Round 2, Vote 13 (HOH Top Wingers)

[tarheelhockey]

For starters, just by the way they were described as players. The eye test.

No eye-test information about Barber has been shared. I can only assume it isn't available?

What we do have, however, is all-star results which reflect a consensus eye test. And Barber's record is better than Middleton's, even if just by a little.

Second, the difference in the help they had from linemates isn't even close. Not that Bourque didn't help, but he wasn't a linemate, and I'm guessing if you remove him, the metrics relating to leading one's team look even more in Middleton's favor.

Not really. Check the seasons yourself. There's only one season where both Middleton was a major figure for the Bruins but outscored by Bourque.

If one is always scoring 80 points with a 100 point linemate and the other is doing it with 60 point linemates, what does that tell you?

Likewise, what does it tell you that in his peak season Barber scored 112 points to Clarke's 119 (this being in 1976), while in his peak season Middleton scored 105 points to Barry Pederson's 116 (this being in 1984)? Think about how that converts for era.

What about years like '79 and '81 when Barber led those linemates rather than trailing them?

I mean, Middleton's already conclusively ahead by a metric, VsX, that is probably equally unfair to both players, and 4-odd points over 7 years isn't a small margin, either, as we've seen on those charts in this project.

Furthermore, even with Bourque considered as help (which he definitely was), Middleton's scoring record indicates he was much more of a catalyst. His "help" rating over his best 7 year period is 1.48 - that's better than guys like Goulet, Robitaille, Shanahan, Recchi, etc, from what I can remember when running those numbers for them. Barber's figure is 1.26 - which is Cournoyer territory.

You'll have to forgive me when I say I don't really trust these home-cooked metrics after seeing them poked full of holes throughout this project.

Middleton scored more by a not insiginificant degree, and he did so with an amount of offensive help that was signficantly less.

Middleton only scored more if you ignore the years where he didn't score much at all, such as his first 4 seasons in the league. For his career he's just below a PPG just like Barber, except he played more in a much higher scoring period.

"You've gotta be kidding me" is the appropriate response because the position that Barber holds an offensive edge is just that fringe. I wouldn't say that position is beneath serious discussion, but it's damn close.

Again, this kind of posturing isn't helpful.

 

[TheDevilMadeMe]

I mean, Middleton's already conclusively ahead by a metric, VsX, that is probably equally unfair to both players, and 4-odd points over 7 years isn't a small margin, either, as we've seen on those charts in this project.

Furthermore, even with Bourque considered as help (which he definitely was), Middleton's scoring record indicates he was much more of a catalyst. His "help" rating over his best 7 year period is 1.48 - that's better than guys like Goulet, Robitaille, Shanahan, Recchi, etc, from what I can remember when running those numbers for them. Barber's figure is 1.26 - which is Cournoyer territory.

Middleton scored more by a not insiginificant degree, and he did so with an amount of offensive help that was signficantly less.

"You've gotta be kidding me" is the appropriate response because the position that Barber holds an offensive edge is just that fringe. I wouldn't say that position is beneath serious discussion, but it's damn close.

Middleton also has a much better array of top 20 finishes than Barber.

What we do have, however, is all-star results which reflect a consensus eye test. And Barber's record is better than Middleton's, even if just by a little.

You're comparing a LW with a RW. Did you see how bad the competition for LW was in the late 70s/early 80s?

You'll have to forgive me when I say I don't really trust these home-cooked metrics after seeing them poked full of holes throughout this project.
.

What is "home cooked" supposed to mean?

VsX doesn't deal with the early 70s very well because of the Bruins breaking the system. That's about it. Other than that, it does exactly what it's supposed to - compare a player's performance vs a typical #2 scorer.

 

[Dennis Bonvie]

Middleton's " edge" really isn't that much considering that Barber had 31 SHG and Middleton had 25 SHG. Barber also lead the league in 1978-79 with 6 SHG. Middleton did have 1 top 5 Selke finish and a top 9 finish, while Barber best finish was 13th.

Gretzky & Mario had way more SHG than either. Pavel Bure & Peter Bondra also had more.

Were they better defensively?

 

[Dennis Bonvie]

Maybe it should. Was Middleton good enough defensively to bridge and then surpass the offensive gap?

I think Middleton was clearly better defensively during his prime.

Truthfully, they are very close for me. Barber was a couple spots ahead on my original list but at this point I'm think Middleton by a nose.

 

[Captain Bowie]

We are voting this weekend right?

I will definitely have Kovalchuk number 1.

Middleton, Naslund, Neely, Gainey and Noble will have a chance for top 4.

Mogilny, Tkachuk, Martin and Olmstead likely to round out the top 10.

Don't see myself voting for McDonald, Barber or Broadbent.

Ilya Kovalchuk

Bob Gainey
Rick Middleton
Markus Naslund
Cam Neely
Reg Noble

Alex Mogilny
Keith Tkachuk
Rick Martin
Bert Olmstead

Lanny McDonald
Bill Barber
Punch Broadbent

 

[TheDevilMadeMe]

Unless I'm missing something, Bill Barber is the easiest player this round to leave off my list. Just seems entirely unremarkable.

Leaning towards leaving Mogilny off my list too - his numbers just aren't good enough for a player who who basically all regular season offense.

Rick Martin - has the talent to be on the list, but with an injury-plagued career, I'm unconvinced he accomplished as much as Middleton or McDonald. Tarheel says he has the higher offense peak, but is a 6th place finish in scoring really that much more impressive than an 8th and a 10th when the later two players have more top 15 finishes, more top 20 finishes, and more relevant seasons in general? Also, he brought less without the puck than McDonald or Middleton, right?

 

[TheDevilMadeMe]

We are voting this weekend right?

I will definitely have Kovalchuk number 1.

Middleton, Naslund, Neely, Gainey and Noble will have a chance for top 4.

Mogilny, Tkachuk, Martin and Olmstead likely to round out the top 10.

Don't see myself voting for McDonald, Barber or Broadbent.

Ilya Kovalchuk

Bob Gainey
Rick Middleton
Markus Naslund
Cam Neely
Reg Noble

Alex Mogilny
Keith Tkachuk
Rick Martin
Bert Olmstead

Lanny McDonald
Bill Barber
Punch Broadbent

Why so down on Olmstead, again?

 

[tarheelhockey]

Middleton also has a much better array of top 20 finishes than Barber.

Fair enough, that's a point in Middleton's favor.

You're comparing a LW with a RW. Did you see how bad the competition for LW was in the late 70s/early 80s?

But it's not like competition suppressed Middleton all that much either. If you look at his best seasons:

1978 and earlier - Didn't get votes
1979 - Behind Lafleur and Bossy, sure, but he's also behind Bob MacMillan.
1980 - He's way guys like Blaine Stoughton, Al MacAdam, Dave Taylor.
1981 - Behind Taylor again, as well as Wayne Babych.
1982 - Behind Bossy, got the 2AS
1983 - Behind Bossy and Lanny McDonald
1984 - Behind Bossy and Kurri, which is a tough duo to break through
1985 and later - Didn't get votes

You could say he could have had an extra 1AS and a couple of 2ASs if Bossy had gotten out of the way during the early 80s. Which would bring him up to one 1AS and three 2ASs.

Barber had a 1AS and two 2ASs in real life, plus you could give him another 2AS in 1982 for Mark Messier. That would make the two of them tied. That still strikes me as being really close.

 

[seventieslord]

No eye-test information about Barber has been shared. I can only assume it isn't available?

of course it's available, but nothing I've ever read has been worth writing home about. He was more of an industrious, two-way player, not a dazzling individual talent.

What we do have, however, is all-star results which reflect a consensus eye test. And Barber's record is better than Middleton's, even if just by a little.

As TDMM said, comparing a LW to a RW is really unfair.

Not really. Check the seasons yourself. There's only one season where both Middleton was a major figure for the Bruins but outscored by Bourque.

Likewise, what does it tell you that in his peak season Barber scored 112 points to Clarke's 119 (this being in 1976), while in his peak season Middleton scored 105 points to Barry Pederson's 116 (this being in 1984)? Think about how that converts for era.

What about years like '79 and '81 when Barber led those linemates rather than trailing them?

OK. Nuts to me for making a few assumptions and generalizations when I lacked the resources to make a better post, and kudos for pouncing on that. But you're still looking at isolated situations rather than the big picture. The big picture quite clearly shows that Middleton was scoring points with players considerably less skilled, on average, than Barber was. He came by his points more honestly.

You'll have to forgive me when I say I don't really trust these home-cooked metrics after seeing them poked full of holes throughout this project.

Whatever, I don't think that comment is really warranted, but ok, look at hf's adjusted points, which is a completely third party creation. Middleton has five seasons with 73+ adjusted points. Barber has one. Not. Even. Close.

Middleton only scored more if you ignore the years where he didn't score much at all, such as his first 4 seasons in the league. For his career he's just below a PPG just like Barber, except he played more in a much higher scoring period.

and typically we do ignore that, don't we? We want to know how a player played at his best. The 7-year primes we tend to look at do a very good job of that.

 

[seventieslord]

Fair enough, that's a point in Middleton's favor.



But it's not like competition suppressed Middleton all that much either. If you look at his best seasons:

1978 and earlier - Didn't get votes
1979 - Behind Lafleur and Bossy, sure, but he's also behind Bob MacMillan.
1980 - He's way guys like Blaine Stoughton, Al MacAdam, Dave Taylor.
1981 - Behind Taylor again, as well as Wayne Babych.
1982 - Behind Bossy, got the 2AS
1983 - Behind Bossy and Lanny McDonald
1984 - Behind Bossy and Kurri, which is a tough duo to break through
1985 and later - Didn't get votes

You could say he could have had an extra 1AS and a couple of 2ASs if Bossy had gotten out of the way during the early 80s. Which would bring him up to one 1AS and three 2ASs.

Barber had a 1AS and two 2ASs in real life, plus you could give him another 2AS in 1982 for Mark Messier. That would make the two of them tied. That still strikes me as being really close.

In 1979, MacMillan had 108 points and was his usual excellent two-way self. Middleton should not have been expected to get votes.

In 1980, Stoughton had 100+ points, and MacAdam and Taylor matched Middleton in points too. MacAdam was intangibles-loaded and Taylor needed only 61 games to score that much. That's a good handful of great seasons there.

In 1981, Taylor had 112 points, and Babych was not that far behind, and led the NHL in even strength goals.

In 1983, McDonald had a remarkable 66 goals.

These are all great seasons, whether they were by great players or not. The point about a guy like Rick Martin (and to a lesser extent, Barber) was that he was facing poor competition at the LW position, not just by looking at the names, but the seasons those players had that he had to exceed to get onto the all-star team... Cashman, Vail, Vickers, Lefley... then, to a lesser extent, Shutt, Simmer, Payne, Gillies... look at the seasons guys like that were having. Heck, just look at the point totals of the top-10 LW and RW players from 1972 to 1985.

the problem with judging players too much by all-star teams as opposed to league-wide metrics is that you are drastically reducing the sample of players they're being compared to, and increasing the likelihood that your result is going to be over-influenced by noise.

 

[tarheelhockey]

^ off the top of my head, Barber was 2AS when Charlie Simmer missed like 15 games and still led the league in goals. How is that different, let alone worse than finishing behind '83 McDonald?

Will try to get back to the other points after NYE hangover.

 

[TheDevilMadeMe]

Fair enough, that's a point in Middleton's favor.



But it's not like competition suppressed Middleton all that much either. If you look at his best seasons:

1978 and earlier - Didn't get votes
1979 - Behind Lafleur and Bossy, sure, but he's also behind Bob MacMillan.
1980 - He's way guys like Blaine Stoughton, Al MacAdam, Dave Taylor.
1981 - Behind Taylor again, as well as Wayne Babych.
1982 - Behind Bossy, got the 2AS
1983 - Behind Bossy and Lanny McDonald
1984 - Behind Bossy and Kurri, which is a tough duo to break through
1985 and later - Didn't get votes

You could say he could have had an extra 1AS and a couple of 2ASs if Bossy had gotten out of the way during the early 80s. Which would bring him up to one 1AS and three 2ASs.

Barber had a 1AS and two 2ASs in real life, plus you could give him another 2AS in 1982 for Mark Messier. That would make the two of them tied. That still strikes me as being really close.

There's a very decent chance that's 3 consecutive 1st Team All-Stars facing the competition Barber faced, isn't it? Bossy and Kurri are easy, but there's also a good case that Lanny McDonald was better than Rick Martin or Steve Shutt (who were Barber's main competition for AS LW).

 

[TheDevilMadeMe]

I guess we're voting soon.

This is where I'm at now:

Will definitely be in my top 4

Ilya Kovalchuk - He's the best scorer by a wide enough margin this round, where I think we are finally at a point where it overcomes his faults as a player

Bert Olmstead - the best of the high intangibles players this round, I think.

Have a decent shot at my top 4

Rick Middleton - compared to the rest of the candidates, he seems average to above average in offense and defense, regular season and playoffs, peak and career. And at this point in the process, simply not being below average in any category is a plus.

Keith Tkachuk - He'd be a shoo-in for the top 4 if he weren't so bad in the playoffs for the final 2/3 of his career. One of the top scoring resumes this round (regular season obviously), and a physically intimidating player, albeit one who sometimes took selfish and lazy penalties. Third best LW of the late 90s behind Kariya and LeClair, and has a massive advantage in career value over Barber and Martin. Was actually considered something of a big game player early in his career, believe it or not. As of the age of 25, he had scored a very respectable 17 goals in his first 32 playoff games for a bad Winnipeg/Phoenix team that was a perennial first round loser. Add in a very solid performance at the 1996 World Cup, and you can see why he was made one of the league's highest paid players in St. Louis. Failed miserably in the playoffs as one of the league's highest paid players, and that's what we remember. So basically, if your best forward is Keith Tkachuk, you probably aren't going anywhere in the playoffs. But at this point, being the best (regular season) forward on your team for the better part of 15 years for teams that usually make the playoffs is pretty impressive, even if you usually suck when you get there. I think a big part of the criticism for Tkachuk is that he was bad... for one of the league's highest paid players.

Lanny McDonald - kind of the forgotten man this round. As a scorer, he seems very close to Middleton. Had something of a gritty game too. Lacks Middleton's defense (I think, am I wrong?) but has a more consistently strong playoff record and was known as a great leader.

Still not sure what I think - convince me why I'm underrating them

Punch Broadbent - his regular season resume is pretty unremarkable - a single scoring title and a big dropoff the rest of his career. Still, he was excellent both defensively and physically and known as a "money player." Definitely needs to be below Olmstead, but how far below?

Reg Noble - another guy who was something of an intangibles king and whose scoring resume would have him clearly below Olmstead - but how far below? I think he was more consistently good in the regular season than Broadbent, despite lacking that one great season. Don't think his playoff record is quite as strong as Broadbent's though.

Cam Neely - I didn't even have him in my top 80. Reconsidering after overpass showed how consistently strong his plus-minus numbers were compared to his teammates when he did play. I think we've reached a point where Neely is probably top 4 this round... on a per-game basis. But man, did he ever miss lots of games. And some of it was his own fault - such as aggravating his knee injury in an ill-conceived attempt at getting revenge on Ulf Samuelsson. That kamikaze style of play that made him so effective on the ice is probably a big reason he was off it so often.

Markus Naslund - Anyone else paying attention to the Naslund vs Palffy thread? Such a short time as an elite player. Definitely below Kovalchuk for me. Strongly considering having him below Tkachuk too (I had them close to back to back on my original list), which would have him out of my top 4 since Tkachuk isn't a guarantee to make it. FWIW, his 10 year VsX is almost identical to Tkachuk, and unlike Keith, Naslund was a softy. Only a slightly better 7 year VsX too. And honestly, Naslund's big game resume is probably even worse than Tkachuk's. Still, what a strong 3 year stretch.

Will probably not make my top 4, but could if there are really good final arguments

Rick Martin - excellent goal scorer, but little else. Top 20 points finishes of 6, 10, 15, 17 are pretty unimpressive for a guy who didn't bring much without the puck. I do realize, however, that injuries held him back.

Bob Gainey - Still don't see why he should necessarily be here before George Armstrong, etc, but at least he has something special about him, which I don't think the guys in my next category really have.

Zero chance at my top 4

Alexander Mogilny - regular season points finishes of 7, 9, 15, 15, regular season goals finishes of 1, 3, 6, 18, 20. Pretty soft and inconsistent defensively. Generally bad in the playoffs. Should be a significant step behind Kovalchuk and even Tkachuk. The ultimate "plays hard for a new contract then slacks off" player. As I said earlier - 2 very good seasons, 2 pretty good ones, and a lot of mediocrity in between.

Bill Barber - A single 4th place finish in points playing next to Hart winner Bobby Clarke - his next highest finish would be 18th. Pretty good without the puck, but we have coach's polls covering his prime very well, and he never received a single vote as the best in the league at anything. Top 100 winger of all time? Definitely. Top 60? Definitely not.

_________

Biggest place I seem to differ with tarheel is with Martin and Barber - I just don't find All-Star voting at LW during their time to be all that impressive, I guess.

 

[Dennis Bonvie]

I guess we're voting soon.

This is where I'm at now:

Will definitely be in my top 4

Ilya Kovalchuk - He's the best scorer by a wide enough margin this round, where I think we are finally at a point where it overcomes his faults as a player

Bert Olmstead - the best of the high intangibles players this round, I think.

Have a decent shot at my top 4

Rick Middleton - compared to the rest of the candidates, he seems average to above average in offense and defense, regular season and playoffs, peak and career. And at this point in the process, simply not being below average in any category is a plus.

Keith Tkachuk - He'd be a shoo-in for the top 4 if he weren't so bad in the playoffs for the final 2/3 of his career. One of the top scoring resumes this round (regular season obviously), and a physically intimidating player, albeit one who sometimes took selfish and lazy penalties. Third best LW of the late 90s behind Kariya and LeClair, and has a massive advantage in career value over Barber and Martin. Was actually considered something of a big game player early in his career, believe it or not. As of the age of 25, he had scored a very respectable 17 goals in his first 32 playoff games for a bad Winnipeg/Phoenix team that was a perennial first round loser. Add in a very solid performance at the 1996 World Cup, and you can see why he was made one of the league's highest paid players in St. Louis. Failed miserably in the playoffs as one of the league's highest paid players, and that's what we remember. So basically, if your best forward is Keith Tkachuk, you probably aren't going anywhere in the playoffs. But at this point, being the best (regular season) forward on your team for the better part of 15 years for teams that usually make the playoffs is pretty impressive, even if you usually suck when you get there. I think a big part of the criticism for Tkachuk is that he was bad... for one of the league's highest paid players.

Lanny McDonald - kind of the forgotten man this round. As a scorer, he seems very close to Middleton. Had something of a gritty game too. Lacks Middleton's defense (I think, am I wrong?) but has a more consistently strong playoff record and was known as a great leader.

Still not sure what I think - convince me why I'm underrating them

Punch Broadbent - his regular season resume is pretty unremarkable - a single scoring title and a big dropoff the rest of his career. Still, he was excellent both defensively and physically and known as a "money player." Definitely needs to be below Olmstead, but how far below?

Reg Noble - another guy who was something of an intangibles king and whose scoring resume would have him clearly below Olmstead - but how far below? I think he was more consistently good in the regular season than Broadbent, despite lacking that one great season. Don't think his playoff record is quite as strong as Broadbent's though.

Cam Neely - I didn't even have him in my top 80. Reconsidering after overpass showed how consistently strong his plus-minus numbers were compared to his teammates when he did play. I think we've reached a point where Neely is probably top 4 this round... on a per-game basis. But man, did he ever miss lots of games. And some of it was his own fault - such as aggravating his knee injury in an ill-conceived attempt at getting revenge on Ulf Samuelsson. That kamikaze style of play that made him so effective on the ice is probably a big reason he was off it so often.

Markus Naslund - Anyone else paying attention to the Naslund vs Palffy thread? Such a short time as an elite player. Definitely below Kovalchuk for me. Strongly considering having him below Tkachuk too (I had them close to back to back on my original list), which would have him out of my top 4 since Tkachuk isn't a guarantee to make it. FWIW, his 10 year VsX is almost identical to Tkachuk, and unlike Keith, Naslund was a softy. Only a slightly better 7 year VsX too. And honestly, Naslund's big game resume is probably even worse than Tkachuk's. Still, what a strong 3 year stretch.

Will probably not make my top 4, but could if there are really good final arguments

Rick Martin - excellent goal scorer, but little else. Top 20 points finishes of 6, 10, 15, 17 are pretty unimpressive for a guy who didn't bring much without the puck. I do realize, however, that injuries held him back.

Bob Gainey - Still don't see why he should necessarily be here before George Armstrong, etc, but at least he has something special about him, which I don't think the guys in my next category really have.

Zero chance at my top 4

Alexander Mogilny - regular season points finishes of 7, 9, 15, 15, regular season goals finishes of 1, 3, 6, 18, 20. Pretty soft and inconsistent defensively. Generally bad in the playoffs. Should be a significant step behind Kovalchuk and even Tkachuk. The ultimate "plays hard for a new contract then slacks off" player. As I said earlier - 2 very good seasons, 2 pretty good ones, and a lot of mediocrity in between.

Bill Barber - A single 4th place finish in points playing next to Hart winner Bobby Clarke - his next highest finish would be 18th. Pretty good without the puck, but we have coach's polls covering his prime very well, and he never received a single vote as the best in the league at anything. Top 100 winger of all time? Definitely. Top 60? Definitely not.

_________

Biggest place I seem to differ with tarheel is with Martin and Barber - I just don't find All-Star voting at LW during their time to be all that impressive, I guess.

Rick Martin was a pure goal scorer. Taking his top 20 finishes in points doesn't tell the story. Martin is 12th all-time in goals per game, just ahead of Esposito, Rocket Ricahrd, Cam Neely and Marcel Dionne. Yes, injuries cut his career short. But, as a goal scorer/prime guy, he'll be in my top 4.

Reg Noble will be in my top 4 also. Seems like a poor man's Dit Clapper. Not as gifted, but more physical. Extremely tough, excellent defensively, good offensive numbers and played for a long time for that era. Led the NHL in assists in its first season, was also 3rd in goals & points. Played on 3 Cup winners.

 

[MXD]

Still not sure what I think - convince me why I'm underrating them

Punch Broadbent - his regular season resume is pretty unremarkable - a single scoring title and a big dropoff the rest of his career. Still, he was excellent both defensively and physically and known as a "money player." Definitely needs to be below Olmstead, but how far below?

I think that's a somewhat optimistic view of Broadbent, in view of the fact that one could claim Broadbent was always the worst player on the ice at a given moment on a given game for the Sens, including the D-Men.

 

[Canadiens1958]

Points to Ponder

Middleton also has a much better array of top 20 finishes than Barber.



You're comparing a LW with a RW. Did you see how bad the competition for LW was in the late 70s/early 80s?



What is "home cooked" supposed to mean?

VsX doesn't deal with the early 70s very well because of the Bruins breaking the system. That's about it. Other than that, it does exactly what it's supposed to - compare a player's performance vs a typical #2 scorer.

But this is a Top Winger project contrary to two position specific project Top LWs and Top RWs.

Home cooked as in a metric that only matters here and nowhere else.

Is it really a question of the Bruins breaking the system? 1970s Bruins did not break any other metric that has validity, yet they broke VsX? Did Gretzky and the Oilers break VsX, early fifties Howe and the Red Wings?

So why specifically does it not work for the 1970s and why was it used in the project?

 

[TheDevilMadeMe]

But this is a Top Winger project contrary to two position specific project Top LWs and Top RWs.

Home cooked as in a metric that only matters here and nowhere else.

Is it really a question of the Bruins breaking the system? 1970s Bruins did not break any other metric that has validity, yet they broke VsX? Did Gretzky and the Oilers break VsX, early fifties Howe and the Red Wings?

So why specifically does it not work for the 1970s and why was it used in the project?

VsX is based on Vs2 - it tries to compare players to the #2 scorer in the NHL, but in the rare seasons when #2 is a statistical outlier, it uses a different standard. But what standard can it use in (for example) 1971?

Points
1. Phil Esposito*-BOS 152
2. Bobby Orr*-BOS 139
3. John Bucyk*-BOS 116
4. Ken Hodge-BOS 105
5. Bobby Hull*-CBH 96
6. Norm Ullman*-TOR 85
7. Wayne Cashman-BOS 79
8. John McKenzie-BOS 77
9. Fred Stanfield-BOS 76
Dave Keon*-TOR 76
Jean Beliveau*-MTL 76

If you use Orr, Bucyk, or Hodge as a standard, you are decimating the scores of the field. If you use Hull or Ullman, you are giving Bucyk, Hodge, and co unrealistically high scorers. Only the early 70s Bruins have this issue of their secondary players looking so dominant in the scoring tables. 1971 is the worst year of this, but there are others.

In actuality, VsX uses a standard of 100 points (the average of the top 10 scorers) to compare everyone else to, which is compromise that I don't think serves anyone well. In the early 70s, you are basically dealing with 2 sets of data - the Bruins and the non-Bruins, and in my opinion, trying to come up with one standard does do either set justice.

Whereas, for the 80s Oilers, generally only Gretzky was a statistical outlier. Take 1983-84

Points
1. Wayne Gretzky*-EDM 205
2. Paul Coffey*-EDM 126
3. Michel Goulet*-QUE 121
4. Peter Stastny*-QUE 119
5. Mike Bossy*-NYI 118
6. Barry Pederson-BOS 116
7. Jari Kurri*-EDM 113
8. Bryan Trottier*-NYI 111
9. Bernie Federko*-STL 107
10. Rick Middleton-BOS 105

Do you use Paul Coffey (the typical #2 and what VsX actually uses) as a standard or do you use Michel Goulet? Considering how close they were in scoring, it really doesn't make much of a difference.

 

[Morgoth Bauglir]

VsX is based on Vs2 - it tries to compare players to the #2 scorer in the NHL, but in the rare seasons when #2 is a statistical outlier, it uses a different standard. But what standard can it use in (for example) 1971?

Points
1. Phil Esposito*-BOS 152
2. Bobby Orr*-BOS 139
3. John Bucyk*-BOS 116
4. Ken Hodge-BOS 105
5. Bobby Hull*-CBH 96
6. Norm Ullman*-TOR 85
7. Wayne Cashman-BOS 79
8. John McKenzie-BOS 77
9. Fred Stanfield-BOS 76
Dave Keon*-TOR 76
Jean Beliveau*-MTL 76

If you use Orr, Bucyk, or Hodge as a standard, you are decimating the scores of the field. If you use Hull or Ullman, you are giving Bucyk, Hodge, and co unrealistically high scorers. Only the early 70s Bruins have this issue of their secondary players looking so dominant in the scoring tables. 1971 is the worst year of this, but there are others.

In actuality, VsX uses a standard of 100 points (the average of the top 10 scorers) to compare everyone else to, which is compromise that I don't think serves anyone well. In the early 70s, you are basically dealing with 2 sets of data - the Bruins and the non-Bruins, and in my opinion, trying to come up with one standard does do either set justice.

Whereas, for the 80s Oilers, generally only Gretzky was a statistical outlier.

Points
1. Wayne Gretzky*-EDM 205
2. Paul Coffey*-EDM 126
3. Michel Goulet*-QUE 121
4. Peter Stastny*-QUE 119
5. Mike Bossy*-NYI 118
6. Barry Pederson-BOS 116
7. Jari Kurri*-EDM 113
8. Bryan Trottier*-NYI 111
9. Bernie Federko*-STL 107
10. Rick Middleton-BOS 105

Do you use Paul Coffey (the typical #2 and what VsX actually uses) as a standard or do you use Michel Goulet? Considering how close they were in scoring, it really doesn't make much of a difference.

So what, do you think, are possible solutions to the conundrum? I don't have the math background to even contemplate trying.

 

[TheDevilMadeMe]

So what, do you think, are possible solutions to the conundrum? I don't have the math background to even contemplate trying.

1) Long term, using two different standards for the Bruins and non-Bruins might have to happen. If you completely ignore all the Bruins, Norm Ullman would be the standard used for non-Bruins. This might be too generous to the non-Bruins, however. Bruins would be compared to Bobby Orr, the actual #2 scorer behind Phil Esposito. It's also really messy, considering one of the points of VsX is to have a single standard that all players can be compared to.

2) For now, just realize that VsX like every other method of measuring scoring across different eras has its flaws, recognize what those flaws are, and make mental adjustments. For early 70s players, I would mentally adjust their 7 and 10 year scores up a few points, and pay more attention to top 10/top 20 finishes than I would for other players. This really only affects Rick Martin this round (only one of these guys in his prime in the first half of the 1970s), and I purposely only talked about his top 20 finishes, not his VsX when discussing him in my big post.

 

[Canadiens1958]

Not So

VsX is based on Vs2 - it tries to compare players to the #2 scorer in the NHL, but in the rare seasons when #2 is a statistical outlier, it uses a different standard. But what standard can it use in (for example) 1971?

Points
1. Phil Esposito*-BOS 152
2. Bobby Orr*-BOS 139
3. John Bucyk*-BOS 116
4. Ken Hodge-BOS 105
5. Bobby Hull*-CBH 96
6. Norm Ullman*-TOR 85
7. Wayne Cashman-BOS 79
8. John McKenzie-BOS 77
9. Fred Stanfield-BOS 76
Dave Keon*-TOR 76
Jean Beliveau*-MTL 76

If you use Orr, Bucyk, or Hodge as a standard, you are decimating the scores of the field. If you use Hull or Ullman, you are giving Bucyk, Hodge, and co unrealistically high scorers. Only the early 70s Bruins have this issue of their secondary players looking so dominant in the scoring tables. 1971 is the worst year of this, but there are others.

In actuality, VsX uses a standard of 100 points (the average of the top 10 scorers) to compare everyone else to, which is compromise that I don't think serves anyone well. In the early 70s, you are basically dealing with 2 sets of data - the Bruins and the non-Bruins, and in my opinion, trying to come up with one standard does do either set justice.

Whereas, for the 80s Oilers, generally only Gretzky was a statistical outlier.

Points
1. Wayne Gretzky*-EDM 205
2. Paul Coffey*-EDM 126
3. Michel Goulet*-QUE 121
4. Peter Stastny*-QUE 119
5. Mike Bossy*-NYI 118
6. Barry Pederson-BOS 116
7. Jari Kurri*-EDM 113
8. Bryan Trottier*-NYI 111
9. Bernie Federko*-STL 107
10. Rick Middleton-BOS 105

Do you use Paul Coffey (the typical #2 and what VsX actually uses) as a standard or do you use Michel Goulet? Considering how close they were in scoring, it really doesn't make much of a difference.

Please review the data with a more critical eye.

The bolded all played on teams that used a three line rotation. The Canadiens and Leafs used a four line rotation:

Leafs:
http://www.hockey-reference.com/teams/TOR/1971.html

Canadiens:
http://www.hockey-reference.com/teams/MTL/1971.html

The problem with VsX goes back to the fifties when first the Canadiens then the Leafs went to four line rotations.

So the issue is not the 1970s Bruins.

 

[Canadiens1958]

This Round

1) Long term, using two different standards for the Bruins and non-Bruins might have to happen. If you completely ignore all the Bruins, Norm Ullman would be the standard used for non-Bruins. This might be too generous to the non-Bruins, however. Bruins would be compared to Bobby Orr, the actual #2 scorer behind Phil Esposito. It's also really messy, considering one of the points of VsX is to have a single standard that all players can be compared to.

2) For now, just realize that VsX like every other method of measuring scoring across different eras has its flaws, recognize what those flaws are, and make mental adjustments. For early 70s players, I would mentally adjust their 7 and 10 year scores up a few points, and pay more attention to top 10/top 20 finishes than I would for other players. This really only affects Rick Martin this round (only one of these guys in his prime in the first half of the 1970s), and I purposely only talked about his top 20 finishes, not his VsX when discussing him in my big post.

But this round is affected by all the preceeding rounds.

Tossing the VsX metric would be the best idea to come out of this project. Adjusting
across era for roster size and line rotations - 3 vs 2, 4 vs 3 might work but doubt that the required work would be undertaken.

 

[TheDevilMadeMe]

Please review the data with a more critical eye.

The bolded all played on teams that used a three line rotation. The Canadiens and Leafs used a four line rotation:

Leafs:
http://www.hockey-reference.com/teams/TOR/1971.html

Canadiens:
http://www.hockey-reference.com/teams/MTL/1971.html

The problem with VsX goes back to the fifties when first the Canadiens then the Leafs went to four line rotations.

So the issue is not the 1970s Bruins.

To the extent that this is a real issue, it would affect all ways of measuring offense, not just VsX.

But this round is affected by all the preceeding rounds.

Tossing the VsX metric would be the best idea to come out of this project. Adjusting
across era for roster size and line rotations - 3 vs 2, 4 vs 3 might work but doubt that the required work would be undertaken.

I disagree with tossing what I think the single best method of comparing regular season offense between eras that has been created, just because there may be room for improvement.

 

[tarheelhockey]

and typically we do ignore that, don't we? We want to know how a player played at his best. The 7-year primes we tend to look at do a very good job of that.

When did this become a consensus decision? It seems to me that there's a VERY wide spectrum here in regard to how people evaluate peak vs prime vs total career. I suspect that most people take them in balance, rather than ignoring one or the other.

 

[tarheelhockey]

There's a very decent chance that's 3 consecutive 1st Team All-Stars facing the competition Barber faced, isn't it? Bossy and Kurri are easy, but there's also a good case that Lanny McDonald was better than Rick Martin or Steve Shutt (who were Barber's main competition for AS LW).

As mentioned earlier -- Barber was blocked from a 1AS by a Charlie Simmer spike season in 1980 and from a 2AS by Mark Messier in 1982.

I originally said I had Barber and Middleton so close that I could easily flip them, and I think the case for Middleton has been made well enough to make me do that. But they really are close.

 

[seventieslord]

When did this become a consensus decision? It seems to me that there's a VERY wide spectrum here in regard to how people evaluate peak vs prime vs total career. I suspect that most people take them in balance, rather than ignoring one or the other.

No one is saying we should outright ignore anything. But when you make statements like "For his career he's just below a PPG just like Barber, except he played more in a much higher scoring period" then you're essentially considering it just as important/meaningful that barber outscored Middleton in their respective 9th-13th best seasons, as it was that Middleton outscored barber in their respective best five. And you can't really think that, can you?

There are reasons to put barber ahead. It's defensible. If you get questioned on it, just say you greatly value cups and defense. But offensively? That was why this conversation started - I piped in when you said barber had an offensive edge. I mean, c'mon. Adjusted points, VsX, top 20 points finishes, importance to team offense, eye test, divisional factors (which you were kind enough to calculate for us but then unfortunately misinterpreted) - he's ahead in every way, usually by a lot, too. Then the linemate issue just puts a massive conclusive stamp on it all for good measure. Barber's not better offensively; there's just nothing to support that. One can still claim he's better overall if they really value defense and cups above all else, though.

(Looking back through our exchange, I think maybe when you brought up all-star teams in response to the eye test, that was when you forgot that I was referring to the issue of offense specifically. I never realized the goalposts were moved and kept on talking about offense)

...And I know I may be beating a dead horse by now, with you saying you've changed your mind.