HFBoards

Go Back   HFBoards > General Hockey Discussion > By The Numbers
By The Numbers Hockey Analytics... the Final Frontier. Explore strange new worlds, to seek out new algorithms, to boldly go where no one has gone before.

Scouting, Statistics, and St. Louis

Reply
 
Thread Tools
Old
08-22-2012, 09:07 PM
  #51
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Not Really

Quote:
Originally Posted by Iain Fyffe View Post
If that's why they wrote him off, they put too much emphasis on those things.
The scouts grade all players by the same criteria. Analogy to education. A class may have 20 assignments and 20 tests during a school year. Certain students may not hand in 5 assignments and skip 5 tests. Evaluations will take this into account in various fashions. Some simply fail the student, others generate a score where the student may pass with a C when his submitted work projects to an A+, while others may ignore the missed work and give the student an A+. Projectionating fits the analogy. Your system tends to the third.

Really no different that considering a player with a history of injury. Two of the top three picks in the 2012 NHL entry Draft were coming of major injuries.With advances in sports medicine, teams are comfortable taking the chance. In the past teams were not willing to take such chances - evidenced by Dino Ciccarelli.

BTW your 1994 projectionating completely missed Tim Thomas as a top 30 - a teammate of St. Louis and Perrin at the University of Vermont. You rank Jose Theodore, Eric Fichaud and Dan Cloutier in the top 30. Yet the NHL scouting saw Thomas drafted in the 9th round by the Nordiques. Tim Thomas has a very late NHL entry, high peak with multiple honours. That's the way it goes - win some, lose some.


Last edited by Canadiens1958: 08-22-2012 at 09:08 PM. Reason: typo
Canadiens1958 is offline   Reply With Quote
Old
08-22-2012, 09:30 PM
  #52
Irato99
Registered User
 
Join Date: Nov 2010
Posts: 168
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
Originally Posted by Iain Fyffe View Post
If that's why they wrote him off, they put too much emphasis on those things.


The scouts grade all players by the same criteria. Analogy to education. A class may have 20 assignments and 20 tests during a school year. Certain students may not hand in 5 assignments and skip 5 tests. Evaluations will take this into account in various fashions. Some simply fail the student, others generate a score where the student may pass with a C when his submitted work projects to an A+, while others may ignore the missed work and give the student an A+. Projectionating fits the analogy. Your system tends to the third.

Really no different that considering a player with a history of injury. Two of the top three picks in the 2012 NHL entry Draft were coming of major injuries.With advances in sports medicine, teams are comfortable taking the chance. In the past teams were not willing to take such chances - evidenced by Dino Ciccarelli.

BTW your 1994 projectionating completely missed Tim Thomas as a top 30 - a teammate of St. Louis and Perrin at the University of Vermont. You rank Jose Theodore, Eric Fichaud and Dan Cloutier in the top 30. Yet the NHL scouting saw Thomas drafted in the 9th round by the Nordiques. Tim Thomas has a very late NHL entry, high peak with multiple honours. That's the way it goes - win some, lose some.
I think you are a bit off topic guys, remember...Wayne Gretzky?

Irato99 is offline   Reply With Quote
Old
08-22-2012, 09:35 PM
  #53
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Projectinating

Quote:
Originally Posted by Irato99 View Post
I think you are a bit off topic guys, remember...Wayne Gretzky?
Very on topic. Projectionating results. Gretzky from the 1980s to today is not different than projectinating a draft eligible player in 1984 or before or after. Its about understanding and appreciating the process involved.

Canadiens1958 is offline   Reply With Quote
Old
08-22-2012, 09:36 PM
  #54
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
The scouts grade all players by the same criteria. Analogy to education. A class may have 20 assignments and 20 tests during a school year. Certain students may not hand in 5 assignments and skip 5 tests. Evaluations will take this into account in various fashions. Some simply fail the student, others generate a score where the student may pass with a C when his submitted work projects to an A+, while others may ignore the missed work and give the student an A+. Projectionating fits the analogy. Your system tends to the third.
Projectinating. I made up the word, so I get to say what it is!

And if using another system has some examples of abject failure, such system should be placed under strenuous review.

Quote:
Originally Posted by Canadiens1958 View Post
BTW your 1994 projectionating completely missed Tim Thomas as a top 30 - a teammate of St. Louis and Perrin at the University of Vermont. You rank Jose Theodore, Eric Fichaud and Dan Cloutier in the top 30. Yet the NHL scouting saw Thomas drafted in the 9th round by the Nordiques. Tim Thomas has a very late NHL entry, high peak with multiple honours. That's the way it goes - win some, lose some.
Those three were drafted in the top 44 picks in 1994, so my having them in the top 30 seems pretty fair. As for Thomas, you don't have enough information to say where I have him ranked, since you know only the top 30. As it happens his rating is 4.0, which would put him at somewhere around #80-100 I believe, which as you might notice is higher than #217.

You're criticizing my system for missing Thomas as a top 30, while the scouts also missed it. That's hardly damning evidence that the system doesn't compare well to scouts.

All that aside, I think I've only mentioned several hundred times since I started this work that numerical systems will be better on some players than scouts are, worse on others, and close to the same on others. Go way back to the point that started all this, that using one's eyes cannot be improved upon by also using stats. The best approach, to my mind, is to use both. Hopefully the numbers would help scouts (and vice-versa) identify the players that might be falling down the rankings unfairly.

Iain Fyffe is offline   Reply With Quote
Old
08-22-2012, 09:38 PM
  #55
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Irato99 View Post
I think you are a bit off topic guys, remember...Wayne Gretzky?
It was a pretty natural evolution from the discussion about Gretzky. It happens. You're liable to restrict interesting discussion if you put hard boundaries on it.

Iain Fyffe is offline   Reply With Quote
Old
08-22-2012, 10:30 PM
  #56
Hardyvan123
tweet@HardyintheWack
 
Join Date: Jul 2010
Location: Vancouver
Country: Canada
Posts: 11,386
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
Yes but they did not get to see them play a 60+ game season, any significant international games or within and against varied or complex systems.
Forgive me if I misread one of your earlier posts but didn't you say that you could tell if a guy was going to be a star in a couple of shifts or something?

Hardyvan123 is offline   Reply With Quote
Old
08-22-2012, 10:32 PM
  #57
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Hardyvan123 View Post
Forgive me if I misread one of your earlier posts but didn't you say that you could tell if a guy was going to be a star in a couple of shifts or something?
Ooh, I missed that. Interesting question.

Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 12:43 AM
  #58
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Misread

Quote:
Originally Posted by Hardyvan123 View Post
Forgive me if I misread one of your earlier posts but didn't you say that you could tell if a guy was going to be a star in a couple of shifts or something?
You did misread and misquote. Goalies do not play shifts.

The actual quote had a key sentence that you left out. "Goalies are trickier."

Post #384 carries the actual complete quote and follow-up explanation.

Canadiens1958 is offline   Reply With Quote
Old
08-23-2012, 07:21 AM
  #59
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
You did misread and misquote. Goalies do not play shifts.

The actual quote had a key sentence that you left out. "Goalies are trickier."
No, not Thomas; St. Louis. If it only takes a few shifts or even games to figure a skater out, then why does it matter that scout didn't see St. Louis play 60+ games etc?

Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 07:44 AM
  #60
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Martin St.Louis

Quote:
Originally Posted by Iain Fyffe View Post
No, not Thomas; St. Louis. If it only takes a few shifts or even games to figure a skater out, then why does it matter that scout didn't see St. Louis play 60+ games etc?
Pre midget it was obvious to an experienced observer that Martin St.Louis had the skating ability and the hockey sense and skills to eventually play in the NHL and that he was one of the few players from his team worth tracking thru the next levels. But these qualities were also observable throughout the hockey world for other players from the 1994 NHL Entry Draft eligible class.

Then it becomes a question of doing all the follow-up work, tracking and rating all the players over the next 4-5 seasons.
The more complete a players background becomes the more revealing the ratings become as the seasons move towards the draft in question.

The issue is not drafting from the limited bantam team pool that St.Louis played for but drafting from the extensive draft pool covering the hockey playing world.

Canadiens1958 is offline   Reply With Quote
Old
08-23-2012, 08:04 AM
  #61
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
Then it becomes a question of doing all the follow-up work, tracking and rating all the players over the next 4-5 seasons.
The more complete a players background becomes the more revealing the ratings become as the seasons move towards the draft in question.
Post 384, as you mentioned, contains the comment in question. You told us that it takes only 1 or 2 shifts to see what you need to see in great players, their "skating, execution, hockey sense." Excellent to average players, you told us, can be evaluated in 3 to 6 shifts, below average 6 to 10.

This was all in response to the criticism of eyes, that you can't see all players all the time; that stats, at least, capture what the player did in every game regardless of whether you were there to see it or not. So when it comes to numbers covering many more games than eyes, your response is that you don't need to see many games to evaluate a player.

But when it comes to a player like St. Louis, you tell us that his low (non-existant even?) rating by scouts was justified, because he didn't play as many games as many other players who were being evaluated. Why would this matter, if you only need 6 to 10 shifts to evaluate a player?

It's not like St. Louis was ranked a little lower than he should have been. Competition with other top available players is not an issue, because there were far more than enough draft picks available to take all of the great players, all of the very good ones, all the above-average ones, and many below-average. It's not like he was drafted 50th overall. What actually happened was that in his two years of draft eligibility, 520 players were selected and he was not one of them.

If you respond to this post, please do so by directly answering this question: why does it matter how many games St. Louis played, if you only need a few shifts to evaluate him?

Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 08:36 AM
  #62
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Evaluations

Quote:
Originally Posted by Iain Fyffe View Post
Post 384, as you mentioned, contains the comment in question. You told us that it takes only 1 or 2 shifts to see what you need to see in great players, their "skating, execution, hockey sense." Excellent to average players, you told us, can be evaluated in 3 to 6 shifts, below average 6 to 10.

This was all in response to the criticism of eyes, that you can't see all players all the time; that stats, at least, capture what the player did in every game regardless of whether you were there to see it or not. So when it comes to numbers covering many more games than eyes, your response is that you don't need to see many games to evaluate a player.

But when it comes to a player like St. Louis, you tell us that his low (non-existant even?) rating by scouts was justified, because he didn't play as many games as many other players who were being evaluated. Why would this matter, if you only need 6 to 10 shifts to evaluate a player?

It's not like St. Louis was ranked a little lower than he should have been. Competition with other top available players is not an issue, because there were far more than enough draft picks available to take all of the great players, all of the very good ones, all the above-average ones, and many below-average. It's not like he was drafted 50th overall. What actually happened was that in his two years of draft eligibility, 520 players were selected and he was not one of them.

If you respond to this post, please do so by directly answering this question: why does it matter how many games St. Louis played, if you only need a few shifts to evaluate him?
You only need a few shifts to evaluate a bantam/midget player's NHL potential. The rest is follow-up to see how the potential is developed and compares to the rest of the draft eligible players for a specific year.

Your point about the 520 players is strictly Monday morning quarterbacking. At the bantam/midget age they were contemporaries of Martin St. Louis and elite players that stood out on their team. Your rating of them from great to below average overlooks this aspect of the process by avoiding the the issue of how and why their status went from elite on a bantam/midget team to their NHL draft level.

You can go observe the NB/PEI midget AAA - Fredericton has a team, or access the individual stats:

http://www.midgetaaacanada.com/leagu...leagueid=10987

and go thru the process leading up to the QMJHL Midget Draft and the NHL Entry Draft. You will see first hand how elite at the Midget AAA level goes to not drafted at the NHL Entry Draft.

Canadiens1958 is offline   Reply With Quote
Old
08-23-2012, 08:44 AM
  #63
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
You only need a few shifts to evaluate a bantam/midget player's NHL potential. The rest is follow-up to see how the potential is developed and compares to the rest of the draft eligible players for a specific year.
So the comment about a few shifts was not relevant to the discussion, which was about evaluating draft-age players for the NHL Entry Draft.

So now we're back to: if you need more than a few shifts/games to evaluate a player properly, how do you account for the fact that you cannot see every player in every game he plays, or even most games he plays, without neglecting all the other games being played at the same time? This is circling back to the eyes v. numbers issue, which prompted your comment in the first place.

Stats have the distinct advantage over eyes of not needing your presence at the particular game to have meaning. Which is why both should be used, which was my point to the comment that started all this (not by you) that eyes cannot be improved upon by using numbers.

I was not discussing why an elite Midget AAA player goes on to not be drafted. I was discussing why an elite NCAA Div. I player was not drafted.

Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 09:42 AM
  #64
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Elite NCAA

Quote:
Originally Posted by Iain Fyffe View Post
So the comment about a few shifts was not relevant to the discussion, which was about evaluating draft-age players for the NHL Entry Draft.

So now we're back to: if you need more than a few shifts/games to evaluate a player properly, how do you account for the fact that you cannot see every player in every game he plays, or even most games he plays, without neglecting all the other games being played at the same time? This is circling back to the eyes v. numbers issue, which prompted your comment in the first place.

Stats have the distinct advantage over eyes of not needing your presence at the particular game to have meaning. Which is why both should be used, which was my point to the comment that started all this (not by you) that eyes cannot be improved upon by using numbers.

I was not discussing why an elite Midget AAA player goes on to not be drafted. I was discussing why an elite NCAA Div. I player was not drafted.
Well, this simplifies things quite a bit. Look where a very solid OHL player - Steve Sullivan was drafted in the 1994 NHL Entry Draft:

http://www.hockey-reference.com/play...sullist01.html

http://www.hockey-reference.com/draf...994_entry.html

late 9th round.

Sullivan had better lead-up numbers in tougher leagues than St. Louis did in the leagues he played in. Yet your evaluation does not include Steve Sullivan who was/is 5'9" / 161 lbs, in the top 30 nor does your evaluation include the following late round picks in the top 30, goalies,Thomas, Hedberg, Nabokov, Vokoun, skaters Holmstrom, Boulton, Zednick, Berezin,Tarnstrom, Johnsson.

Miss rating 10+ and overrating 1 - Martin St.Louis is hardly an endorsement of your method/system.


Last edited by Canadiens1958: 08-23-2012 at 09:50 AM. Reason: addition
Canadiens1958 is offline   Reply With Quote
Old
08-23-2012, 11:20 AM
  #65
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
Well, this simplifies things quite a bit. Look where a very solid OHL player - Steve Sullivan was drafted in the 1994 NHL Entry Draft:

http://www.hockey-reference.com/play...sullist01.html

http://www.hockey-reference.com/draf...994_entry.html

late 9th round.

Sullivan had better lead-up numbers in tougher leagues than St. Louis did in the leagues he played in.
Sullivan's OHL numbers are not better than St. Louis' ECAC numbers. That's the whole point of the system: a sophisticated analysis of which numbers are better. They may appear better superficially, but if you dig into them, they're not. That's what the system does: it digs into the numbers, to see if they're really as impressive as they might appear at first glance.

Steve Sullivan's numbers are simply not impressive, once you factor in his age. His career certainly exceeded what a numerical projection would have pegged him for. Of course, given that he was drafted 233rd overall in his final year of draft eligibility, his career also certainly exceeded what the scouts pegged him for.

Quote:
Originally Posted by Canadiens1958 View Post
Yet your evaluation does not include Steve Sullivan who was/is 5'9" / 161 lbs, in the top 30 nor does your evaluation include the following late round picks in the top 30, goalies,Thomas, Hedberg, Nabokov, Vokoun, skaters Holmstrom, Boulton, Zednick, Berezin,Tarnstrom, Johnsson.

Miss rating 10+ and overrating 1 - Martin St.Louis is hardly an endorsement of your method/system.
Once again you're talking about players not in the top 30 of all players that year, as if the system believes they're not worth drafting. You're arguing from ignorance, because you're ignorant of what each of these players' ratings are.

On top of that fact, you're going about this wrongly in at least two major ways:

1. You're listing a bunch of players the scouts did not rate highly (drafted in the 200s), and then claiming the fact that my system does not rate them highly (although you don't actually know where it rates them) is a point against it. If this is a point against the system, it's also a point against scouting.

What you should be doing is looking at players that the system rates low but scouts rate high, and who turn out to have good careers. There are a good number of those.

Unfortunately, there are also a good number of players in just the opposite situation: rated low by the scouts, high by the system, good careers. Which leads to point #2.

2. You can't evaluate anything by just looking at a few players in a single draft year. Your analysis has to be thorough and systematic in order for it to be valid.

So, as for your 10/1 comment: you have no idea if those 10+ players are mis-rated, because you don't know where they're rated. And now you're saying that Martin St. Louis should not have been rated one of the best players available in 1994? I think his career suggests otherwise.

(Also, Eric Boulton? Really?)


Last edited by Iain Fyffe: 08-23-2012 at 11:21 AM. Reason: Grammar
Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 11:59 AM
  #66
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Specifics

Quote:
Originally Posted by Iain Fyffe View Post
Sullivan's OHL numbers are not better than St. Louis' ECAC numbers. That's the whole point of the system: a sophisticated analysis of which numbers are better. They may appear better superficially, but if you dig into them, they're not. That's what the system does: it digs into the numbers, to see if they're really as impressive as they might appear at first glance.

Steve Sullivan's numbers are simply not impressive, once you factor in his age. His career certainly exceeded what a numerical projection would have pegged him for. Of course, given that he was drafted 233rd overall in his final year of draft eligibility, his career also certainly exceeded what the scouts pegged him for.


Once again you're talking about players not in the top 30 of all players that year, as if the system believes they're not worth drafting. You're arguing from ignorance, because you're ignorant of what each of these players' ratings are.

On top of that fact, you're going about this wrongly in at least two major ways:

1. You're listing a bunch of players the scouts did not rate highly (drafted in the 200s), and then claiming the fact that my system does not rate them highly (although you don't actually know where it rates them) is a point against it. If this is a point against the system, it's also a point against scouting.

What you should be doing is looking at players that the system rates low but scouts rate high, and who turn out to have good careers. There are a good number of those.

Unfortunately, there are also a good number of players in just the opposite situation: rated low by the scouts, high by the system, good careers. Which leads to point #2.

2. You can't evaluate anything by just looking at a few players in a single draft year. Your analysis has to be thorough and systematic in order for it to be valid.

So, as for your 10/1 comment: you have no idea if those 10+ players are mis-rated, because you don't know where they're rated. And now you're saying that Martin St. Louis should not have been rated one of the best players available in 1994? I think his career suggests otherwise.

(Also, Eric Boulton? Really?)
Your comments lack specifics. Name the players that support your contentions like I have named the players that illustrate my points.

The 10 not rated in the top 30 are simply red flags given that at least 9 players you list cannot justify their rating in your top 30 given the glaring omissions from the middle - which I have not listed, or late rounds.

Rather obvious comparing the dates of birth Sullivan July 6, 1974, St. Louis June 1975, Jose Theodore September 16, 1976, Limited sampling to illustrate, that your system rates the date of birth in the draft year as more important than most evaluation systems do.

Eric Boulton has over 600 NHL games for better or worse and was an August 17,1976 birth, 1 month older than Theodore, so he should have been on your radar.

Canadiens1958 is offline   Reply With Quote
Old
08-23-2012, 12:20 PM
  #67
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
Your comments lack specifics. Name the players that support your contentions like I have named the players that illustrate my points.
Go to Hockey Prospectus. There's a whole series of articles there, as I've mentioned before. Read to your heart's content.

Quote:
Originally Posted by Canadiens1958 View Post
The 10 not rated in the top 30 are simply red flags given that at least 9 players you list cannot justify their rating in your top 30 given the glaring omissions from the middle - which I have not listed, or late rounds.
What you seem to suggesting is that since Eric Boulton is not listed in the top 30 players that year, there's a problem with the system.

If these players are red flags for the system, they're also red flags for scouting. You keep missing that point. You can't point to players that the system misses if the scouts also miss them, as evidence that the system is inferior to scouting.

Quote:
Originally Posted by Canadiens1958 View Post
Rather obvious comparing the dates of birth Sullivan July 6, 1974, St. Louis June 1975, Jose Theodore September 16, 1976, Limited sampling to illustrate, that your system rates the date of birth in the draft year as more important than most evaluation systems do.
Now this is entirely possible. But it's also easily justified. A player's birth date is extremely important when evaluating his junior numbers. Just massively important. If you don't properly consider the player's age, you will see impressive numbers where they're actually humdrum. This could be a big part of the problem here: you're greatly underrating the importance of age in these evaluations.

If you study the numbers in depth, this is one of the biggest lessons you'll learn: a player's age is one of the most important factors to consider when evaluating his numbers.

And it's also extremely important to consider which league the player is in when looking at his age. Being 19 years old in major junior has a very different effect on the raw stats than does being 19 years old in the NCAA.

Quote:
Originally Posted by Canadiens1958 View Post
Eric Boulton has over 600 NHL games for better or worse and was an August 17,1976 birth, 1 month older than Theodore, so he should have been on your radar.
He has a rating. It's just a very low one. Much like the scouts had for him apparently.

Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 03:15 PM
  #68
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Date of Birth

Quote:
Originally Posted by Iain Fyffe View Post
Go to Hockey Prospectus. There's a whole series of articles there, as I've mentioned before. Read to your heart's content.


What you seem to suggesting is that since Eric Boulton is not listed in the top 30 players that year, there's a problem with the system.

If these players are red flags for the system, they're also red flags for scouting. You keep missing that point. You can't point to players that the system misses if the scouts also miss them, as evidence that the system is inferior to scouting.


Now this is entirely possible. But it's also easily justified. A player's birth date is extremely important when evaluating his junior numbers. Just massively important. If you don't properly consider the player's age, you will see impressive numbers where they're actually humdrum. This could be a big part of the problem here: you're greatly underrating the importance of age in these evaluations.

If you study the numbers in depth, this is one of the biggest lessons you'll learn: a player's age is one of the most important factors to consider when evaluating his numbers.

And it's also extremely important to consider which league the player is in when looking at his age. Being 19 years old in major junior has a very different effect on the raw stats than does being 19 years old in the NCAA.


He has a rating. It's just a very low one. Much like the scouts had for him apparently.
Players have to impress by the age of 17-18 in terms of the draft. 19 is too old.

Daniel Alfredsson was a 6th round pick as a 22 year old in the 1994 draft. No one picked him in the draft when he was first eligible.Neither was Nicklas Lidstrom.

Nothing new about the age factor - when the player was born during the hockey year ,In the sixties, the organization I played for and many other used this to balance house league teams Year was divided into quarters and the teams were balanced in proportion. Hockey organizations and jurisdictions recognized this in the seventies.Lead by Ontario, categories were divided into minor - first year and major - second year in a category.At the novice level the gap is much larger than at the junior level.

Also true in education and acceptance to advanced programs - date of birth in the academic year is a strong determining factor.

Scouts recognized this years ago. The gentleman that ran our organization scouted the Quebec minor leagues and juniors.

Scouts and organized hockey are well aware of the weaknesses inherent in their models and systems. They are also aware of their successes.

Canadiens1958 is offline   Reply With Quote
Old
08-23-2012, 03:17 PM
  #69
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
Your point about the 520 players is strictly Monday morning quarterbacking.
No it's not. As I said, numerical analysis could have been done at the time, using information available at the time, that would indicate St. Louis (and Perrin) were players you really wanted to look at.

I took another hour today and adapted the QMJHL method I developed yesterday, this time using ECAC player data from 1980-1982. All of the data used existed, and was easily available, in 1994 and 1995. Again, the output in this case is peak (best 5 of 10 seasons) points-per-game production at the NHL level, assuming a scoring environment of 3.75 goals per game. Results for contemporaries of St. Louis and Perrin:

PlayerSeasonEst PPG
St Louis M19951.22
Perrin E19951.21
St Louis M19940.98
Perrin E19940.77
Palmer S19940.55
Martins S19920.54
Fogarty R19920.52
Ducharme D19930.50
Majic X19930.48
Ducharme D19920.48
Richardson B19930.47
Sancimino M19940.21
Majic X19920.19
Perreault N1992.018
Murphy B19930.18
Clark W19920.17
Houle J19940.16
Regan T19940.16
Tuomainen M19920.15
Regan T19930.15
Palmer S19930.15

So now we can stop pretending that numerical analysis could not have been done, using only information available at the time, that would show the seasons put up by St. Louis and Perrin to be as impressive as they actually are.

Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 03:21 PM
  #70
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
Players have to impress by the age of 17-18 in terms of the draft. 19 is too old.
This is largely true for major junior players (which is why Steve Sullivan's numbers are not impressive). But it's not at all true for NCAA players, who are playing against men several years older than them. There are no 22- and 23-year-old CHLers, but there are plenty of NCAA players of that age, and that's why a 19-year-old NCAAer who produces is a viable prospect.

And of course, in St. Louis' first NCAA season he was 18. Still time to impress, apparently.

Quote:
Originally Posted by Canadiens1958 View Post
Scouts and organized hockey are well aware of the weaknesses inherent in their models and systems. They are also aware of their successes.
I am also well aware of the weaknesses inherent in my model, and you'll certainly grant me these. I am also, however, aware of its successes, which you will not admit to.


Last edited by Iain Fyffe: 08-23-2012 at 03:30 PM.
Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 03:58 PM
  #71
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Draft Age

Quote:
Originally Posted by Iain Fyffe View Post
This is largely true for major junior players (which is why Steve Sullivan's numbers are not impressive). But it's not at all true for NCAA players, who are playing against men several years older than them. There are no 22- and 23-year-old CHLers, but there are plenty of NCAA players of that age, and that's why a 19-year-old NCAAer who produces is a viable prospect.

And of course, in St. Louis' first NCAA season he was 18. Still time to impress, apparently.


I am also well aware of the weaknesses inherent in my model, and you'll certainly grant me these. I am also, however, aware of its successes, which you will not admit to.
But the first eligible draft age is when a player is 17-18, not 19.Why a sophmore 19 year old in the NCAA does not get drafted while a 19 year old + CHL or European player gets drafted is not strictly a question of talent but a question of signability. The 19 year old CHL player or the European drafted in June could be in a team's September training camp. St. Louis and Perrin were set on finishing their NCAA careers so NHL teams were not willing to wait.

Now if you look at the 1995 draft, the number and quality of post 6th round successes is much lower than the 1994 draft. So the scouts and NHL GMs obviously adapted.

At best you have enjoyed some success with St. Louis balanced against the listed bumps in the road that I provided. Do the same for some of the players like Lidstrom - older than the 1988 first overall pick Mike Modano or Daniel Alfredsson and you will add a few points to your scoreboard. Project some of the undrafted 2012 players and we'll see roughly 3 - 7 years down the road.


Last edited by Canadiens1958: 08-23-2012 at 04:00 PM. Reason: addition
Canadiens1958 is offline   Reply With Quote
Old
08-23-2012, 04:14 PM
  #72
Iain Fyffe
Hockey fact-checker
 
Iain Fyffe's Avatar
 
Join Date: Feb 2009
Location: Fredericton, NB
Country: Canada
Posts: 2,172
vCash: 500
Quote:
Originally Posted by Canadiens1958 View Post
St. Louis and Perrin were set on finishing their NCAA careers so NHL teams were not willing to wait.
This must be why teams never draft players who complete their NCAA careers, right? If what you're saying here is true (which it isn't), NCAA players would rarely be drafted.

St. Louis could not have been drafted in his age-17 season, of course. In order to maintain his NCAA eligibility one assumes he did not opt in.

Quote:
Originally Posted by Canadiens1958 View Post
Now if you look at the 1995 draft, the number and quality of post 6th round successes is much lower than the 1994 draft. So the scouts and NHL GMs obviously adapted.
And if you look at the 1995 draft, St. Louis' number were that much more impressive than even his very impressive 1994. And any conclusion that anything significant changed from one draft to the next, based on a few players selected in the late rounds, is pure assumption. You don't have nearly enough evidence to make such a claim.

Quote:
Originally Posted by Canadiens1958 View Post
At best you have enjoyed some success with St. Louis balanced against the listed bumps in the road that I provided.
With St. Louis and many other players, you mean. You have read all the stuff available at Hockey Prospectus, yes?

Quote:
Originally Posted by Canadiens1958 View Post
Do the same for some of the players like Lidstrom - older than the 1988 first overall pick Mike Modano or Daniel Alfredsson and you will add a few points to your scoreboard.
How is Henrik Zetterberg as an example? The system ranks him 28th in 1999, a far sight better than the 210th he was actually drafted. Henrik Lundqvist is rated 11th in 2000, Marc Savard 5th in 1995, Hejduk 11th in 1994, Sami Salo 12th in 1993, Pavol Demitra 17th in 1993, Sami Kapanen 8th in 1992. These are just from what's been published at Hockey Prospectus.

I know the state of my scoreboard, because I have more information about the rankings than you do. It's understandable that you don't know all the ratings, what's not defensible is making assumptions about them, when you don't know what they are.

Iain Fyffe is offline   Reply With Quote
Old
08-23-2012, 05:11 PM
  #73
Canadiens1958
Moderator
 
Canadiens1958's Avatar
 
Join Date: Nov 2007
Posts: 10,457
vCash: 500
Signability

Quote:
Originally Posted by Iain Fyffe View Post
This must be why teams never draft players who complete their NCAA careers, right? If what you're saying here is true (which it isn't), NCAA players would rarely be drafted.

St. Louis could not have been drafted in his age-17 season, of course. In order to maintain his NCAA eligibility one assumes he did not opt in.


And if you look at the 1995 draft, St. Louis' number were that much more impressive than even his very impressive 1994. And any conclusion that anything significant changed from one draft to the next, based on a few players selected in the late rounds, is pure assumption. You don't have nearly enough evidence to make such a claim.


With St. Louis and many other players, you mean. You have read all the stuff available at Hockey Prospectus, yes?


How is Henrik Zetterberg as an example? The system ranks him 28th in 1999, a far sight better than the 210th he was actually drafted. Henrik Lundqvist is rated 11th in 2000, Marc Savard 5th in 1995, Hejduk 11th in 1994, Sami Salo 12th in 1993, Pavol Demitra 17th in 1993, Sami Kapanen 8th in 1992. These are just from what's been published at Hockey Prospectus.

I know the state of my scoreboard, because I have more information about the rankings than you do. It's understandable that you don't know all the ratings, what's not defensible is making assumptions about them, when you don't know what they are.
NCAA players or NCAA bound players are drafted by NHL teams after being qualified by the scouts as to whether they are willing to leave early and turn pro if the opportunity exists.

Same is true for Europeans, especially those taken past their first year of eligibility. If a team is willing to wait and have solid contacts they can wait and gamble at the round they feel appropriate.

As for assumptions, that is your interpretation. All that matters is the results.

Canadiens1958 is offline   Reply With Quote
Old
08-23-2012, 05:44 PM
  #74
Leafs87
Mr. Steal Your Job
 
Leafs87's Avatar
 
Join Date: Aug 2010
Location: Richmond Hill
Country: Romania
Posts: 4,253
vCash: 500
For every st luis there is like 50 guys who busted. You can take a chance, but no way can you tell if it is an educated one. players get drafted on IQ, skill, size, strength, etc. Skilled guys always get passed up, usually its for good reasons, some times its a mistake.

Leafs87 is offline   Reply With Quote
Old
08-23-2012, 06:21 PM
  #75
Pietraneglo222
WwWwW
 
Pietraneglo222's Avatar
 
Join Date: Jul 2009
Location: Gatineau
Country: Canada
Posts: 3,271
vCash: 500
I didn't read the whole thread but when it comes to scouting small players, there is a perverse consideration I wanna talk about.

If I'm a scout from a an average NHL team and I wanna make a name for myself. Why would I draft a small player even if I do believe that size isn't that important? I know that:

1) My people will not give him a proper opportunity
2) Even in the event that he is given an opportunity, the coach and the GM will not rate him objectively and he is very likely to discriminated against in the future.
3) if my average-to-good size player busts, the player will take most of the blame. If a small player busts, it'll be charged doubly harshly to my account.

Pietraneglo222 is offline   Reply With Quote
Reply

Forum Jump


Bookmarks

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -5. The time now is 01:34 PM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"
Contact Us - HFBoards - Archive - Privacy Statement - Terms of Use - Advertise - Top - AdChoices

vBulletin Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. 2014 All Rights Reserved.