final league standings)

Detroit Red Wings - 4.6
Toronto Maple Leafs - 2.8
Boston Bruins - 5.3
Chicago Black Hawks - 4.0
New York Rangers - 6.1

That picture certainly looks like Richard pouring it on against real competition, while taking a breather during the joke matchups. IMO, has very strong vibes of a player being taken off the ice during uncontested matchups. To verify that theory with isolated results from "extreme" games:

11-09-44 - Habs 9, Hawks 2. Richard = 3
12-16-44 - Habs 8, Bruins 5. Richard = 2
11-27-44 - Habs 11, Bruins 3. Richard = 1
12-28-45 - Habs 9, Wings 1. Richard = 5
1-06-45 - Habs 10, Hawks 1. Richard = 1
1-13-45 - Habs 8, Wings 3. Richard = 1
2-08-45 - Habs 9, Rangers 4. Richard = 2
3-11-45 - Habs 11, Rangers 5. Richard = 2
TOTAL: Habs 75, Richard 17 (23%)

So the only case of Richard really running up stats during a blowout win was against the Wings, a solid team against whom Richard probably played the entire game and really brought it.

On the other end of the spectrum:
11-02-44 Habs 1, Leafs 4. Richard = 0
11-11-44 Habs 1, Leafs 3. Richard = 0
12-03-44 Habs 1, Hawks 2. Richard = 1
12-14-44 Habs 2, Leafs 2. Richard = 1
12-23-44 Habs 1, Hawks 2. Richard = 0
1-04-45 Habs 2, Leafs 4. Richard = 0
2-01-45 Habs 1, Hawks 1. Richard = 1
3-03-45 Habs 2, Leafs 3. Richard = 0
3-15-45 Habs 1, Wings 2. Richard = 1
TOTAL: Habs 12, Richard 4 (33%)

So Richard actually produced a higher proportion of the Habs' offense during these tight, low-scoring games than he did during blowouts. Again, this IMO points toward him being leveraged in games where it really counted, and rested (or just taking it easy) when things got circus-like.

There is at least something of an argument here that in a more intense, standard NHL environment, Richard would have been deployed more aggressively and would have trended closer to that more-competitive 33% of team scoring, which on a high-scoring team with a 50-game schedule would have meant a 1/3 share of about 150 goals, which would put him right around...
... wait for it...

... 50 goals