Round 2, Vote 7 (HOH Top Centers)

[Sturminator]

I wouldn't be so sure about Stewart, personally. His 1926 season would boost his vsx7 score, but not really by much, right?

Well, if his 7th lowest season in VsX was an 80, replacing it with a 100 would boost his 7-season average by almost exactly three points. That's actually a pretty substantial difference, as you know, though not an enormous one. It is definitely worth nothing, at the very least.

The objection about Stewart's points occuring in a goals-heavy league is entirely fair. It is really hard to know how to account for this factor, but in the case of Stewart, I do not get the impression that he was Esposito-like around the net in that he was equally good at passing and shooting in tight. Every description I've heard of Stewart paints the picture of a shoot first, and second and third type of player, and this disposition would almost certainly lead to somewhat favorable points totals (and thus VsX totals) in comparison to how such a player would score in the modern game. I'm sure that his shooting generated a certain amount of rebound chances for his linemates, but I don't get the impression that he was particularly skilled at or, frankly, interested in passing.

 

[Sturminator]

By the way, here is a season-by-season summary of NHL assists per goal. It steadily rose throughout Stewart’s career, and by the end of it, it had gotten to more or less modern levels, but in his first few it was extremely low, which I can only assume benefited him in the points leaders:
*
http://www.quanthockey.com/TS/TS_AssistsPerGoal.php
*
Let me know what you think is appropriate and I’ll whip something up.

That's excellent data, 70's. Man, the 1929-30 season was really the wild west, wasn't it?

 

[seventieslord]

It is definitely worth nothing, at the very least.
.

...yes, it's worth at least nothing! I like when typos are funny... don't mind me. Anyway, I agree with everything you said. I didn't realize it was going to.affect him that much. I can run some numbers once a couple of you agree on a framework , so there's less chance of my work being shot down off the start as too "fantasy". For the record, I'm not saying I should just multiply out the assists to match today's rate of 1.7. That would be going too far.

That's excellent data, 70's. Man, the 1929-30 season was really the wild west, wasn't it?

Yeah, certainly an out of whack season with odd results in a few ways.

 

[MadArcand]

Who? 5.7 seasons worth of games so your 'opinion' is based on a lack of knowledge.

What does that have to do with anything? 3-4 great seasons. Partial & fractional seasons, PPG and other woulda-shoulda nonsense don't really matter to me, sorry.

 

[MadArcand]

Oh please. Don't take it as an attack, use it as an opportunity to defend your position.

I should point out though, that these three include a hurricane/whaler and a Slovak, two demographics you've always shown a, let's say, affinity for.

Whaler, sure.

Slovak? Please search for my posts on Stastny in this section and ATD. I just happen to think that being #3 during one's career - only behind the two greatest centers in history - should count for something.

 

[BraveCanadian]

Agreed.

I'm very much a prime & peak guy and thus (I think) in a minority here. Most seem to favor high quality longevity.

And here is Ron Francis, 2nd in assists only to Gretzky and 4th in points only to Gretzky, Howe and Messier in his career. A Selke Trophy winner. Was great on face-offs. A clutch player who scored big goals, and a captain.

Yet he is being dismissed immediately as a possible top 30 center.

I realize I am clearly biased here. I saw more of Ron Francis live than any other great player. But it should be noted that I was not a Whalers fan. I was not rooting for him as I would an Esposito. Still, I fully understand how underwhelming Ron Francis could seem to fans that didn't see that much of him.

It seems that in some cases the high peak cancels out the longevity and in some cases the longevity makes up for a lower peak.

Except for a guy like Francis.

Most people here probably didn't see him at his best, though. And since the numbers don't do him justice....

Finally Francis is up. It's weird how much sooner Fedorov came up - also compared to Gilmour, as has been mentioned already.

Interesting bunch of players without too much of clear separation in the pack. Initial thoughts would be Francis, Stastny & Lach for top 3, but it's hardly set in stone for me.

Agreed. Feds is easily the most purely talented of them but I'll stick to my point that I think they are all in the same range as two-way all around centermen based on their results.

Very funny. I don't understand why do you Venerators Of The Holy Peak feel the constant need to belittle anyone and anything opposing your dogma.

FYI I was one of the two who didn't vote for that three-year wonder last round.

I think people discount how difficult it is just to stay in the NHL for 15-20 years let alone be one of the better players.

Does one or two real standout seasons outweigh 5-6 very good ones?

Oh please. Don't take it as an attack, use it as an opportunity to defend your position.

I should point out though, that these three include a hurricane/whaler and a Slovak, two demographics you've always shown a, let's say, affinity for.

So don't take it as an attack.. here is an attack. lol

For the record I am not a Whaler/Cane or Penguin fan and I still think Francis is underrated in these parts.

 

[seventieslord]

Look, I've been In about a dozen drafts with madarcand and I feel I can say I have a good handle on what players he tends to be partial to. That's not an attack.

 

[MXD]

First -bias is a normal thing in our endeavor. As long as its not ridiculous, I can live with it. The point is to determine what's ridiculous and what's not.

Second, next time I see something like 15-20 years of very good output with good-to-great 2-way play in this thread, Im starting to cry. Like a bearded man in his mid 30ies.

 

[unknown33]

What does that have to do with anything? 3-4 great seasons. Partial & fractional seasons, PPG and other woulda-shoulda nonsense don't really matter to me, sorry.

You whole opinion is based on a lie that a player who has - woulda-shoulda - 5 seasons were he was voted at least Top 3 Center AS is a 3 season wonder.


http://hfboards.mandatory.com/showpost.php?p=75215481&postcount=42
You already got called out and seem not to care.

 

[Sturminator]

Look, I've been In about a dozen drafts with madarcand and I feel I can say I have a good handle on what players he tends to be partial to. That's not an attack.

I would say that madar's opinions on players he's actually seen quite often "go to eleven", so to speak. He has a tendency to either really love or really hate modern players, and is firm in those judgments. Everyone has a right to his opinion, but with all due respect (because I like the guy), I also consider madar's opinions on modern players, or at least the way that he expresses them, somewhat exaggerated.

But that's neither here nor there, and we're not here to talk about one another. On the subject of Peter Stastny, I think some people fall into the "Brad Park fallacy" when discussing his career. That is, they go on the assumption that Gretzky and Mario should be removed, and then say "look, without Wayne, Stastny is the highest scorer of the 80's". I refer to Brad Park because this argument is often made in his favor on the assumption that Bobby Orr should be removed, and that if he is, Park wins eleventeen Norris Trophies and farts chocolate.

This argument is faulty because it assumes that the mere existence of a generational talent is some kind of outlier, which is silly. What made Gretzky and Orr special is not that they lived or played hockey or even dominated, but rather that they dominated to the extent that they did. When evaluating Park, one should not remove Orr, but rather normalize him to an average "best defenseman in the NHL" over a reasonable lenth of time. Same for Stastny vis-á-vis Wayne Gretzky. You can't just arbitrarily remove Gretzky, but you can ask how Stastny would compare to a more down-to-earth best scorer of the decade, like say a Jagr-type, or a Lafleur-type, or even a Bathgate-type.

And this is where the fallacy becomes pretty clear. Stastny was a great player, but he doesn't compare all that well as a scorer to anybody who was really capable of holding that best scorer mantle over an extended period of time. If one completely removed Gretz and Mario he looks like that kind of scorer, but that result only reveals the fact that removing those players completely is a junky method.

 

[Canadiens1958]

Joe Malone I

Québec City newspaper reports are extremely sparse from his era so I will post what I can find from Montreal papers.

http://news.google.com/newspapers?id=VRYqAAAAIBAJ&sjid=_YQFAAAAIBAJ&hl=fr&pg=6706,1737176

January 15, 1913 game, Canadiens beat the Bulldogs 5-4 in a very rough game. Malone is not mentioned as being involved in the rough play, unlike Lalonde. Also he is mentioned as having passing skills on Québec,s first goal and working in tandemon the second goal.

http://news.google.com/newspapers?id=NxwqAAAAIBAJ&sjid=d4YFAAAAIBAJ&hl=fr&pg=6721,2183151

February 19, 1913 game Bulldogs win 4-2 Malone gets two goals, little mention of his opposing center Newsy Lalonde.

http://news.google.com/newspapers?id=OhwqAAAAIBAJ&sjid=d4YFAAAAIBAJ&hl=fr&pg=6778,2614726

February 22, 1913 game Bulldogs win 7-6 clinching league championship Joe Malone scores a hat trick including the tying and winning goals within four seconds. Joe Malone was not penalized in a very rough game.

 

[BraveCanadian]

What made Gretzky and Orr special is not that they lived or played hockey or even dominated, but rather that they dominated to the extent that they did.

The fact that they dominated to the extent they did is why they are considered outliers.

I agree that you then have to judge who is left based on some sort of average top player but that is exactly what the top of the pack is, in theory, after you remove those outliers.

 

[Cruor]

Look, I've been In about a dozen drafts with madarcand and I feel I can say I have a good handle on what players he tends to be partial to. That's not an attack.

A wild guess, none of them are Forsberg or Lundqvist?

 

[Sturminator]

The fact that they dominated to the extent they did is why they are considered outliers.

I agree that you then have to judge who is left based on some sort of average top player but that is exactly what the top of the pack is, in theory, after you remove those outliers.

You're somewhat missing the point. Generational talents are what they are because they were able to be consistently outliers, and in that they are special. Some people think they should be treated differently from normal outliers in standardized systems and the point is that outliers need to be handled the same across all eras of hockey. There has been an idea in circulation that generational talents should be removed in addition to the normal handling of outliers, and I am saying that this is a faulty idea. Systems which remove outliers, like VsX, are already removing Gretzky and Lemieux with regularity when setting the benchmarks. It is not necessary to remove them, and then remove the next best scorer after them, as well. Doing so leads to gross distortions in favor of everyone who played during their respective careers.

The point here is that Stastny's VsX score already normalizes for the effect of the generational talents against whom he competed, and his score even without those guys in the equation simply isn't that special, in this round or overall. The "Stastny was the #1 scorer of the 80's without Gretzky...ZOMG!" (which I realize you are not making) argument is a lot of smoke.

 

[BraveCanadian]

The point here is that Stastny's VsX score already normalizes for the effect of the generational talents against whom he competed, and his score even without those guys in the equation simply isn't that special, in this round or overall. The "Stastny was the #1 scorer of the 80's without Gretzky...ZOMG!" (which I realize you are not making) argument is a lot of smoke.

I agree with you that the VsX already accounts for Gretzky, obviously.

Also agree that I don't think that being the leading scorer of the 80s outside of Gretzky tells us anything about Stastny that we didn't already know about him. It is just neat trivia.

There is a cluster of great players very close to Stastny at that somewhat arbitrary cutoff.

 

[MXD]

Sid Abel
Max Bentley
Bill Cowley
Alex Delvecchio
Elmer Lach
Sergei Fedorov
Ron Francis
Dave Keon
Nels Stewart
Peter Stastny
Norm Ullman

Since submitting my initial list, I really soured on Delvecchio --- and I had him in the mid-30ies. Can't tell why -- I just don'T have the impression he was a difference-maker at any point in his career. Something everybody on that list was, and quite a few that aren't up for voting at the moment. Bottom-line, if I had to cast my list again, he probably be something like 10 spots lower (roughly where I had Marty Barry, actually), at the very least.

Does any of the oldtimers has some insight on Delvecchio?

Going by last round, Cowley has the inside track at top spot, but my round 6 vote was cast in a hurry cause I was afraid to miss the deadline (which was extended in the end). AT the moment, my gut feeling regarding this group is probably Stewart on top... but 70ies made some very valid points which, IMO, cannot be ignored.

 

[tony d]

Delvecchio's an interesting case. I think he should get voted in next time but I think he's either someone you like or you don't.

 

[Hawkey Town 18]

Don't have my data accessible. Can someone run the vsX numbers for Cowley Thornton and Oates best 4, 7, and 10 seasons with some kind if discount to cowleys war years. Thought I remembered them being very close when doing my round 1 list but perhps I'm wrong.

 

[Dennis Bonvie]

You're somewhat missing the point. Generational talents are what they are because they were able to be consistently outliers, and in that they are special. Some people think they should be treated differently from normal outliers in standardized systems and the point is that outliers need to be handled the same across all eras of hockey. There has been an idea in circulation that generational talents should be removed in addition to the normal handling of outliers, and I am saying that this is a faulty idea. Systems which remove outliers, like VsX, are already removing Gretzky and Lemieux with regularity when setting the benchmarks. It is not necessary to remove them, and then remove the next best scorer after them, as well. Doing so leads to gross distortions in favor of everyone who played during their respective careers.

The point here is that Stastny's VsX score already normalizes for the effect of the generational talents against whom he competed, and his score even without those guys in the equation simply isn't that special, in this round or overall. The "Stastny was the #1 scorer of the 80's without Gretzky...ZOMG!" (which I realize you are not making) argument is a lot of smoke.

Still, Stastny is 7th all-time in points per game and 12th all-time in playoff PPG. And he didn't have a short career.

Excellent power play player. Either at the half-boards or on the extended goal line (with his brother, Anton) seemed to always generate great chances.

This guy was a truly great offensive player.

 

[MXD]

By the way, here is a season-by-season summary of NHL assists per goal. It steadily rose throughout Stewart’s career, and by the end of it, it had gotten to more or less modern levels, but in his first few it was extremely low, which I can only assume benefited him in the points leaders:
*
http://www.quanthockey.com/TS/TS_AssistsPerGoal.php
*
Let me know what you think is appropriate and I’ll whip something up.

Was it possible that there was simply less assists due to forward pass rules, and that it doesn't have that much to do with "counting" techniques?

In which case, wouldn't that be not much more different than the other rule changes that might have affected both the game itself and the statistical output over the years?

 

[seventieslord]

Still, Stastny is 7th all-time in points per game and 12th all-time in playoff PPG. And he didn't have a short career.

Excellent power play player. Either at the half-boards or on the extended goal line (with his brother, Anton) seemed to always generate great chances.

This guy was a truly great offensive player.

*
See, the thing is, and I assume you have to recognize this, Stastny had almost the perfect conditions to post an excellent career PPG average. 23 years old when the clock struck 1980. Didn’t play in the NHL his younger developing seasons. Stepped right into the NHL in his prime. Was part of a generation of players who had poor longevity as a whole and though he outlasted almost everyone his age, he didn’t embarrass himself and once he was old enough to, he only got into 23 games in 2 seasons.
*
Whatever you think of something like VsX, I know you realize that something has to be done to account for higher scoring in different eras and attempt to uncover who were the most dominant scorers relative to their peers. When I read something like “he’s 7th all-time in PPG†I see “let’s take a step backwards and go back to raw unadjusted numbersâ€â€¦ why would we want to do that?

Was it possible that there was simply less assists due to forward pass rules, and that it doesn't have that much to do with "counting" techniques?

In which case, wouldn't that be not much more different than the other rule changes that might have affected both the game itself and the statistical output over the years?

*
Yes, it’s possible; however, it appears that assists per goal were on a drastic rise starting in 1926, a rise which appears completely unaffected by the 1930 rule change and which finally leveled off around 1940. So it really does appear to be a counting thing.
*
Even if the game was that different, I don’t think it affects what I’m saying that much though. Although finishing was still a skill, so was skating with the puck, so was advancing the puck towards the net, so was working the corners, and so was passing it. Just in a different way. The way goals have been awarded has never changed so comparing proficiency in goal scoring across areas is simple. But the way assists are recorded has clearly changed drastically. But does that change the fact that the time a forward spent in the opposition’s zone was spent attempting to score a goal or help a teammate score one? I honestly don’t think the lower assist numbers diminish the value of what the other players on the ice were doing or to help the cause. It just means that there was less statistical tracking of these important tasks.
*
I have no way to prove this but I am quite sure that if you and I watched a 1927 hockey game together and awarded assists the way they are awarded today, we would find A LOT more than 0.37 per goal. Does anyone really doubt that? Could 2/3 of goals really have been unassisted?
*
The biggest downfall is the potential inaccuracy in multiplying small numbers by a certain factor to “modernize†them. You could turn 10, 20 and 30 into 20, 40, and 60, but maybe the “actual†results are 16, 44 and 68. Of course there’s no way to truly know, and we have to simply understand and accept the limitations of extrapolation.

 

[Canadiens1958]

Alex Delvecchio

Since submitting my initial list, I really soured on Delvecchio --- and I had him in the mid-30ies. Can't tell why -- I just don'T have the impression he was a difference-maker at any point in his career. Something everybody on that list was, and quite a few that aren't up for voting at the moment. Bottom-line, if I had to cast my list again, he probably be something like 10 spots lower (roughly where I had Marty Barry, actually), at the very least.

Does any of the oldtimers has some insight on Delvecchio?

Going by last round, Cowley has the inside track at top spot, but my round 6 vote was cast in a hurry cause I was afraid to miss the deadline (which was extended in the end). AT the moment, my gut feeling regarding this group is probably Stewart on top... but 70ies made some very valid points which, IMO, cannot be ignored.

Alex Delvecchio was an excellent center, complete, reliable, equally effective at home and on the road whose game suited all opponents. Did not have the elite offensive skills of a Beliveau or Mikita or defensively like a Henri Richard or Dave Keon. Comfortably the third to fifth best center in the NHL for app. fifteen seasons.

http://www.hockey-reference.com/players/d/delveal01.html

Rarely injured or penalized. Excellent positioning.

 

[seventieslord]

Rarely injured or penalized. Excellent positioning.

This sounds just like Ron Francis. How do you compare the two?

 

[MXD]

*



*
Yes, it’s possible; however, it appears that assists per goal were on a drastic rise starting in 1926, a rise which appears completely unaffected by the 1930 rule change and which finally leveled off around 1940. So it really does appear to be a counting thing.
*
Even if the game was that different, I don’t think it affects what I’m saying that much though. Although finishing was still a skill, so was skating with the puck, so was advancing the puck towards the net, so was working the corners, and so was passing it. Just in a different way. The way goals have been awarded has never changed so comparing proficiency in goal scoring across areas is simple. But the way assists are recorded has clearly changed drastically. But does that change the fact that the time a forward spent in the opposition’s zone was spent attempting to score a goal or help a teammate score one? I honestly don’t think the lower assist numbers diminish the value of what the other players on the ice were doing or to help the cause. It just means that there was less statistical tracking of these important tasks.
*
I have no way to prove this but I am quite sure that if you and I watched a 1927 hockey game together and awarded assists the way they are awarded today, we would find A LOT more than 0.37 per goal. Does anyone really doubt that? Could 2/3 of goals really have been unassisted?
*
The biggest downfall is the potential inaccuracy in multiplying small numbers by a certain factor to “modernize†them. You could turn 10, 20 and 30 into 20, 40, and 60, but maybe the “actual†results are 16, 44 and 68. Of course there’s no way to truly know, and we have to simply understand and accept the limitations of extrapolation.

Well I can certainly see how ban on forward pass could effect the way the game is played, in a manner that would result in lower assists.

THIS SAID, yeah, the rise predates forward passing, and the said rise was pretty much consistent 1926-onwards. The game was possibly different, which COULD mean less real assists. But not at such a point.

 

[Canadiens1958]

Certain Points

*
See, the thing is, and I assume you have to recognize this, Stastny had almost the perfect conditions to post an excellent career PPG average. 23 years old when the clock struck 1980. Didn’t play in the NHL his younger developing seasons. Stepped right into the NHL in his prime. Was part of a generation of players who had poor longevity as a whole and though he outlasted almost everyone his age, he didn’t embarrass himself and once he was old enough to, he only got into 23 games in 2 seasons.
*
Whatever you think of something like VsX, I know you realize that something has to be done to account for higher scoring in different eras and attempt to uncover who were the most dominant scorers relative to their peers. When I read something like “he’s 7th all-time in PPG†I see “let’s take a step backwards and go back to raw unadjusted numbersâ€â€¦ why would we want to do that?


*
Yes, it’s possible; however, it appears that assists per goal were on a drastic rise starting in 1926, a rise which appears completely unaffected by the 1930 rule change and which finally leveled off around 1940. So it really does appear to be a counting thing.
*
Even if the game was that different, I don’t think it affects what I’m saying that much though. Although finishing was still a skill, so was skating with the puck, so was advancing the puck towards the net, so was working the corners, and so was passing it. Just in a different way. The way goals have been awarded has never changed so comparing proficiency in goal scoring across areas is simple. But the way assists are recorded has clearly changed drastically. But does that change the fact that the time a forward spent in the opposition’s zone was spent attempting to score a goal or help a teammate score one? I honestly don’t think the lower assist numbers diminish the value of what the other players on the ice were doing or to help the cause. It just means that there was less statistical tracking of these important tasks.
*
I have no way to prove this but I am quite sure that if you and I watched a 1927 hockey game together and awarded assists the way they are awarded today, we would find A LOT more than 0.37 per goal. Does anyone really doubt that? Could 2/3 of goals really have been unassisted?
*
The biggest downfall is the potential inaccuracy in multiplying small numbers by a certain factor to “modernize†them. You could turn 10, 20 and 30 into 20, 40, and 60, but maybe the “actual†results are 16, 44 and 68. Of course there’s no way to truly know, and we have to simply understand and accept the limitations of extrapolation.

Let's reword certain assumptions in play. Instead of accounting for higher scoring in different eras let's look at accounting for scoring differences across eras. Just looking at centers it has become rather clear that the distribution of center scoring in the same era was impacted by various factors such as whether the team rolled lines around three or four centers. You saw and supported this point when the Mikita example was provided. VsX has certain attributes but it fails to account for line rotations dictated by roster size and until it does it offers little insight.

Assists and the 1929-30 Forward Pass Rule. Most of the impact on assists happened later when the granting of assists was liberalized thru amendments to how assists were granted. The introduction of the Blue Line offside a few weeks into the season is deeply under appreciated here. Previously, as long as a player held the puck, no offside was possible. Eventually even passing the puck backwards could create offside situations at the Blue Line.

1927, counting and the old time machine argument. You are totally ignoring strategy and style of play having an impact on assists.

Specifically, the strategy of crashing or cluttering the opponents net did not exist. Offensive players could not risk creating offside situations since it was hard to clearly distinguish a shot from a pass. Likewise rebound assists. Wanting to avoid offsides had the downside that offensive players did not camp in prime rebound areas.

Instead of looking for multipliers, a simple appreciation for the nuances of an era are sufficient.