ATD 2012 - Draft Thread I

[Dreakmur]

Dreak, those consolidated stats are simply brutal. Just for example: how on earth can you consider Frank Nighbor 4th in consolidated scoring in 1916-17? He and one other guy in the NHA lapped the league in scoring that season. Do you seriously think that there were three better scorers in the PCHL that season? Those stats need to come with an enormous asterisk:

*this is my completely unscientific opinion*

...just so younger posters don't get confused and take that business for some kind of canon.

As much as you were quite rude, I appreciate you asking.

You actually found a typo when I was punching the data in. I had Nighbor with 1 assist that season instead of his 10. That correction bumbs him into 3rd in assists and 1st in points.

Anything else that "simply brutal"?

 

[monster_bertuzzi]

Notes on excellent early 2nd rounders who basically are franchise anchoring selections (1st round quality picks in the 2nd round):

Halifax has Cyclone and Kharlamov! Talk about speed and creativity on transition!

Garnish has Messier and Hall. Now there's strength up the middle.

Glenn Hall is a centre all of a sudden?

 

[Sturminator]

Anything else that "simply brutal"?

Well, your apparent use of unofficial NHA assists, presumably from SIHR, is highly questionable. I would have to dig into the finer details of your finished product to find other questionable decisions. There is such a huge volume of caveats, qualifications, provisos and clarifications that need to be made when presenting this data. I find it brutal that you publish these comparative numbers without all that explanation.

In the ATD, there are old GMs and there are new GMs. The old GMs mostly don't care about these numbers and the new GMs mostly don't understand them. Hockey in the split-league years was a complicated mosaic of shifting players, teams, rules and scoring systems. Some leagues and teams were well-documented, and others hardly at all. Following the rabbit down every hole is simply not possible with such a rough, brute-force system. You paint over far more truth than you illuminate with these numbers. I think that is brutal.

Complicated systems simply don't lend themselves well to simple methods of comparison.

 

[tony d]

Glenn Hall is a centre all of a sudden?

I'm wondering about this, myself.

 

[VanIslander]

Glenn Hall is a centre all of a sudden?

He's up the **** middle. I will cite an undrafted coach's definition of such when it is appropriate to do so.

 

[Stoneberg]

Notes on excellent early 2nd rounders who basically are franchise anchoring selections (1st round quality picks in the 2nd round):

Halifax has Cyclone and Kharlamov! Talk about speed and creativity on transition!

Garnish has Messier and Hall. Now there's strength up the middle.

I thought it may have been a slight reach for Kharlamov at 39 given his usual draft position but couldn't resist putting him with Cyclone. Should be a line that will give opposition teams (especially slower ones) fits once raleh and I decide on who will be the final piece.

 

[TheDevilMadeMe]

Considering where we are in the draft, there is no reason why he has to be the best offensive player on Reen's team. I find your criticism strange, though. Does anyone say that of Bobby Clarke when he is drafted? Nighbor and Clarke are extremely similar in terms of what they bring, and the level at which they bring it. Clarke is more physical and Nighbor's scoring is more balanced, but other than that, there is very little to seperate them.

I would like to see Frank Nighbor make it into the first round one of these days. It has taken us a long time to tease "the truth" out of the legends of old-time hockey, but the more information we have, the clearer it becomes that Nighbor was the greatest player of the pre-consolidation age (meaning before Morenz). This was obviously the consensus of the people who actually saw those players on the ice. In Cyclone Taylor's case, they are difficult to compare because Taylor played most of his best years in the relative obscurity of the west, but the case for Nighbor over Lalonde is open and shut. I have no idea why Newsy continues to be drafted ahead of Frank...probably his last name and the Habs connection. Nighbor rightfully belongs in the Clarke - Esposito tier of all-time centers, or at the very least on the same level with Taylor. It's been a slow climb, but I'm sure he'll eventually get there.

Lalonde was undisputably a better goal scorer than Nighbor. The proven leadership and physicality doesn't hurt either. I don't see why that one is open and shut. Nighbor is easier to build around and therefore might have more "ATD value"

As for the Clarke comparison, I know the statistical case, but we also know that Clarke was the best offensive player on his line and the primary cause for his assists. Nighbor, we aren't quite as sure.

(Nothing I said he disputes the fact that Nighbor was a very good pick. Nighbor at 41 is obviously better value than Clarke at 16).

 

[BraveCanadian]

Glenn Hall is a centre all of a sudden?

You guys do know something about hockey, right?

It is common to refer to goalies as part of the "building up the middle" when constructing teams.

 

[TheDevilMadeMe]

Heh...yeah, a Harvey / Nighbor core has been tried before. It worked out well the first time, though that was back in ATD#11 when Nighbor could still be had far below his actual value. I don't think teams had any idea how to attack that first unit.

Yeah, good luck getting Nighbor in the third round now.

 

[Dreakmur]

Well, your apparent use of unofficial NHA assists, presumably from SIHR, is highly questionable. I would have to dig into the finer details of your finished product to find other questionable decisions. There is such a huge volume of caveats, qualifications, provisos and clarifications that need to be made when presenting this data. I find it brutal that you publish these comparative numbers without all that explanation.

I've gone over the specifics of my project numerous times before....

Basically, I used formulas to equalize games played and goals per game averages. I have just completed formulas to equalize the assists per goals differences.

In the ATD, there are old GMs and there are new GMs. The old GMs mostly don't care about these numbers and the new GMs mostly don't understand them. Hockey in the split-league years was a complicated mosaic of shifting players, teams, rules and scoring systems. Some leagues and teams were well-documented, and others hardly at all. Following the rabbit down every hole is simply not possible with such a rough, brute-force system. You paint over far more truth than you illuminate with these numbers. I think that is brutal.

There's nothing complicated about it. The rules led to higher and lower scoring rates. If you equalize the scoring rates, you have essentilly equalized the different rules.

The leagues were basically equal in terms of talent, especially from 1912 to 1921.

The time period between 1922 and 1924 is a little tricky, since the NHL was likely stronger than each of the Western leagues, but people can account for that as they like.

 

[Sturminator]

I don't see why that one is open and shut.

The observers who watched both men play don't seem to have had your reservations on the subject. Nighbor is mentioned many, many times from all sorts of different sources as being the greatest player of all time as late as the 50's, even drawing favorable comparisons to Morenz. Perhaps I am mistaken, but I am not aware of anyone who held Lalonde in that kind of esteem. There is no one piece of evidence that paints the whole picture, but the preponderance of documentation that Nighbor was considered the greatest of all time is pretty compelling.

The MacLeans top players of all-time 1925 list has Nighbor as the best center of all time and Lalonde third. We can give Taylor a break in the comparison because he spent most of his best years out west and a lot of time at left wing, but don't you think it is a rather telling data point vis-a-vis Nighbor vs. Lalonde?

As far as Nighbor being the best offensive player on his line - at the very least, this was the case in 1916-17, his best offensive season. Over the rest of his career, yes, it is less clear. But Bobby Clarke only had two really high peak offensive seasons, himself (twice finishing 2nd in points). Going be offense only, you wouldn't want Clarke to be your best ATD player, either, and that's before we get into playoff records. Nighbor led all players in postseason scoring twice, iirc, while Clarke never did.

 

[chaosrevolver]

You guys do know something about hockey, right?

It is common to refer to goalies as part of the "building up the middle" when constructing teams.

First I've heard of it.

When referring to goalies, I'm used to hearing "building from the net out"..of course that wouldn't refer to Messier at all.

 

[TheDevilMadeMe]

Well, your apparent use of unofficial NHA assists, presumably from SIHR, is highly questionable. I would have to dig into the finer details of your finished product to find other questionable decisions. There is such a huge volume of caveats, qualifications, provisos and clarifications that need to be made when presenting this data. I find it brutal that you publish these comparative numbers without all that explanation.

In the ATD, there are old GMs and there are new GMs. The old GMs mostly don't care about these numbers and the new GMs mostly don't understand them. Hockey in the split-league years was a complicated mosaic of shifting players, teams, rules and scoring systems. Some leagues and teams were well-documented, and others hardly at all. Following the rabbit down every hole is simply not possible with such a rough, brute-force system. You paint over far more truth than you illuminate with these numbers. I think that is brutal.

Complicated systems simply don't lend themselves well to simple methods of comparison.

Good to see I'm not the only one who considers the uncritical use of "reconstructed assists" to be a questionable practice.

I think Dreakmur's study could have its use as a rough guide of these players, but he really does need to give an explanation as to what those numbers mean, so nobody is tricked into taking "consolidated stats" as something that actually happened.

As far as I know, you've never publicly shown your work, have you Dreakmur? Doing so would definitely leave it open for criticism, but such criticism could help make it better. Without a better idea of where those numbers come from, I really don't pay them much attention.

 

[Hawkey Town 18]

That Clancy pick from DoMakc really hurt my plans. IMO he was easily the best Dman available. I think I know what I'm going to do, but I need a little more time to research. My pick should come in the next couple hours.

 

[TheDevilMadeMe]

The observers who watched both men play don't seem to have had your reservations on the subject. Nighbor is mentioned many, many times from all sorts of different sources as being the greatest player of all time as late as the 50's, even drawing favorable comparisons to Morenz. Perhaps I am mistaken, but I am not aware of anyone who held Lalonde in that kind of esteem. There is no one piece of evidence that paints the whole picture, but the preponderance of documentation that Nighbor was considered the greatest of all time is pretty compelling.

The MacLeans top players of all-time 1925 list has Nighbor as the best center of all time and Lalonde third. We can give Taylor a break in the comparison because he spent most of his best years out west and a lot of time at left wing, but don't you think it is a rather telling data point vis-a-vis Nighbor vs. Lalonde?

As far as Nighbor being the best offensive player on his line - at the very least, this was the case in 1916-17, his best offensive season. Over the rest of his career, yes, it is less clear. But Bobby Clarke only had two really high peak offensive seasons, himself (twice finishing 2nd in points). Going be offense only, you wouldn't want Clarke to be your best ATD player, either, and that's before we get into playoff records. Nighbor led Cup winners in scoring twice, iirc, while Clarke never did.

Well, in an ideal world, Bobby Clarke isn't your best offensive player either. Maybe I'm spoiled by having Howe and Richard in the past two drafts, though.

 

[monster_bertuzzi]

That Clancy pick from DoMakc really hurt my plans. IMO he was easily the best Dman available. I think I know what I'm going to do, but I need a little more time to research. My pick should come in the next couple hours.

There are 4 or 5 defenceman from differen't era's still left on the board who are on par (or better) with Clancy in my gods honest opinion..

 

[TheDevilMadeMe]

That Clancy pick from DoMakc really hurt my plans. IMO he was easily the best Dman available. I think I know what I'm going to do, but I need a little more time to research. My pick should come in the next couple hours.

Clancy was a great pick from a talent perspective, but his lack of size could be an issue in the ATD.

 

[TheDevilMadeMe]

There are 4 or 5 defenceman from differen't era's still left on the board who are on par (or better) with Clancy in my gods honest opinion..

You must have a really dim opinion of pre-World War 2 hockey to say such a thing.

 

[monster_bertuzzi]

You must have a really dim opinion of pre-World War 2 hockey to say such a thing.

You're making me seem like the anti-christ. I don't understand this sudden surge up the rankings for Clancy.

 

[Dreakmur]

Good to see I'm not the only one who considers the uncritical use of "reconstructed assists" to be a questionable practice.

Since when are they reconstructed numbers? They started recording assists in the PCHA in 1913, and in the NHA in 1914. They are official records, not reconstructed.

I think Dreakmur's study could have its use as a rough guide of these players, but he really does need to give an explanation as to what those numbers mean, so nobody is tricked into taking "consolidated stats" as something that actually happened.

They are designed to be a rough guide.

With the split leagues, it's nice to see what effect that had on scoring finishes.

As far as I know, you've never publicly shown your work, have you Dreakmur? Doing so would definitely leave it open for criticism, but such criticism could help make it better. Without a better idea of where those numbers come from, I really don't pay them much attention.

I definately have shown the specifics of the study - both here and on LC.

The Basics:
NHA and PCHA played different # of games, so the league that played more game has an advantage in point totals. League with lower games played has their stats boosted by the difference.

NHA and PCHA had different goals per game averages each season. The players in the higher scoring league had an advantage, so the lower scoring league has their stats boosted.

NHA and PCHA handed out assists at a different rate. The league with the lower assist per goal gets a boost to their assist totals.

 

[Sturminator]

Regarding Dreak's project, I'd like to see the specific formulas and a list of the sources of data, at the very least. Before we have that, we can't really debate the merits of his system.

 

[chaosrevolver]

There are 4 or 5 defenceman from differen't era's still left on the board who are on par (or better) with Clancy in my gods honest opinion..

I see maybe 1..perhaps 2 on his level..that's about it.

Clancy was 5"7...in his time..which doesn't make him 5"7 in an ATD view. He was also a fantastic skater , great with the puck and a tremendous leader.

I had him best available..not sure if it is "by far" however.

 

[nik jr]

Lalonde was undisputably a better goal scorer than Nighbor. The proven leadership and physicality doesn't hurt either. I don't see why that one is open and shut. Nighbor is easier to build around and therefore might have more "ATD value"

As for the Clarke comparison, I know the statistical case, but we also know that Clarke was the best offensive player on his line and the primary cause for his assists. Nighbor, we aren't quite as sure.

(Nothing I said he disputes the fact that Nighbor was a very good pick. Nighbor at 41 is obviously better value than Clarke at 16).

i read more than a little about lalonde when researching nighbor, and lalonde's physicality could hurt. physicality has certainly been used against eddie shore, for example.

lalonde was an extremely dirty player.

in ATD 2010, i quoted a post of yours about a funny incident of a cheapshot. lalonde intentionally bashed nighbor in the face with his stick, which knocked nighbor out of the game, but lalonde then apologized b/c he thought it was a different player.

those kinds of incidents were not so uncommon for lalonde. 70s quoted sturminator's report about lalonde's faceoff trick called "the scratch." lalonde would rake his stick across the opponent's face. nighbor said he hardly had any eyebrows left b/c of lalonde.



i will also add that my bio of nighbor has some nice things about his leadership, in games, off the ice and in practice.

 

[chaosrevolver]

You're making me seem like the anti-christ. I don't understand this sudden surge up the rankings for Clancy.

Ah..I remember you saying something similar about Nighbor last draft.

I hope Domakc can change your mind then with a really solid bio!

 

[monster_bertuzzi]

Ah..I remember you saying something similar about Nighbor last draft.

I hope Domakc can change your mind then with a really solid bio!

What did I say about Nighbor? I didn't think he was good enough offensively to be picked so high that GM's were building their forward group around him, but he has a place in the top 50.

There is no reason for Hawkey town to be bummed about not getting Clancy when IMO there are two D on the board better - and others who are comparable.