The Hockey News - Top 50 Players of All-Time By Franchise

[quoipourquoi]

I dig the new magazine format. I'm moving right now, so all of the boxes of 80s/90s/early-00s newspapers are probably annoying Mrs. Quoipourquoi right about now.

 

[seventieslord]

It's been at this new size since 2009. It was only slightly larger from 07-09. 2007 is when it went full glossy. From 05 to 07 it was newspaper stock and larger. From 02 to 07 it was larger still. Prior to 2002 it as the newspaper stock we all probably remember best.

 

[decma]

I've had a chance to read the issue a little more closely the last couple days and have two quick thoughts.

In the Introduction Kay refers to the list as being of "the 1,500 best players of all-time." Surely he doesn't believe that this list represents that?

In the Canucks section I was suprised Snepts was so high (higher than Brodeur, Gradin, and MacLean, for example). Obviously hard to compare a stay-at-home, goon defenceman with goalies or forwards, but curious as to the reasoning behind having Snepts so high. Butcher played a similar role over a comparable length of time a decade later, but is much lower on list (38th).

But that is a minor quibble. Several times my initial reaction was that a placement was way off, but upon further reflection found myself in general agreement. A testament to the work that went into this.

 

[seventieslord]

I've had a chance to read the issue a little more closely the last couple days and have two quick thoughts.

In the Introduction Kay refers to the list as being of "the 1,500 best players of all-time." Surely he doesn't believe that this list represents that?

In the Canucks section I was suprised Snepts was so high (higher than Brodeur, Gradin, and MacLean, for example). Obviously hard to compare a stay-at-home, goon defenceman with goalies or forwards, but curious as to the reasoning behind having Snepts so high. Butcher played a similar role over a comparable length of time a decade later, but is much lower on list (38th).

But that is a minor quibble. Several times my initial reaction was that a placement was way off, but upon further reflection found myself in general agreement. A testament to the work that went into this.

No, Kay definitely does not think these are the 1500 best players, nor is it even 1500 players because of the dozens who show up 2, 3, 4 and even 5 times. I agree it was maybe not the best way to state it on his part.

Snepsts is a very rich man's Butcher. Butcher became a quite useful and utilized player as a St. Louis Blue, but in Vancouver was a depth player until his last two seasons, when he finally topped 20 minutes, but overall played 18.5 minutes per game for them. The team was 18% worse than average during his time there.

In comparison, Snepsts:

- played 21.2 minutes a game for the Canucks including his 2nd stint there late in his career (basically was a #2/3 for them on average compared to a #5/6 on average)
- played for them for 28% more regular season games and 2.6 times as many playoff games
- the team was only 12% worse than average in his time there, as opposed to 18% worse
- led the team to the finals in 1982 (and I don't say that lightly, see below)
- represented the team in two all-star games
- most importantly, was named the Babe Pratt trophy winner four times as the Canucks' best defenseman (one of four defensemen to earn the award four times, the others being Ohlund, Lumme and Lidster)

To refer to him as a goon defenseman doesn't do him justice, because he was a very important defensive player for a long time. I mean, you seem generally aware of what type of player he was, but unaware of just how good he was at it.

As for the other names you dropped, I know at first glance it might seem strange that a defensive defenseman is ahead of a scoring forward and starting goalies, but:

- Gradin only played 78% as many games and 0.9 PPG back then is not THAT special - he was only the team's best forward twice and went to one ASG; Snepsts has him doubled in both areas

- McLean, aside from 1989, 1992 and the 1994 playoffs, posted consistently below average numbers - often very far below. He's a little like vernon in that he does well by "accomplishment counting" but if you look at his everyday level of play, it wasn't that great. He put his time in, though, which helps his case.

- Brodeur was by no means a bad goalie, but his decline due to age started midway through his Vancouver career. He had one excellent season (and playoff), one very good one before that, and otherwise struggled to keep this mediocre team afloat. You could say 1982 is a higher peak than Snepsts and you wouldn't necessarily be wrong*, but I don't think his 1984-1987 seasons as a mediocre-to-poor starter do his full body of work any favors.




*even if Snepsts is a major reason Brodeur was able to post those stats and advance that far in the playoffs. Snepsts in that season was the #2 in total TOI for Vancouver behind McCarthy, who got tons of PP TOI but Snepsts was the runaway leader in ESTOI and SHTOI and here's the kicker - McCarthy, his only real TOI competition on the team, missed the entire playoffs, leaving Snepsts as the overall #1 defenseman through four playoff rounds, without a second of PP time padding that total.

 

[Howie Hodge]

The Hockey News was the bomb!

Shame it became blogs and analytics rather than well written and insightful articles.

As a youngster I used to rush to my Elementary Schools Library to get each weeks new copy of THN.

The Macon Whoopees, The Suncoast Suns, The Seattle Totems, The New Wesminster Bruins, The Denver Spurs, The Beauce Jaros...

Andy Hebenton - Clay Hebenton, Willie Marshall, Len Thornson, Keke Mortson, Joe Hardy.

I'm sorry to get off topic; as you were saying....

 

[Thenameless]

Thank you for the write-up seventieslord, I rank Harold Snepsts quite highly myself.

My 6 Canucks defensemen (not nearly the best, but pretty good ones that I liked): Snepsts, Ohlund, Jeff Brown, Jonanovski, Brian Glynn (yes, just because of that magical '94 run), Bieksa

 

[decma]

No, Kay definitely does not think these are the 1500 best players, nor is it even 1500 players because of the dozens who show up 2, 3, 4 and even 5 times. I agree it was maybe not the best way to state it on his part.

Snepsts is a very rich man's Butcher. Butcher became a quite useful and utilized player as a St. Louis Blue, but in Vancouver was a depth player until his last two seasons, when he finally topped 20 minutes, but overall played 18.5 minutes per game for them. The team was 18% worse than average during his time there.

In comparison, Snepsts:

- played 21.2 minutes a game for the Canucks including his 2nd stint there late in his career (basically was a #2/3 for them on average compared to a #5/6 on average)
- played for them for 28% more regular season games and 2.6 times as many playoff games
- the team was only 12% worse than average in his time there, as opposed to 18% worse
- led the team to the finals in 1982 (and I don't say that lightly, see below)
- represented the team in two all-star games
- most importantly, was named the Babe Pratt trophy winner four times as the Canucks' best defenseman (one of four defensemen to earn the award four times, the others being Ohlund, Lumme and Lidster)

To refer to him as a goon defenseman doesn't do him justice, because he was a very important defensive player for a long time. I mean, you seem generally aware of what type of player he was, but unaware of just how good he was at it.

As for the other names you dropped, I know at first glance it might seem strange that a defensive defenseman is ahead of a scoring forward and starting goalies, but:

- Gradin only played 78% as many games and 0.9 PPG back then is not THAT special - he was only the team's best forward twice and went to one ASG; Snepsts has him doubled in both areas

- McLean, aside from 1989, 1992 and the 1994 playoffs, posted consistently below average numbers - often very far below. He's a little like vernon in that he does well by "accomplishment counting" but if you look at his everyday level of play, it wasn't that great. He put his time in, though, which helps his case.

- Brodeur was by no means a bad goalie, but his decline due to age started midway through his Vancouver career. He had one excellent season (and playoff), one very good one before that, and otherwise struggled to keep this mediocre team afloat. You could say 1982 is a higher peak than Snepsts and you wouldn't necessarily be wrong*, but I don't think his 1984-1987 seasons as a mediocre-to-poor starter do his full body of work any favors.


*even if Snepsts is a major reason Brodeur was able to post those stats and advance that far in the playoffs. Snepsts in that season was the #2 in total TOI for Vancouver behind McCarthy, who got tons of PP TOI but Snepsts was the runaway leader in ESTOI and SHTOI and here's the kicker - McCarthy, his only real TOI competition on the team, missed the entire playoffs, leaving Snepsts as the overall #1 defenseman through four playoff rounds, without a second of PP time padding that total.

Where are you getting the ice time stats for Snepsts and Butcher? Do you have the data for Gradin and other Canuck forwards of his era?

In any case, is 21.2 MPG that impressive for a late 70s/early 80s D-man? For most of his time as a Canuck, 21MPG would be about average for a D-man (with only 17 skaters dressed).

Do you put much stock in team awards? Stecher won the Pratt award last year, despite being 5th among Canuck d-men in TOI per game. Hutton won the year before despite being 4th in TOI per game. Neither hit 20 minutes per game. Do you believe they were Vancouver's best during those seasons?

These awards are largely popularity contests (and there is no doubt that Snepsts was very popular in Vancouver).

But if you want to go the team awards route, it is interesting that the three names I dropped won the Cyclone Taylor award (team MVP) six times (Brodeur 3, McLean 2, and Gradin 1), whereas Snepsts won 0.

Those three won the Molson Cup seven times (Brodeur 4, McLean 3, and Gradin 1), whereas Snepsts won 0.

Re longevity, Snepsts has the advantage. But was he adding much value during his second stint? The gap in games played narrows considerably without those 98 games as a 34- and35-year old. All of Gradin's seasons with Vancouver were when he was still quite productive. He was Vancouver's #1 C for seven seasons. And Butcher's Vancouver seasons were all before his decline.

 

[seventieslord]

Where are you getting the ice time stats for Snepsts and Butcher? Do you have the data for Gradin and other Canuck forwards of his era?

TOI estimates were put together a decade ago by Iain Fyffe. They are based on situational GF/GA based on the idea that players who are on the ice for the most goals for and against, are on the ice the most. The methodology was tested against known results in post-99 seasons and was found to be very accurate.

However, Snepsts, a very low-event player, is likely underestimated by this metric.

Yes, the TOI data exists for forwards as well. I don't have time to type it all out - Do you want the file? Gradin was 1st among forwards once, 2nd another two times, and 3rd-6th in his other seasons. I called him "best" in two seasons by his point totals.

In any case, is 21.2 MPG that impressive for a late 70s/early 80s D-man? For most of his time as a Canuck, 21MPG would be about average for a D-man (with only 17 skaters dressed).

The simple answer is, it's pretty impressive, but slightly less impressive than if a player played that much today. I tracked the average TOI of a top defenseman going back to expansion, and it does tick slightly upwards as you go back, but really only slightly. The roster size difference affected forwards much more than defensemen. As it applies to Butcher, these guys' careers overlapped three seasons in Vancouver; the differences would be very minimal - Snepsts was obviously a much more valued player there.

(also, keep in mind this is with nearly zero PP time aside from the 1979 season. Other guys get to pad their minutes with 2-4 more easier PP minutes per game, he never did and still played 21+ on average)

Do you put much stock in team awards? Stecher won the Pratt award last year, despite being 5th among Canuck d-men in TOI per game. Hutton won the year before despite being 4th in TOI per game. Neither hit 20 minutes per game. Do you believe they were Vancouver's best during those seasons?

These awards are largely popularity contests (and there is no doubt that Snepsts was very popular in Vancouver).

No, I don't. I think last year's award flies in the face of the fact that the coaching staff clearly thought Edler was far more valuable a defenseman than anyone else on the team. (to a lesser degree, the year before that, too, though Edler missed a lot of games but if that excludes him, should've been Tanev). But if a player was the top ES defender on his team, AND wins the team award you can be pretty sure those two things mean something. Someone who would still claim otherwise wouldn't have much of a leg to stand on.

But if you want to go the team awards route, it is interesting that the three names I dropped won the Cyclone Taylor award (team MVP) six times (Brodeur 3, McLean 2, and Gradin 1), whereas Snepsts won 0.

Looks like a "best non-defenseman" award to me.

Those three won the Molson Cup seven times (Brodeur 4, McLean 3, and Gradin 1), whereas Snepsts won 0.

Again, defensemen do very badly in three star selections (Salming was the Leafs' best player the whole time he was there and won it 4 times, no defenseman since, Anderson has beaten a healthy Karlsson a few times, MacInnis never won one in Calgary, no Edmonton or Vancouver defenseman has ever won one, and Robinson and Housley are the only ones to win one - just one - for Winnipeg and Montreal. I believe that's 8 times in 231 opportunities) - and of course, a defensive defenseman will do a lot worse. To understand the value of a defensive defenseman, you have to dig deeper than awards that other positions are eligible to win. He was voted the team's best defenseman four times and he had the numbers to back it up every one of those times.

There's no "best goalie" award - that's just the one who played all the games that season. There's no "best forward" award - typically it's the guy who scored the most, give or take an intangibles case here and there. But defensemen get their own award, so they're not gonna get any votes for the Taylor, and I think you can see that too.

Re longevity, Snepsts has the advantage. But was he adding much value during his second stint? The gap in games played narrows considerably without those 98 games as a 34- and35-year old. All of Gradin's seasons with Vancouver were when he was still quite productive. He was Vancouver's #1 C for seven seasons. And Butcher's Vancouver seasons were all before his decline.

No, he wasn't - he was just a #6 guy at the end of his career. I included it for completeness' sake, but if you remove that, he's a 21.6 minute player in his first stint, a more impressive number yet. Or, if you want to go with his 6-year prime, 23.1 minutes - without PP time. That's not that common.

Gradin had the 59th best PPG average in the NHL during his Vancouver years. I know as an offensive forward he gets more attention, but try to put into perspective how relatively unimpressive his level of offense was. He was basically neck and neck with Smyl offensively, minus all of Smyl's valuable intangibles.

I don't think that number ("59th") is far off from where a player like Gradin would have ranked among all forwards in the NHL over that period as a whole. Snepsts, I think, would be relatively in the same range among D-men considering his usage (#2 D on average) and team awards. Based on 8 seasons, they'd be pretty much even, but Snepsts put in time in 4 more seasons and yes, considering context of awards, I think his awards are more impressive.

 

[decma]

TOI estimates were put together a decade ago by Iain Fyffe. They are based on situational GF/GA based on the idea that players who are on the ice for the most goals for and against, are on the ice the most. The methodology was tested against known results in post-99 seasons and was found to be very accurate.

However, Snepsts, a very low-event player, is likely underestimated by this metric.

Yes, the TOI data exists for forwards as well. I don't have time to type it all out - Do you want the file? Gradin was 1st among forwards once, 2nd another two times, and 3rd-6th in his other seasons. I called him "best" in two seasons by his point totals.



The simple answer is, it's pretty impressive, but slightly less impressive than if a player played that much today. I tracked the average TOI of a top defenseman going back to expansion, and it does tick slightly upwards as you go back, but really only slightly. The roster size difference affected forwards much more than defensemen. As it applies to Butcher, these guys' careers overlapped three seasons in Vancouver; the differences would be very minimal - Snepsts was obviously a much more valued player there.

(also, keep in mind this is with nearly zero PP time aside from the 1979 season. Other guys get to pad their minutes with 2-4 more easier PP minutes per game, he never did and still played 21+ on average)



No, I don't. I think last year's award flies in the face of the fact that the coaching staff clearly thought Edler was far more valuable a defenseman than anyone else on the team. (to a lesser degree, the year before that, too, though Edler missed a lot of games but if that excludes him, should've been Tanev). But if a player was the top ES defender on his team, AND wins the team award you can be pretty sure those two things mean something. Someone who would still claim otherwise wouldn't have much of a leg to stand on.



Looks like a "best non-defenseman" award to me.



Again, defensemen do very badly in three star selections (Salming was the Leafs' best player the whole time he was there and won it 4 times, no defenseman since, Anderson has beaten a healthy Karlsson a few times, MacInnis never won one in Calgary, no Edmonton or Vancouver defenseman has ever won one, and Robinson and Housley are the only ones to win one - just one - for Winnipeg and Montreal. I believe that's 8 times in 231 opportunities) - and of course, a defensive defenseman will do a lot worse. To understand the value of a defensive defenseman, you have to dig deeper than awards that other positions are eligible to win. He was voted the team's best defenseman four times and he had the numbers to back it up every one of those times.

There's no "best goalie" award - that's just the one who played all the games that season. There's no "best forward" award - typically it's the guy who scored the most, give or take an intangibles case here and there. But defensemen get their own award, so they're not gonna get any votes for the Taylor, and I think you can see that too.



No, he wasn't - he was just a #6 guy at the end of his career. I included it for completeness' sake, but if you remove that, he's a 21.6 minute player in his first stint, a more impressive number yet. Or, if you want to go with his 6-year prime, 23.1 minutes - without PP time. That's not that common.

Gradin had the 59th best PPG average in the NHL during his Vancouver years. I know as an offensive forward he gets more attention, but try to put into perspective how relatively unimpressive his level of offense was. He was basically neck and neck with Smyl offensively, minus all of Smyl's valuable intangibles.

I don't think that number ("59th") is far off from where a player like Gradin would have ranked among all forwards in the NHL over that period as a whole. Snepsts, I think, would be relatively in the same range among D-men considering his usage (#2 D on average) and team awards. Based on 8 seasons, they'd be pretty much even, but Snepsts put in time in 4 more seasons and yes, considering context of awards, I think his awards are more impressive.

Yes, good point re Molson Cup being heavily skewed toward G and F. I agree that it should not be used in this context.

And thanks for the ice time estimate information. I know we had this discussion several months ago, but I had forgotten the details and haven't been able to access old PMs since the site transition. I found a Fyffe article explaining his method and am going to review it.

Re effect of roster rize on d-man ice time, in games when 6 d-men dress, the average TOI per D-man is obviously 20 minutes. However, in games when 5 d-men dress, the average TOI per D-man is 24 minutes.

I don't know how many times the Canucks dressed only 5 D and 12 F (instead of 6 D and 11 F) before the roster size change, but I think a reasonable estimate can be made by looking at games played. In 81/82, for example, tallying up the games totals for Vancouver's d-men indicates 460 total games for their D-men, which is consistent with 60 games with 6D dressed and 20 games with 5D dressed. This means the average TOI per D-man for the season, weighted by number of games when 6 d-men dressed vs. 5, would be 21 MPG.

21 MPG is less impressive in those circumstances than post 83, when the average D played 20 MPG.

In any case, regardless of whether 5 or 6 D dressed, I suspect teams used their top 4 guys more pre roster change than after.

But this is just conjecture on my part and if you tracked it back to expansion and found that the average TOI of top D only ticked slightly upward going back, then I accept that. Out of curiosity, what how big was the uptick?

 

[vadim sharifijanov]

Do you put much stock in team awards? Stecher won the Pratt award last year, despite being 5th among Canuck d-men in TOI per game. Hutton won the year before despite being 4th in TOI per game. Neither hit 20 minutes per game. Do you believe they were Vancouver's best during those seasons?

No, I don't. I think last year's award flies in the face of the fact that the coaching staff clearly thought Edler was far more valuable a defenseman than anyone else on the team. (to a lesser degree, the year before that, too, though Edler missed a lot of games but if that excludes him, should've been Tanev).

not really on topic but it should have been tanev both years, not close.

 

[seventieslord]

Re effect of roster rize on d-man ice time, in games when 6 d-men dress, the average TOI per D-man is obviously 20 minutes. However, in games when 5 d-men dress, the average TOI per D-man is 24 minutes.

Yes, that's true. Had you not gone deeper into this, I'd have pointed out that this difference was usually found in the forwards, not in the defensemen, but your results below demonstrate that already.

I don't know how many times the Canucks dressed only 5 D and 12 F (instead of 6 D and 11 F) before the roster size change, but I think a reasonable estimate can be made by looking at games played. In 81/82, for example, tallying up the games totals for Vancouver's d-men indicates 460 total games for their D-men, which is consistent with 60 games with 6D dressed and 20 games with 5D dressed. This means the average TOI per D-man for the season, weighted by number of games when 6 d-men dressed vs. 5, would be 21 MPG.

I agree with your calculations. I checked the two previous seasons, and the total GP were 465 and 462, meaning a defenseman's "fair share" over those three seasons would be in the neighbourhood of 20.9 minutes. Snepsts still played about 22.4 during that time. During Butcher's career with Vancouver he played 18.5 when his "fair share" would have been 20.0.

(for the record, those numbers are 435, 438 and 439 in the preceding three seasons, meaning a defenseman's "fair share" of minutes would have been approximately 21.9 per game over that time. Snepsts' actual average over that time: 23.9 minutes per game.

I think you have successfully pointed out that the difference in their usage is not quite as vast as I initially claimed; however, the differences between their results are still much closer to my initial assertion, than they are to showing they were even/equal as players.

21 MPG is less impressive in those circumstances than post 83, when the average D played 20 MPG.

Yes and no. Depends on the team situation. If that team was choosing to dress 6 defensemen most/all of the time pre-1983, it's more or less the same. On the other hand, if they were dressing only 5 and playing those 5 approximately 20% more, that should speak to the perceived value of those five. Numerically there may need to be some mental adjustments when comparing players from directly pre- and post-1983, but to simply lop off 20% would overstate the difference, too.

Out of curiosity, what how big was the uptick?

I had a post where I alluded to this before during ATD 2011. I will see if I can find it. It doesn't outright describe the results, but it makes adjustments to defensemen's TOI from the 70s-90s based on what the uptick was that I found, so I can probably reverse engineer that.

 

[seventieslord]

Found it: http://hfboards.mandatory.com/threads/atd-2012-draft-thread-viii.1134463/page-24#post-47027609

Unfortunately, it looks like all the data has completely disappeared. I can't edit or quote the post to see if the numbers are in there. This is the first I've seen of this - I saw some stuff get its formatting messed up, but completely disappearing would be a new thing.