Robson Division Semifinals: Orillia Terriers vs Regina Pats

[seventieslord]

Nearly equal is not the same as equal. As shown earlier, Orillia has an offensive edge on both the top two lines.

Most importantly, Regina relies very heavily on one player to create a huge amount of their offense... and Orillia has the guys in match-up positions to slow him down significantly.

Again (assuming Lindsay's scores are reflective of his talent and assuming Foyston is equally effective at RW, both of which seem rather dubious), that's great that you got yourself a set of VsX scores that are 2% better all things considered, but that doesn't mean your lines work better than ours. There's more to line building than just assembling three good VsX scores.

As I said before, size is already accounted into these players' legacies. Being big made them what they were, so double-counting it seems off.

That's not the way this works. We still draft players with a consideration to their attributes, strengths and weaknesses - if we could just treat each player as an overall "sum of parts" package there would be no team building aspect to this. But there is - as the GMs, we must make sure we don't take too many similar players, take players who complement eachother well and don't leave the team deficient in any area. Unfortunately, the way we both built our teams, it's just a big physical mismatch.

Well, for starters, if you're on the PK, there's a good chance it's Joe Hall in the box.

good one. But no, it's not like the guy's giving you two powerplays each game, let's be realistic.

Second, going to the net has nothing to do with beating up guys like Joe Hall. All I have to do is go to the front of the net. From there, I keep the goalie in his crease, and also obstruct the goalie's view.

The whole idea of a "net presence" is kind of silly. Anyone can be a net presence... just go stand there.

I don't expect you to have a player who can "beat up" Joe Hall, either. But you do need someone with the size and/or strength and/or courage to stand there and take a pounding. You're right that obstructing the goalie is important, but the one guy with size on the 2nd unit is also the softest forward in the series, so that's not a good matchup for you.

As a coach and a person who watches a lot of hockey, I'm really surprised that you think anyone can do this.

Yes, Regina has a better group of PK forwards on the 2nd unit. On the first unit, Orillia very clearly has the better forwards. Ramsay > Savard and Walker > Gilmour.

As a PK defenseman, I don't think there's much of an argument against Larry Robinson being the best among these groups. He was widely viewed as the best defensive defenseman of his era, and everything we know about how he played translates well to the PK.

I don't agree. This is why the first units are practically even. Robinson spent most of his career on a very dominant team, yet, was used far less on the PK than one would think and attained very good, but not "omg tha best in tha series" results... 45% is just not that much. Tom Johnson was Doug Harvey's teammate, yet was the leader of the PK and I don't need to tell you that they were an obscenely dominant team. He has a legitimate case as the best PK defenseman of the entire pre-expansion era, while the numbers for the post-expansion era would place a fair number of them ahead of Robinson, and that's even if one is charitable because of his size and style of play translating well to the PK, as you said.

Toe Blake is way better than Babcock. It's not even close really. The gap is wide.

Orillia's actual coach, Dick Irvin, is a lot closer to Babcock. It's a safe bet that Babs eventually overtakes I'm, but as of right now Irvin has the edge.

embarrassing.

Irvin is a little less fearsome.

I'm a competitor. You clearly are as well. you put up a good fight till the end. And it may not even be over yet. Best of luck to you. It was fun.

 

[Dreakmur]

Again (assuming Lindsay's scores are reflective of his talent and assuming Foyston is equally effective at RW, both of which seem rather dubious), that's great that you got yourself a set of VsX scores that are 2% better all things considered, but that doesn't mean your lines work better than ours. There's more to line building than just assembling three good VsX scores.

I realize that, which is why I like my line construction. Both my lines are balanced with multiple threats, which makes them tougher to shut down.

That's not the way this works. We still draft players with a consideration to their attributes, strengths and weaknesses - if we could just treat each player as an overall "sum of parts" package there would be no team building aspect to this. But there is - as the GMs, we must make sure we don't take too many similar players, take players who complement eachother well and don't leave the team deficient in any area. Unfortunately, the way we both built our teams, it's just a big physical mismatch.

That's correct, you have larger guys. The question is, what does that actually accomplish.

Duncan Keith is small for a defenseman now. How has that impacted his career thus far? JC Tremblay was small - and actually well known to be physically soft. How did that impact his career? Doug Gilmour was small too, and he did fine.

All the small guys managed to overcome that by being better in other areas. That still applies in the ATD.

I don't expect you to have a player who can "beat up" Joe Hall, either. But you do need someone with the size and/or strength and/or courage to stand there and take a pounding. You're right that obstructing the goalie is important, but the one guy with size on the 2nd unit is also the softest forward in the series, so that's not a good matchup for you.

It doesn't even have to be my guy who obstructs the view. Joe Hall standing there trying to beat my guy up would to it too, right?

It's not even about just standing there for the full 2 minutes anyway. A guy can just cruise through the slot at the right time and be just as effective at obscuring the goalie's vision.

As a coach and a person who watches a lot of hockey, I'm really surprised that you think anyone can do this.

Anyone can do it. Most people just aren't willing to do it.

I don't agree. This is why the first units are practically even. Robinson spent most of his career on a very dominant team, yet, was used far less on the PK than one would think and attained very good, but not "omg tha best in tha series" results... 45% is just not that much. Tom Johnson was Doug Harvey's teammate, yet was the leader of the PK and I don't need to tell you that they were an obscenely dominant team. He has a legitimate case as the best PK defenseman of the entire pre-expansion era, while the numbers for the post-expansion era would place a fair number of them ahead of Robinson, and that's even if one is charitable because of his size and style of play translating well to the PK, as you said.

There's a big difference between usage an ability.

Tom Johnson was not as good defensively as Harvey, but was given more PK time. Why? Most likely because Harvey couldn't play the whole game, and his coach thought he was more valuable in offensive situations. Doug Harvey was a better PKer than Tom Johnson despite the fact that Johnson was used more.

 

[BraveCanadian]

Duncan Keith and J.C. Tremblay, for example, have been pretty unflappable in real life against NHL/WHA level competition, but in an ATD environment they're a bit shrimpy and stand out as guys who the likes of Tocchet, Hunter, Nolan and Tkachuk can really wear down. Here are the adjusted sizes of our blueliners:

Pratt: 6'6", 242
Savard: 6'5", 230
Seibert: 6'5", 228
Hall: 6'3", 215
Johnson: 6'3", 210
Vadnais: 6'3", 205

Robinson: 6'5", 235
Ramsey: 6'4", 205
Harper: 6'3", 220
Larson: 6'1", 205
Keith: 6'1", 192
Tremblay: 6'1", 190

Yikes... don't like this argument, at all.

 

[seventieslord]

I realize that, which is why I like my line construction. Both my lines are balanced with multiple threats, which makes them tougher to shut down.

This is a really liberal definition of balanced. There's no size and no significant defensive presence on the first line - that's not balanced.

That's correct, you have larger guys. The question is, what does that actually accomplish.

Duncan Keith is small for a defenseman now. How has that impacted his career thus far? JC Tremblay was small - and actually well known to be physically soft. How did that impact his career? Doug Gilmour was small too, and he did fine.

All the small guys managed to overcome that by being better in other areas. That still applies in the ATD.

And every one of these guys is now being put in a situation very similar to that of a player being called up from the minors to the NHL. Their size didn't hurt them at that lower level, but here, where everyone is bigger, faster and better, it can.

Gilmour is a small guy who could get roughed up, sure. But he's a rarity on my roster and he'd fit right in with the majority of Orillia's forwards.

It doesn't even have to be my guy who obstructs the view. Joe Hall standing there trying to beat my guy up would to it too, right?

It's not even about just standing there for the full 2 minutes anyway. A guy can just cruise through the slot at the right time and be just as effective at obscuring the goalie's vision.

Anyone can do it. Most people just aren't willing to do it.

Perhaps. I was equating "willing to" and "can" as the same thing. It's like backchecking. So I don't really see the value of saying your guys all "can" do it. Who WILL?

It's not a 100% prerequisite for a successful power play, but bodies in front absolutely helps and it's why guys like Dave Andreychuk, Dino Ciccarelli, Joe Nieuwendyk and Tim Kerr often feel overvalued compared to their overall resumes.

There's a big difference between usage an ability.

Tom Johnson was not as good defensively as Harvey, but was given more PK time. Why? Most likely because Harvey couldn't play the whole game, and his coach thought he was more valuable in offensive situations. Doug Harvey was a better PKer than Tom Johnson despite the fact that Johnson was used more.

...is that your assumption just because Harvey was Harvey?

Yikes... don't like this argument, at all.

I've seen you say you don't like adjusted size in the past, but what I've never seen you say is why, in any kind of way that would compel someone to agree. As I said many times before, I was not a 100% believer in it because it was just something I made up and people ran with it. But this year I tested it and it was far more accurate than I ever thought, until we got to players born before 1910, in which case a couple tweaks needed to be made. I doubt you read the entire draft thread, so here's the post:

By the way, speaking of adjusted height and weight, has anyone ever taken a deeper look at the averages over the years? I designed it off the cuff and have always said it passed the smell test, but never went any further than that. Look at this:

https://hockey-graphs.com/2015/02/19...-a-brief-look/

- Call 2006 the modern peak. It's where NHL weight hit its maximum of 205.6 and players were 6'1.2" tall. It was very close to this peak both 10 years before, and 10 years after (never lower than 201.2 or 6'0.9"). A player born in 1970 would be at the midpoint of their career during this era, as would a player born in 1990, all of whom get no adjustment, as they played in the "modern era".

- Go back 20 years, 1986. Hopefully the average is about 195 and 6'0.2". What does the chart say? 194.7 and 6'0.3".

- 20 more years, in 1966, it's 5'11.2" and 183.2; again, almost exactly one inch and ten pounds lower.

- 20 more years back, in 1946, it's 5'10.4" and 175.5, but many 173 averages in the years surrounding it, so nearly an inch and ten pounds yet again. 175 would still represent exactly 30 pounds less than it would be in 60 years.

- 20 more years back in 1926, it's 5'9.5" and 170.8, so the height adjustment still looks legit, but weights have only changed 5 pounds this time.

- This isn't quite 20 years back, but I took 1913 and checked the height and weight of every player who played in the NHA and PCHA that season (I chose this year for a good sized sample; the 1906 ECAHA and FAHL were smaller, comprised a lower percentage of the talent pool we draft from compared to the PCHA/NHA, and (inevitably would have a lot of unknown heights and weights). Surprisingly, I found that the average height in this season was 5'9.2" and 171.2", so it was largely unchanged from the 1926 numbers.

- So I begrudgingly checked the 1906 season in order to go back a full 20 years. I was able to get a sample of 59 players from the ECAHA, FAHL, MHL and IHL, which is where 90+% of players we draft in the ATD/MLD were playing if they were active in 1906. I found that the averages were 5'8.9" and 168.0, so 0.6" and 3 pounds from 1926.

I would therefore propose a change to the "formula":

born 1970-present| No adjustment
born 1950-1969 |+1"|\, +10 lbs
born 1930-1949 |+2", +20 lbs
born 1910-1929 |+3", +30 lbs
born 1890-1909 |+4", +35 lbs
born 1870-1889 |+5", +40 lbs

The only change was to scale back the weight increases in the two oldest brackets, from 40 and 50, to 35 and 40. Instead of adhering to the 10 pounds per 20 years standard, I followed what the evidence clearly said. However, these are just rounded numbers, as the oldest generation is not quite an inch and five pounds smaller than the the 2nd oldest, but it's better than no adjustment, and I am trying to use simple to remember numbers. A player born all the way back in 1873 like Alf Smith would be a 33 year old graybeard at the time of the 1906 snapshot, whereas some players in that range would only be 17 that season. Those younger ones are getting a slightly larger upgrade than they deserve, but the way the average size curve moves over time, I think the older players like Smith would be perfectly appropriately adjusted. Anyway, this is the only generation that threatens to break the adjustment formula, and it's a pretty small percentage of the players we draft.

using adjusted sizes is currently the only way anyone has devised, to give a clear explanation of what an older player's size would be in a modern context. Following the above will give a pretty accurate representation of how a player's height and weight compared to the average of the top league at the time that they played. What's wrong with that?

 

[Dreakmur]

This is a really liberal definition of balanced. There's no size and no significant defensive presence on the first line - that's not balanced.

Ted Lindsay is the 2nd best defensive player I either of our top-6s. He's not elite, but he's definitely good.

And every one of these guys is now being put in a situation very similar to that of a player being called up from the minors to the NHL. Their size didn't hurt them at that lower level, but here, where everyone is bigger, faster and better, it can.

That applies to everything, not just size.

All the giant forecheckers get to try to attack guys that are faster, more skilled, and smarter than they did in their NHL careers.

All the super skilled guys are making plays against defensemen who are better at taking away passing lanes and closing the gaps I coverage.

etc....

...is that your assumption just because Harvey was Harvey?

Absolutely.

Same for Robinson vs. Savard. Robinson was pretty widely viewed as the better defensive player.

 

[seventieslord]

Ted Lindsay is the 2nd best defensive player I either of our top-6s. He's not elite, but he's definitely good.

Really, so now a guy who was never on a defensive line or the defensive player on his own line is better than Lewis based on.... 2 quotes that make brief mentions of defense??

Absolutely.

Same for Robinson vs. Savard. Robinson was pretty widely viewed as the better defensive player.

How can it possibly be described as "widely" when Savard finished 2nd in voting by coaches (past his prime and right in Robinson's prime) as the league's best defensive defenseman? Do you have the detailed voting results showing that every coach voted Robinson 1st and Savard 2nd? Because that's about the only way you can credibly use the word "widely" here.

And defensive play at even strength and on the PK are separate skills. Savard was clearly used by the Habs a lot more on the penalty kill and the team was better when he was used more.

 

[Dreakmur]

Really, so now a guy who was never on a defensive line or the defensive player on his own line is better than Lewis based on.... 2 quotes that make brief mentions of defense??

I think you're confusing Lewis with a guy who has more than a couple quotes about his defensive play. There's very little in his bio to support the reputation you claim he had.

How can it possibly be described as "widely" when Savard finished 2nd in voting by coaches (past his prime and right in Robinson's prime) as the league's best defensive defenseman? Do you have the detailed voting results showing that every coach voted Robinson 1st and Savard 2nd? Because that's about the only way you can credibly use the word "widely" here.

Widely, as in most people believed.

And defensive play at even strength and on the PK are separate skills. Savard was clearly used by the Habs a lot more on the penalty kill and the team was better when he was used more.

According to your bio, he was used as a forward most of the time.

That actually indicates that he wasn't that great as a defenseman on the PK, doesn't it? Why move him up front if he was so irreplaceable on the bottom of the box?

 

[BraveCanadian]

I've seen you say you don't like adjusted size in the past, but what I've never seen you say is why, in any kind of way that would compel someone to agree.

Because it is silly.

Average size in most of the developed world has gone up but it isn't a given that a subset of that population will be.

For example, the NHL is actually smaller and lighter now than it was in the late 90s when teams were actually passing over skill for size.

Meanwhile if you were to do the same thing for the NBA it would be drastically wrong because the NBA shifted from looking for athletes to looking for tall people.

As I said many times before, I was not a 100% believer in it because it was just something I made up and people ran with it.

Yes, it is something you just made up, and that is why people should not have run with it.

I agree we have to kind of eyeball how big a 6ft guy from 1900 would be today.. but the idea that we can simply adjust for it based on averages is nonsense.

 

[jarek]

Quite honestly I'm surprised size is even being brought up here. We've seen bigger teams get beaten by smaller teams with more skill in the past. Truth be told, I'm not sure that I've ever even considered size before when determining a matchup winner.

 

[BraveCanadian]

Quite honestly I'm surprised size is even being brought up here. We've seen bigger teams get beaten by smaller teams with more skill in the past. Truth be told, I'm not sure that I've ever even considered size before when determining a matchup winner.

To me it isn't just size as level of physicality -- hopefully without ending up in the box all the time -- and whether the opposition had players known for avoiding the rough stuff etc.

You don't have to be a physical monster to beat size.. as long as you're willing to pay the price then speed and puck movement can frustrate and wear out big guys looking to mix it up too.

 

[jarek]

That's exactly it. I'm much more interested in the willingness of a player to engage in physical play. You need look no further than Mitch Marner to realize that size isn't everything. Despite being maybe the smallest player in the league, he constantly wins physical battles he isn't "supposed" to win.

 

[seventieslord]

Widely, as in most people believed.

source?

According to your bio, he was used as a forward most of the time.

That actually indicates that he wasn't that great as a defenseman on the PK, doesn't it? Why move him up front if he was so irreplaceable on the bottom of the box?

That's a terrible conclusion to draw from that.

 

[Dreakmur]

source?

Let's start with the coaches' polls, where Robinson was voted better every year....

That's a terrible conclusion to draw from that.

Is it really?

For starters, you said that Savard was a better PKer because he was used more. If most of that time was spent as a forward, I don't understand what makes it comparable to other defensemen. An argument of "he played more PK than XXXX" is basically meaningless if they didn't play the same position on the PK.

Also, if I'm a coach, there's almost no way I take my best defensive defenseman and put him on forward for the PK. First, the forwards would have to be a lot weaker than the defensemen. Second, the rest of the defenseman would have to be similarly effective. Finally, none of the lower defensemen would be able to play forward.

 

[seventieslord]

Let's start with the coaches' polls, where Robinson was voted better every year....

How about a source I didn't already cite and/or one that helps confirm your statement. There's nothing "widely" about a poll that puts a prime Robinson 1st and a past-prime Savard 2nd, unless, everyone voted the former 1st and the latter 2nd.

Is it really?

For starters, you said that Savard was a better PKer because he was used more. If most of that time was spent as a forward, I don't understand what makes it comparable to other defensemen. An argument of "he played more PK than XXXX" is basically meaningless if they didn't play the same position on the PK.

Sure it is, if defensemen across the board play twice as much on the PK than forwards (their even share would be 33% as opposed to 17% for a forward) and yet the guy playing forward was out there 30% more often.

Also, if I'm a coach, there's almost no way I take my best defensive defenseman and put him on forward for the PK. First, the forwards would have to be a lot weaker than the defensemen. Second, the rest of the defenseman would have to be similarly effective. Finally, none of the lower defensemen would be able to play forward.

I'm sorry, but there's no arguing with the results and it's not like the person who employed this strategy the most often was some kind of idiot, either.

Why does Savard's usage as a forward instead of as a defenseman say so much about him, but Robinson's relatively low usage doesn't say the same about him?

 

[Dreakmur]

How about a source I didn't already cite and/or one that helps confirm your statement. There's nothing "widely" about a poll that puts a prime Robinson 1st and a past-prime Savard 2nd, unless, everyone voted the former 1st and the latter 2nd.

Why do I need another source when the first is quite clear?

Sure it is, if defensemen across the board play twice as much on the PK than forwards (their even share would be 33% as opposed to 17% for a forward) and yet the guy playing forward was out there 30% more often.

They are not comparable because they are not the same position.

I'm sorry, but there's no arguing with the results and it's not like the person who employed this strategy the most often was some kind of idiot, either.

I didn't say it was a bad strategy. I told you the situations that I would employ it.

Why does Savard's usage as a forward instead of as a defenseman say so much about him, but Robinson's relatively low usage doesn't say the same about him?

They both say something. I'm just trying to separate the false comparisons.

 

[seventieslord]

Why do I need another source when the first is quite clear?

Because that's not clear at all. 21 NHL coaches answered, and we have no idea if they ranked top-3 or just one, and all we know is Robinson came 1st and Savard 2nd. It might have been 8-7 or it might have been 14-5; these polls don't prove that your use of the word "widely" was at all appropriate.

They are not comparable because they are not the same position.

Yeah, it's comparable in the same way we can say Jake Gardiner is a better offensive player than Zach Hyman. Defensemen typically score half as many points as forwards, and yet Gardiner scored significantly more points than Hyman.

 

[Dreakmur]

Because that's not clear at all. 21 NHL coaches answered, and we have no idea if they ranked top-3 or just one, and all we know is Robinson came 1st and Savard 2nd. It might have been 8-7 or it might have been 14-5; these polls don't prove that your use of the word "widely" was at all appropriate.

Ok, so you just don't like the word widely? Want me to say the majority opinion was that Robinson was better?

Yeah, it's comparable in the same way we can say Jake Gardiner is a better offensive player than Zach Hyman. Defensemen typically score half as many points as forwards, and yet Gardiner scored significantly more points than Hyman.

Yes, Gardiner is better for a defenseman than Hyman is for a forward. Why do you think they don't put Gardiner up front to replace Hyman?

 

[seventieslord]

Ok, so you just don't like the word widely? Want me to say the majority opinion was that Robinson was better?

Yeah, I thought I made that pretty clear in three separate posts...

and yes, it was the majority opinion when Robinson was at his peak and Savard was not, whatever can be inferred from that...

Yes, Gardiner is better for a defenseman than Hyman is for a forward. Why do you think they don't put Gardiner up front to replace Hyman?

Gardiner's not just "better for a defenseman", he's just plain better. And switching a player's position full-time is not even close to the same thing as playing them in a different position on special teams when it's been discovered to be advantageous.

 

[Dreakmur]

Gardiner's not just "better for a defenseman", he's just plain better. And switching a player's position full-time is not even close to the same thing as playing them in a different position on special teams when it's been discovered to be advantageous.

Yes, yes, we all know that. The question I asked though, is why would you not make the switch? It's the same reason I wouldn't. It's the same reason his coach's don't.

 

[seventieslord]

Yes, yes, we all know that. The question I asked though, is why would you not make the switch? It's the same reason I wouldn't. It's the same reason his coach's don't.

You've got this backwards.

My assertion is that Savard is a better penalty killer than Robinson because of his superior results and because his amount of usage was not only significantly higher, but especially significantly higher because of the position he played (it is over 3X higher than the average forward, while Robinson was only used 37% more often than the average defenseman).

You claim that the fact that they put him at forward on the PK instead back where a defenseman would usually go, means he wasn't as good as Robinson.

I'm saying that although 58% is obviously more than 45%, it's even more greatly pronounced when you consider the demands of the position and statistical trends that have emerged over the last 50 years for obvious reasons - defensemen naturally have higher PK usage numbers across the board, so when a forward (or in this case a defenseman playing as one) has a higher usage number than a defenseman, that's very noteworthy.

As an analogy, I used the well-known fact that forwards tend to score more across the board than defensemen, but that this is not a hard and fast rule - some defensemen outscore some forwards, like, for example, Jake Gardiner and Zach Hyman. Jake Gardiner outscores Zach Hyman by a fairly significant amount, despite the offensive disadvantage of the defense position and the statistical trends we've seen emerge over a century of hockey. Therefore, not only would a blind look at their offensive numbers all but confirm Gardiner was the better offensive player, once their positions are known, it only drills that point home further.

Jake Gardiner is the Serge Savard in this analogy - the guy whom is outperforming the other guy so much that it transcends position.

You're asking "Why move him up front if he was so irreplaceable on the bottom of the box?", and "Why do you think they don't put Gardiner up front to replace Hyman?" as though the questions are asking the same thing - they're not. One is regarding the statistical significance of the actual results of a player who has actually changed positions in a certain circumstance, the other is regarding the hypothetical of moving a player to a different position based on his proficiency in something that's not his primary role, when the only reason that player was even brought up was to make an analogy about statistical significance

 

[seventieslord]

I agree we have to kind of eyeball how big a 6ft guy from 1900 would be today.. but the idea that we can simply adjust for it based on averages is nonsense.

The data tells us that a 6'0" player born in 1900 is approximately 3-4" taller than average in big league hockey around the midpoint of his career. There's nothing wrong with saying that at all, right?

So if we then take that number and add it to today's average in order to give a visual of what that means, why would you have such a problem with it?

 

[Dreakmur]

You've got this backwards.

My assertion is that Savard is a better penalty killer than Robinson because of his superior results and because his amount of usage was not only significantly higher, but especially significantly higher because of the position he played (it is over 3X higher than the average forward, while Robinson was only used 37% more often than the average defenseman).

You claim that the fact that they put him at forward on the PK instead back where a defenseman would usually go, means he wasn't as good as Robinson.

I'm saying that although 58% is obviously more than 45%, it's even more greatly pronounced when you consider the demands of the position and statistical trends that have emerged over the last 50 years for obvious reasons - defensemen naturally have higher PK usage numbers across the board, so when a forward (or in this case a defenseman playing as one) has a higher usage number than a defenseman, that's very noteworthy.

As an analogy, I used the well-known fact that forwards tend to score more across the board than defensemen, but that this is not a hard and fast rule - some defensemen outscore some forwards, like, for example, Jake Gardiner and Zach Hyman. Jake Gardiner outscores Zach Hyman by a fairly significant amount, despite the offensive disadvantage of the defense position and the statistical trends we've seen emerge over a century of hockey. Therefore, not only would a blind look at their offensive numbers all but confirm Gardiner was the better offensive player, once their positions are known, it only drills that point home further.

Jake Gardiner is the Serge Savard in this analogy - the guy whom is outperforming the other guy so much that it transcends position.

You're asking "Why move him up front if he was so irreplaceable on the bottom of the box?", and "Why do you think they don't put Gardiner up front to replace Hyman?" as though the questions are asking the same thing - they're not. One is regarding the statistical significance of the actual results of a player who has actually changed positions in a certain circumstance, the other is regarding the hypothetical of moving a player to a different position based on his proficiency in something that's not his primary role, when the only reason that player was even brought up was to make an analogy about statistical significance

Serge Savard was a better forward PKer than Robinson. That's not up for debate. The question was who was the better defenseman PKer, and you don't get that from looking at how much one of them played forward.

The fact that the forward core was so thin with PK options, and the defense core so overstocked, that a defenseman was moved says a lot about how much competition for "PK usage" was at both positions.

 

[seventieslord]

Serge Savard was a better forward PKer than Robinson. That's not up for debate. The question was who was the better defenseman PKer, and you don't get that from looking at how much one of them played forward.

The fact that the forward core was so thin with PK options, and the defense core so overstocked, that a defenseman was moved says a lot about how much competition for "PK usage" was at both positions.

The forward corps wasn't thin, though. Gainey, Roberts, and Jarvis alone are proof of that.

defense was overstocked, sure, in the top-3 at least. It does make sense that they'd want to get three out there at once on the top unit. But it's complete conjecture to speculate that Savard was the one who was chosen because he was the worst at playing D, especially considering he was doing this before Robinson had finished high school.

One thing needs to be made clear here that you may not be aware of - you're probably assuming Robinson was always on the first PK unit playing back, while Savard was up front, and that's your proof he was more valued. Well, the numbers don't really support that.

- In 1974, Robinson was clearly a 2nd unit penalty killer, with Savard, Lapointe and Laperriere carrying the bulk of the weight
- in 1975, Awrey (for 2/3 of a season), Lapointe, Roberts and Savard (likely at F) all played significantly more than him on the PK, he was again a 2nd unit guy
- In 1976, Lapointe and Savard manned the 1st unit with Robinson and Awrey on the 2nd.
- In 1977, Lapointe and Savard both doubled Robinson's usage, he was definitely a 2nd unit guy.
- In 1978, Robinson finally joined the top unit with Savard and Lapointe as a 26 year old (Savard was 32).
- In 1979, he remained there but was still playing significantly less than Savard and Lapointe.
- In 1980, all three were top unit guys again.
- In 1981, Lapointe took a backseat while Robinson and Savard manned the top unit.

There appears to have been only two seasons where savard and Robinson were two of three defensemen with high usage numbers (indicating it was Robinson and another guy playing back, with savard up), but typically what was happening while both were in Montreal was Robinson just wasn't killing many penalties.

 

[jarek]

I'm not certain that Robinson's usage really paints a fair picture of his value as a PK'er. It probably has a lot more to do with Robinson likely being far and away the best ES defenseman for the team and thus getting a great deal of minutes there. Although obviously I can't prove that, but that's my read.

 

[BraveCanadian]

The data tells us that a 6'0" player born in 1900 is approximately 3-4" taller than average in big league hockey around the midpoint of his career. There's nothing wrong with saying that at all, right?

So if we then take that number and add it to today's average in order to give a visual of what that means, why would you have such a problem with it?

For the reasons I stated already regarding adjusting via averages, and because it drives me nuts seeing these numbers cited in ATD profiles of older players.

We do a ton of other great research for all these guys and then people go and stick in Mickey Mouse numbers that do nothing but muddy the water and take away from real finds.

But this is all beside the point here.

I'm not certain that Robinson's usage really paints a fair picture of his value as a PK'er. It probably has a lot more to do with Robinson likely being far and away the best ES defenseman for the team and thus getting a great deal of minutes there. Although obviously I can't prove that, but that's my read.

With how stacked that team was they could afford to play specialists and play people to their strengths (or for greatest advantage over the opposition).. so I would agree with your theory being a good possibility.