The Hockey News Top 100 Defensemen of all-time (special edition magazine)

[Johnny Engine]

I'm really puzzled by "journalistic integrity". What journalistic code of ethics is being violated by publishing an aggregation of several hockey writers' opinions here?

 

[quoipourquoi]

In 2013 or so, I received a more concise burial of my work from someone I had an incredible amount of respect for, and I basically shrunk into my chair.

That sucks. I hope the positive praise outweighs the bad.

 

[Professor What]

Okay, now that I'm sitting down, taking more time for a real response to that letter rather than just a joke, there are some pretty serious logistical issues with some of the "points."

First, with the idea that +/- is the definitive stat to rank defensemen, then it only stands to reason that points are the definitive stat to rank forwards, which would mean that Ron Francis is the fifth-greatest forward of all time, and that the entire hockey world has been horribly mistaken by including Lemieux as a member of the "Big Four," since he's only the eighth best forward of all time. I have a feeling that the writer of that letter would dismiss that as being ridiculous, but it has as much logistical grounding as the argument in the letter does. The assessment also fails to acknowledge that by the writer's own methodology, Orr had the greatest ever defensive season when he posted a +124. I wonder what their view on Dallas Smith's +98 is.

Second, while the writer accuses the panel of reinventing the definition of "defenseman" as support of their argument against Orr, it only seems right to point out that Orr reinvented the concept of the position. That even seems appropriate linguistically, since, culturally, words' meanings shift over time, just as Orr's arrival on the NHL represented a shift in hockey's "culture." Think about what the word "gay" means in today's culture, and compare that to what it meant 100 years ago. To think that things don't change, whether it's the perception of a word or the way a game is played is simply wrong.

Third, given that line of thinking, why not simply say that lists like the top defensemen project are pointless and should never be done. Since there's a clear statistical ranking based on +/-, points, wins, or what have you. In fact, we're all wasting our time with the project we're working on. Great to know we're wasting months of our lives, right?

I'm watching "Corner Gas" while I'm writing this (for any of you who are familiar with that show), and it's making me think, I've heard sounder arguments on there from Hank and Oscar...

 

[seventieslord]

The idea that +/- is a defensive stat comes from, of course, the idea that minuses are assigned for goals against. So really when someone says that +/- is a measure of defense what they really mean to say is that the minus is a defensive stat. Of course, there's a lot more to it than that, but at least if we get the pluses out of there we are eliminating a great deal of noise from +/-.

Orr had 526 adjusted ESGA scored against him in 657 games, for an average of 0.80 per game. Robinson had 1022 scored against him in 1384 games, for an average of 0.74 per game. If one is really trying to determine their defensive efficiency, that would be where you could start. Then you'd have to consider things like the quality of players on the ice with each of them. Robinson's teammates scored at a rate of 1.34 goals for, for every goal against during his entire career without him on the ice. Orr's teammates scored just 1.03 goals per goal against with him off the ice. Is that clear difference in teammate quality enough to account for a 7.5% difference in goals against per game? probably.

 

[VanIslander]

...
4. Eddie Shore
5. Ray Bourque

I can't agree there have been 4 greater defensemen than Bourque. But Shore might have been better. Both facts could be true.

9. Paul Coffey
10. Chris Chelios

In what universe?

27. Karlsson
31. Doughty
45. Weber
48. Burns

Do you se a pattern here? Ab-surd.

 

[Hockey Outsider]

The idea that +/- is a defensive stat comes from, of course, the idea that minuses are assigned for goals against. So really when someone says that +/- is a measure of defense what they really mean to say is that the minus is a defensive stat. Of course, there's a lot more to it than that, but at least if we get the pluses out of there we are eliminating a great deal of noise from +/-.

Orr had 526 adjusted ESGA scored against him in 657 games, for an average of 0.80 per game. Robinson had 1022 scored against him in 1384 games, for an average of 0.74 per game. If one is really trying to determine their defensive efficiency, that would be where you could start. Then you'd have to consider things like the quality of players on the ice with each of them. Robinson's teammates scored at a rate of 1.34 goals for, for every goal against during his entire career without him on the ice. Orr's teammates scored just 1.03 goals per goal against with him off the ice. Is that clear difference in teammate quality enough to account for a 7.5% difference in goals against per game? probably.

All those points are valid, not to mention ESGA really should be looked at on a per-minute (rather than per-game) basis. It's likely that Orr was getting a bit more ES ice time than Robinson (on top of way more on special teams) so his per-minute goal allowance rate would be even lower.

 

[quoipourquoi]

All those points are valid, not to mention ESGA really should be looked at on a per-minute (rather than per-game) basis. It's likely that Orr was getting a bit more ES ice time than Robinson (on top of way more on special teams) so his per-minute goal allowance rate would be even lower.

I think that effect might be lessened a little by their teams’ respective penalties. In the late-1960s and early-1970s, Boston was among the more penalized teams, whereas the mid-1970s and onward Montreal usually were among the least. I think Orr certainly had a bigger slice of his teams’ ES minutes than Robinson, but the pie itself may have been a little smaller, putting them on more similar footing.

On the flip side, Boston had goaltending that was more middle-of-the-pack. It doesn’t surprise me all that much that two of Robinson’s higher ESGA seasons from his 20s were 1974 and 1980, whereas Orr’s 1971 and 1972 (when Boston’s goaltending was more towards the high end of save percentage) ESGA numbers were lower than even some of his surrounding seasons where he played a dozen fewer games.

 

[ted2019]

The idea that +/- is a defensive stat comes from, of course, the idea that minuses are assigned for goals against. So really when someone says that +/- is a measure of defense what they really mean to say is that the minus is a defensive stat. Of course, there's a lot more to it than that, but at least if we get the pluses out of there we are eliminating a great deal of noise from +/-.

Orr had 526 adjusted ESGA scored against him in 657 games, for an average of 0.80 per game. Robinson had 1022 scored against him in 1384 games, for an average of 0.74 per game. If one is really trying to determine their defensive efficiency, that would be where you could start. Then you'd have to consider things like the quality of players on the ice with each of them. Robinson's teammates scored at a rate of 1.34 goals for, for every goal against during his entire career without him on the ice. Orr's teammates scored just 1.03 goals per goal against with him off the ice. Is that clear difference in teammate quality enough to account for a 7.5% difference in goals against per game? probably.

Question, is this the sheet in which Larry Robinson's EV+/- is 611? If it is, is there a way to sort the charts from largest to smallest?

 

[seventieslord]

Question, is this the sheet in which Larry Robinson's EV+/- is 611? If it is, is there a way to sort the charts from largest to smallest?

Yes, that's the one. and sorting is really easy, just use excel's native sort function, or click on the filter arrows at the tops of the columns.

 

[MadLuke]

Based on this assessment, the greatest defensive defenceman is Larry Robinson.

Robinson’s career Plus/Minus: Plus 722 over 20 seasons & 1384 games.

Orr’s career Plus/Minus: Plus 582 over 12 seasons & 657 games.

This 140 point differential, represents an increase in magnitude of 24% in efficiency over Orr. The magnitude of this differential increases significantly when taking into consideration that Robinson played 802 more games than Orr.

Imagine a conversation about who was greater offensively between Ron Franchis and Mario Lemieux, and someone saying not only Francis was greater because he had more points but he achieved to have more points while playing almost twice as many games.

plus/minus by game here is important, the fact that Robinson took almost twice has many game to get +24% in +/- is not increasing the magnitude of that difference, it is flattening it completely.

Now considering a bit latter you are writing:
Orr had 526 adjusted ESGA scored against him in 657 games, for an average of 0.80 per game. Robinson had 1022 scored against him in 1384 games, for an average of 0.74 per game. If one is really trying to determine their defensive efficiency, that would be where you could start.

You seem to completely agree already, so now I am just completely confused.

 

[seventieslord]

Imagine a conversation about who was greater offensively between Ron Franchis and Mario Lemieux, and someone saying not only Francis was greater because he had more points but he achieved to have more points while playing almost twice as many games.

plus/minus by game here is important, the fact that Robinson took almost twice has many game to get +24% in +/- is not increasing the magnitude of that difference, it is flattening it completely.

Now considering a bit latter you are writing:
Orr had 526 adjusted ESGA scored against him in 657 games, for an average of 0.80 per game. Robinson had 1022 scored against him in 1384 games, for an average of 0.74 per game. If one is really trying to determine their defensive efficiency, that would be where you could start.

You seem to completely agree already, so now I am just completely confused.

Yep, you're definitely confused! The first part is not me, the second part is, that's why they are completely contradictory.

 

[ted2019]

Yes, that's the one. and sorting is really easy, just use excel's native sort function, or click on the filter arrows at the tops of the columns.

Thanks, I'll try that.

 

[ted2019]

Yes, that's the one. and sorting is really easy, just use excel's native sort function, or click on the filter arrows at the tops of the columns.

It says that I need the password to sort the columns.

 

[Professor What]

Imagine a conversation about who was greater offensively between Ron Franchis and Mario Lemieux, and someone saying not only Francis was greater because he had more points but he achieved to have more points while playing almost twice as many games.

plus/minus by game here is important, the fact that Robinson took almost twice has many game to get +24% in +/- is not increasing the magnitude of that difference, it is flattening it completely.

Now considering a bit latter you are writing:
Orr had 526 adjusted ESGA scored against him in 657 games, for an average of 0.80 per game. Robinson had 1022 scored against him in 1384 games, for an average of 0.74 per game. If one is really trying to determine their defensive efficiency, that would be where you could start.

You seem to completely agree already, so now I am just completely confused.

If you're referring to the mention of Lemieux and Francis in this thread, that was me, though I wasn't arguing that Francis was greater. Quite the contrary, actually. I was pointing out that using raw plus/minus to rate defensemen, as the writer of the letter @seventieslord posted seemed to suggest would be something akin to saying that Francis was greater than Lemieux because of raw points.

 

[MadLuke]

If you're referring to the mention of Lemieux and Francis in this thread, that was me, though I wasn't arguing that Francis was greater. Quite the contrary, actually. I was pointing out that using raw plus/minus to rate defensemen, as the writer of the letter @seventieslord posted seemed to suggest would be something akin to saying that Francis was greater than Lemieux because of raw points.

We just shared mind about an obvious example to use to show that total +/- instead of +/- by game was a strange metric to use (I had yet to read your message when I wrote mine).

 

[seventieslord]

It says that I need the password to sort the columns.

That's really strange, none of the versions I have ever seen have a password. Sometimes when you get a spreadsheet from a third party you have to click another button inside the sheet to confirm you want to unprotect it to make it editable. That might be what is happening?

 

[ted2019]

That's really strange, none of the versions I have ever seen have a password. Sometimes when you get a spreadsheet from a third party you have to click another button inside the sheet to confirm you want to unprotect it to make it editable. That might be what is happening?

Yes, I received it from a valued member of his section. My original got lost when I switched computers earlier this Summer.

 

[Professor What]

We just shared mind about an obvious example to use to show that total +/- instead of +/- by game was a strange metric to use (I had yet to read your message when I wrote mine).

That being the case, I misunderstood your post, but yeah, we completely agree. It just goes to show again though why I tend to cringe when I see plus/minus brought up in these conversations. It's not totally useless (even if I'm not a fan of it overall), but the applications people try to make usually are.

 

[Habsfan18]

In the most recent issue of THN, they added an additional 10 active defensemen (in alphabetical order) who although missing the cut this time, they feel could potentially make the list one day:

Mattias Ekholm
Oliver Ekman-Larsson
Dougie Hamilton
Torey Krug
Seth Jones
Ivan Provorov
Morgan Rielly
Jaccob Slavin
Marc-Edouard Vlasic
Zach Werenski

 

[seventieslord]

In the most recent issue of THN, they added an additional 10 active defensemen (in alphabetical order) who although missing the cut this time, they feel could potentially make the list one day:

Mattias Ekholm
Oliver Ekman-Larsson
Dougie Hamilton
Torey Krug
Seth Jones
Ivan Provorov
Morgan Rielly
Jaccob Slavin
Marc-Edouard Vlasic
Zach Werenski

I don't see any universe in which we ever consider Ekholm a top-100 defenseman. Are we ever going to consider him better than Mike Ramsey or Andrei Markov?

Krug is defintely no Markov. That one's easy.

Rielly seems like a reeeeeal longshot. He's already a 7-year vet at 26 who's achieved very, very little. I think he's a legit #1 NHL defenseman, and might be for 5-6 years, but that doesn't get you on this list.

Vlasic is turning 34 shortly, and he's no Mike Ramsey.

Slavin, on the other hand? he is a Mike Ramsey. He has a real shot.

Jones will obviously make it, barring a major injury.

Provorov and Werenski are so young and so much can happen, but it appears they'll be strong candidates.

What's OEL trending towards? Numminen plus? Numminen minus? Teppo made #89 so there's a chance.

Hamilton is a wildcard because his career might go exactly as it has, in which case he's the next Al Iafrate and doesn't make it, but it may also get significantly better or worse.

 

[Habsfan18]

Yeah, I especially thought Ekholm stood out on the list and not in a good way. I mean, he’s an important part of the Preds D. But has a single person even thought for a single second that be could one day be a top 100 D-man of ALL-TIME?

Their reasoning: “One of the game’s longest reaches, making it extremely difficult for opponents to get around him. Crucial on the PK, and Nashville rode him for more than 25 minutes a night during its run to the 2017 Stanley Cup final. Overshadows by Josi but if he can maintain his smothering standard a few more seasons, he’ll be remembered as one of the best scoring-chance killers of his era.”

 

[JackSlater]

The ravings in that two part letter/magnum opus seemed very familiar. I do not think that the person is a poster, just that their position and manner of making that position is reminiscent of a poster or two on here. Grandiose allusions to having some sort of elevated understanding? Yes. Criticisms of dissenting opinions, and opinions in general, because they don't match other opinions? Indeed. A general misunderstanding of a fairly common statistic and how to interpret it? Oh yeah. Cloaking that general misunderstanding by attempting to co-opt math, poorly? Of course. Ending with an appeal to authority? It's all there. Good stuff overall. I didn't agree with any of it but I did enjoy the read.

 

[Kosher Halal]

Charlie Huddy and Ken Daneyko did not make the list.

 

[Yozhik v tumane]

Charlie Huddy and Ken Daneyko did not make the list.

Two points in its favor, I’d say.

 

[Deep Blue Metallic]

Quite the bump.

Interested in the Lidstrom versus Bourque conversation. I can certainly understand Ray's supporters, but the outrage at Lidstrom being ranked above him seems unjustified.

Bourque: 5 Norris Trophies; 13 times 1st Team All-Star; Regular Season: 1,612 games, 410G, 1,169A; Playoffs: 1 Stanley Cup, 214 games, 41G, 139A

Lidstrom: 7 Norris Trophies; 10 times 1st Team All-Star; Regular Season: 1,564 games, 264G, 878A; Playoffs: 4 Stanley Cups, 1 Conn Smythe, 263 games, 54G, 129A

Bourque was better offensively, and I don't know any serious hockey fan who doesn't cherish seeing him finally lift the Cup in his last game as one of the greatest memories of the sport. Lidstrom was rightly celebrated for the ease with which he could dominate defensively, with his positioning and anticipation much more so than with physicality. He also benefited from being on better teams than Borque.

Seems like a tossup to a neutral fan like me.