HFBoards

Go Back   HFBoards > NHL Western Conference > Pacific Division > Vancouver Canucks
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
Notices

Prospect Thread - Part XII

Closed Thread
 
Thread Tools
Old
11-06-2012, 04:55 PM
  #1
Mr. Canucklehead
Mod Supervisor
Kitimat Canuck
 
Mr. Canucklehead's Avatar
 
Join Date: Dec 2002
Location: Kitimat, BC
Country: Canada
Posts: 23,790
vCash: 500
Prospect Thread - Part XII

And the beat goes on. Last thread was at 1,000 posts.

Continue.

Mr. Canucklehead is online now  
Old
11-06-2012, 07:48 PM
  #2
Vankiller Whale
And so it begins
 
Vankiller Whale's Avatar
 
Join Date: May 2012
Location: Toronto
Country: Canada
Posts: 24,777
vCash: 500
Any word on Tommernes? Does he intend on coming to North America any time soon?

Vankiller Whale is online now  
Old
11-06-2012, 10:12 PM
  #3
vanuck
Griffiths Way Goons
 
vanuck's Avatar
 
Join Date: Dec 2009
Country: Hong Kong
Posts: 10,139
vCash: 500
Advanced metrics at lower levels of hockey

Given what we know about advanced stats and how they've proven to be useful for quantifying NHL players' performance, does anyone else think the Canucks implement their own system for stat-tracking like the numbers we see on nhl.com, but at the junior and AHL levels?

You would think they'd have the financial resources to hire enough people to do this at each and every arena at the CHL and AHL levels - all kept hush-hush of course. Obviously they'd need a huge division of people just devoted to measuring statistics.

If it's true that good players drive possession, naturally they would be looking for the kids who have the best underlying numbers, and find value where other teams may not be as clued in about someone, while using the perception of value to trade down and get more picks.

But I was just wondering if this was at all feasible and if it's possible VAN is already doing this. And if so, don't we gain a massive edge in terms of amateur scouting?

Any flaws about such a hypothetical method of finding players?

Does this make sense? Am I going crazy here?

vanuck is online now  
Old
11-06-2012, 10:24 PM
  #4
Nuck This
Registered User
 
Nuck This's Avatar
 
Join Date: May 2011
Location: Vancouver
Country: Canada
Posts: 3,072
vCash: 500
Quote:
Originally Posted by vanuck View Post
Given what we know about advanced stats and how they've proven to be useful for quantifying NHL players' performance, does anyone else think the Canucks implement their own system for stat-tracking like the numbers we see on nhl.com, but at the junior and AHL levels?

You would think they'd have the financial resources to hire enough people to do this at each and every arena at the CHL and AHL levels - all kept hush-hush of course. Obviously they'd need a huge division of people just devoted to measuring statistics.

If it's true that good players drive possession, naturally they would be looking for the kids who have the best underlying numbers, and find value where other teams may not be as clued in about someone, while using the perception of value to trade down and get more picks.

But I was just wondering if this was at all feasible and if it's possible VAN is already doing this. And if so, don't we gain a massive edge in terms of amateur scouting?

Any flaws about such a hypothetical method of finding players?

Does this make sense? Am I going crazy here?
"I feel like I'm taking crazy pills!"

Nuck This is offline  
Old
11-06-2012, 10:26 PM
  #5
Scurr
Registered User
 
Scurr's Avatar
 
Join Date: Jun 2009
Location: Whalley
Country: Canada
Posts: 7,902
vCash: 500
Quote:
Originally Posted by vanuck View Post
Any flaws about such a hypothetical method of finding players?

Does this make sense? Am I going crazy here?
It's a good thought but I'm not sure it would be worth the investment. The problem with drafting amateurs isn't figuring out which ones are making an impact, it's forecasting which ones can make an impact in the NHL.

Scurr is offline  
Old
11-07-2012, 02:26 PM
  #6
Vankiller Whale
And so it begins
 
Vankiller Whale's Avatar
 
Join Date: May 2012
Location: Toronto
Country: Canada
Posts: 24,777
vCash: 500
Quote:
Originally Posted by Vankiller Whale View Post
Any word on Tommernes? Does he intend on coming to North America any time soon?
Anyone know? He's probably the prospect I know least about right now. What's his upside/playstyle?

Vankiller Whale is online now  
Old
11-07-2012, 04:24 PM
  #7
Hansen 36
BASED JIM BENNING
 
Hansen 36's Avatar
 
Join Date: Oct 2011
Location: Nanaimo, B.C.
Country: Canada
Posts: 6,282
vCash: 500
Quote:
Originally Posted by Vankiller Whale View Post
Anyone know? He's probably the prospect I know least about right now. What's his upside/playstyle?
Solid two-way d-man, good offensive skills and is much better defensively now than he was. I'd say his upside is a slightly better version of Alex Sulzer, and if he decides to come over to North America, he'll likely reach that potential.

Hansen 36 is online now  
Old
11-07-2012, 04:51 PM
  #8
pitseleh
Registered User
 
pitseleh's Avatar
 
Join Date: Jul 2005
Location: Vancouver
Country: Canada
Posts: 17,645
vCash: 500
Quote:
Originally Posted by Scurr View Post
It's a good thought but I'm not sure it would be worth the investment. The problem with drafting amateurs isn't figuring out which ones are making an impact, it's forecasting which ones can make an impact in the NHL.
Iain Fyffe's Pucks from the Past is interesting in that regard - he basically constructed a model using G-A-P-+/- accounting for things like league difficulty and team quality and went through and drafted two decades worth of players using a fake team (pretending that the St. Louis Blues moved to Saskatoon in 1983). Draft picks were determined, I believe, based on simulations of the season:

http://www.puckprospectus.com/news/?author=8

He did at least as well as the NHL teams in picking good players, and probably better. For example, from 1997-2000, he got Hossa, Campbell, Markov, Ribeiro, B. Richards, Gionta, Comrie, Zetterberg, and Lundqvist while only picking above #26 once. He also picked good role players like Andrew Ference, Jarret Stoll, and Nick Schultz.

If simple stats like points can tell you that much about the draft, tracking scoring chances, or Corsi at the very least, would probably tell you a lot more. You could avoid the 'broken leg' pheonomenon by eliminating players who clearly have warts (speed, size, smarts) that limit their potential to ever make it to the NHL.

pitseleh is offline  
Old
11-07-2012, 05:29 PM
  #9
Scurr
Registered User
 
Scurr's Avatar
 
Join Date: Jun 2009
Location: Whalley
Country: Canada
Posts: 7,902
vCash: 500
Quote:
Originally Posted by pitseleh View Post
Iain Fyffe's Pucks from the Past is interesting in that regard - he basically constructed a model using G-A-P-+/- accounting for things like league difficulty and team quality and went through and drafted two decades worth of players using a fake team (pretending that the St. Louis Blues moved to Saskatoon in 1983). Draft picks were determined, I believe, based on simulations of the season:

http://www.puckprospectus.com/news/?author=8

He did at least as well as the NHL teams in picking good players, and probably better. For example, from 1997-2000, he got Hossa, Campbell, Markov, Ribeiro, B. Richards, Gionta, Comrie, Zetterberg, and Lundqvist while only picking above #26 once. He also picked good role players like Andrew Ference, Jarret Stoll, and Nick Schultz.

If simple stats like points can tell you that much about the draft, tracking scoring chances, or Corsi at the very least, would probably tell you a lot more. You could avoid the 'broken leg' pheonomenon by eliminating players who clearly have warts (speed, size, smarts) that limit their potential to ever make it to the NHL.
Cool. I'm not surprised at all, I think overall hockey scouting is pretty poor. I wonder how he'd do against the better teams? I wonder if the better teams are already using something like this?

One of the biggest problems I have with Corsi in junior (and it can apply in the NHL as well) is that coaching and systems have such a strong effect. As you know, there can be a huge discrepancy even in the same league in the quality of coaching. I would worry that we would be missing out on good players on poorly coached teams, though the benefits could easily outweigh that negative.

These stats have a place in hockey, I really like them. I'm still not sure exactly what kind of weight they should carry or how exactly they should be looked at with so many variables involved.

Scurr is offline  
Old
11-07-2012, 05:33 PM
  #10
Luck 6
\\_______
 
Luck 6's Avatar
 
Join Date: Oct 2008
Location: Vancouver
Posts: 7,376
vCash: 500
Quote:
Originally Posted by Hansen 36 View Post
Solid two-way d-man, good offensive skills and is much better defensively now than he was. I'd say his upside is a slightly better version of Alex Sulzer, and if he decides to come over to North America, he'll likely reach that potential.
I'd say his potential is higher than that. He does have top 4 potential, don't see how he doesn't considering hes playing top pairing minutes in the SEL at age 22.

Luck 6 is online now  
Old
11-07-2012, 06:25 PM
  #11
timw33
Registered User
 
Join Date: Nov 2007
Location: Victoria
Country: Canada
Posts: 13,886
vCash: 500
Quote:
Originally Posted by pitseleh View Post
Iain Fyffe's Pucks from the Past is interesting in that regard - he basically constructed a model using G-A-P-+/- accounting for things like league difficulty and team quality and went through and drafted two decades worth of players using a fake team (pretending that the St. Louis Blues moved to Saskatoon in 1983). Draft picks were determined, I believe, based on simulations of the season:

http://www.puckprospectus.com/news/?author=8

He did at least as well as the NHL teams in picking good players, and probably better. For example, from 1997-2000, he got Hossa, Campbell, Markov, Ribeiro, B. Richards, Gionta, Comrie, Zetterberg, and Lundqvist while only picking above #26 once. He also picked good role players like Andrew Ference, Jarret Stoll, and Nick Schultz.

If simple stats like points can tell you that much about the draft, tracking scoring chances, or Corsi at the very least, would probably tell you a lot more. You could avoid the 'broken leg' pheonomenon by eliminating players who clearly have warts (speed, size, smarts) that limit their potential to ever make it to the NHL.
It's kinda like how Nate Silver predicted the US election to a T while all the pundits just went with their "gut", which is pretty much what the scouts do.

"He's got a solid jaw, good head of hair, I have a good feeling about this kid"

timw33 is offline  
Old
11-07-2012, 06:58 PM
  #12
vanuck
Griffiths Way Goons
 
vanuck's Avatar
 
Join Date: Dec 2009
Country: Hong Kong
Posts: 10,139
vCash: 500
^ Just read up a bit on Nate Silver... interesting stuff.

Quote:
Originally Posted by Scurr View Post
It's a good thought but I'm not sure it would be worth the investment. The problem with drafting amateurs isn't figuring out which ones are making an impact, it's forecasting which ones can make an impact in the NHL.
Is it possible that with these stats, we can still find players who are making an 'impact' - or have good possession numbers - but haven't put up the numbers to reflect that for some reason? If, say, Mallet eventually turns out to be worth the 2nd round pick used on him, such a system might be able to tell that he just hadn't gotten the opportunity to shine with more ice time and a bigger role, when in reality he was a good player all along - with good underlying stats - despite playing on the 4th line for Rimouski as an enforcer.

Again this is based on 'good players drive possession', and the belief that the Canucks are looking for those players who drive play in the right direction.

This would all be with the purpose of finding hidden gems who haven't necessarily put up big point totals due to maybe luck, ice time or whatever, but would still fit the definition of a good 'possession player'.

vanuck is online now  
Old
11-08-2012, 12:20 AM
  #13
Scurr
Registered User
 
Scurr's Avatar
 
Join Date: Jun 2009
Location: Whalley
Country: Canada
Posts: 7,902
vCash: 500
Quote:
Originally Posted by vanuck View Post
This would all be with the purpose of finding hidden gems who haven't necessarily put up big point totals due to maybe luck, ice time or whatever, but would still fit the definition of a good 'possession player'.
I'm with you on that. I guess the first thing to do is study how possession numbers in junior hockey and other leagues translate into the NHL. It'd be interesting. I imagine a lot of junior players are able to dominate without much defensive ability or awareness because they can control the play with the puck on their stick.

I'm not against what you're saying, I like it, I'm just not sure about those stats. Zone starts, quality of competition, quality of teammate, quality of the system you're playing and coaching you're getting... a lot to digest.

Scurr is offline  
Old
11-08-2012, 05:03 AM
  #14
DJOpus
Registered User
 
DJOpus's Avatar
 
Join Date: Nov 2011
Posts: 4,777
vCash: 500
Quote:
Originally Posted by timw33 View Post
It's kinda like how Nate Silver predicted the US election to a T while all the pundits just went with their "gut", which is pretty much what the scouts do.

"He's got a solid jaw, good head of hair, I have a good feeling about this kid"
I hope less people bring up Nate Silver, his stats make it easy to win election bets.

In regards to stats, is there any place that track stats about things like winning puck battles to get pucks out of the zone and random stuff like that?

DJOpus is online now  
Old
11-08-2012, 10:19 AM
  #15
Luck 6
\\_______
 
Luck 6's Avatar
 
Join Date: Oct 2008
Location: Vancouver
Posts: 7,376
vCash: 500
Jensen pointless with a -1 in his last 4 games. Let's see if he can respond to this adversity after the break. To be fair though, his team was outscored 8-0 the last 2 games and he was an even player over that span.

Luck 6 is online now  
Old
11-08-2012, 02:07 PM
  #16
Hansen 36
BASED JIM BENNING
 
Hansen 36's Avatar
 
Join Date: Oct 2011
Location: Nanaimo, B.C.
Country: Canada
Posts: 6,282
vCash: 500
Kassian has been suspended 1 game for actions in the game on the 6th.

Hansen 36 is online now  
Old
11-08-2012, 09:36 PM
  #17
Cocoa Crisp
Registered User
 
Cocoa Crisp's Avatar
 
Join Date: Mar 2006
Location: NYC
Country: Hong Kong
Posts: 2,765
vCash: 500
Quote:
Originally Posted by vanuck View Post
Given what we know about advanced stats and how they've proven to be useful for quantifying NHL players' performance, does anyone else think the Canucks implement their own system for stat-tracking like the numbers we see on nhl.com, but at the junior and AHL levels?

You would think they'd have the financial resources to hire enough people to do this at each and every arena at the CHL and AHL levels - all kept hush-hush of course. Obviously they'd need a huge division of people just devoted to measuring statistics.

If it's true that good players drive possession, naturally they would be looking for the kids who have the best underlying numbers, and find value where other teams may not be as clued in about someone, while using the perception of value to trade down and get more picks.

But I was just wondering if this was at all feasible and if it's possible VAN is already doing this. And if so, don't we gain a massive edge in terms of amateur scouting?

Any flaws about such a hypothetical method of finding players?

Does this make sense? Am I going crazy here?
Maybe a simpler solution is to just do a better job with advanced stats and pro scouting. Just cutting down on the number of Keith Ballards and targeting players more likely to fit our system would seem a wiser use of resources.

Cocoa Crisp is offline  
Old
11-08-2012, 09:46 PM
  #18
Taelin
Resident Hipster
 
Taelin's Avatar
 
Join Date: Jan 2012
Location: Vancouver
Country: Canada
Posts: 7,240
vCash: 500
Quote:
Originally Posted by Hansen 36 View Post
Kassian has been suspended 1 game for actions in the game on the 6th.
He threw his stick, which landed on an official.

Taelin is offline  
Old
11-08-2012, 11:24 PM
  #19
vanuck
Griffiths Way Goons
 
vanuck's Avatar
 
Join Date: Dec 2009
Country: Hong Kong
Posts: 10,139
vCash: 500
Quote:
Originally Posted by Cocoa Crisp View Post
Maybe a simpler solution is to just do a better job with advanced stats and pro scouting. Just cutting down on the number of Keith Ballards and targeting players more likely to fit our system would seem a wiser use of resources.
I think the Canucks are already using advanced stats to evaluate their own and other teams' personnel. The system for measuring players' performance possession-wise is already in place though because they keep track of all sorts of stats, whereas it doesn't for junior hockey, NCAA and the AHL.

As for doing a better job with Ballard, well, that one should fall on Bowness. He's coached him before so you'd think he'd already know if he could play the right side or not, as well as his play style, strengths and weaknesses, and whether or not they would fit into the Nucks' system.

vanuck is online now  
Old
11-09-2012, 12:04 AM
  #20
vanuck
Griffiths Way Goons
 
vanuck's Avatar
 
Join Date: Dec 2009
Country: Hong Kong
Posts: 10,139
vCash: 500
Quote:
Originally Posted by Scurr View Post
One of the biggest problems I have with Corsi in junior (and it can apply in the NHL as well) is that coaching and systems have such a strong effect. As you know, there can be a huge discrepancy even in the same league in the quality of coaching. I would worry that we would be missing out on good players on poorly coached teams, though the benefits could easily outweigh that negative.
This is a good point. Really then, if this is a flaw with using Corsi for even NHLers, does it mean that its always going to be inherently flawed for all levels of hockey? Apparently all of Detroit's players have really high Corsi rates... does this really mean every single one of them is a dominant possession player? Or is that a product of Babcock's system that exaggerates everyone's underlying numbers?

I don't fully understand everything there is to know about advanced stats yet, but the hope is that even on a bad team (whether due to coaching or just lack of talent) the numbers would still allow for us to find those diamonds in the rough, who aren't producing as much as they could be due to the quality of their team, but are still good players in their own right.

But the problem seems to be when you're trying to compare players across different teams in this way, and I don't know that even advanced stats could be used for a comparison in that way in the NHL. I'd think traditional scouting is still needed, and a mix of both methods might be the perfect way to go about evaluating players.

Unfortunately however, some of us have little faith in our scouts when it comes to *certain regions*, so it might end up being a moot point anyway! But hopefully advanced metrics would negate some of that ineptitude we seem to suffer from.

Quote:
Originally Posted by Scurr View Post
I'm with you on that. I guess the first thing to do is study how possession numbers in junior hockey and other leagues translate into the NHL. It'd be interesting. I imagine a lot of junior players are able to dominate without much defensive ability or awareness because they can control the play with the puck on their stick.

I'm not against what you're saying, I like it, I'm just not sure about those stats. Zone starts, quality of competition, quality of teammate, quality of the system you're playing and coaching you're getting... a lot to digest.
Agreed. I wouldn't even know where to begin with things like QoC for 16/17 year old players if they've only played in the junior league for a season or so, though I suppose if we did use such a system MG would leave that up to a group of statistics professionals to figure it out. Didn't he say they were consulting stats people for something, back when he first got hired as GM?

Not to mention how much someone's advanced stats can change as they mature, so it seems to be a mystery as to how one can reasonably project a player's growth in future.

But I do wonder if even just a little statistical analysis would already be at least something of an improvement on our scouting as it is right now. So that we don't end up with players like S. Schneider eating up a contract spot for 3 years whenever we go to the WHL.

vanuck is online now  
Old
11-09-2012, 12:17 AM
  #21
Tiranis
Registered User
 
Join Date: Jun 2009
Location: Toronto, ON
Country: Czech_ Republic
Posts: 20,975
vCash: 500
Quote:
Originally Posted by vanuck View Post
This is a good point. Really then, if this is a flaw with using Corsi for even NHLers, does it mean that its always going to be inherently flawed for all levels of hockey? Apparently all of Detroit's players have really high Corsi rates... does this really mean every single one of them is a dominant possession player? Or is that a product of Babcock's system that exaggerates everyone's underlying numbers?
You can still look at Corsi Relative to compare them to their own teammates. While that doesn't give you a league-wide view of how good they are, it does give you a sense of where they stand in relation to their teammates. I think that's probably all you really need anyway.

I could see using advanced stats to uncover overlooked gems, guys that are buried on 3rd/4th lines but have potential for more. I don't think it would be that useful for 1st and 2nd liners due to the nature of the junior game.

Tiranis is offline  
Old
11-09-2012, 02:08 AM
  #22
vanuck
Griffiths Way Goons
 
vanuck's Avatar
 
Join Date: Dec 2009
Country: Hong Kong
Posts: 10,139
vCash: 500
Quote:
Originally Posted by Tiranis View Post
You can still look at Corsi Relative to compare them to their own teammates. While that doesn't give you a league-wide view of how good they are, it does give you a sense of where they stand in relation to their teammates. I think that's probably all you really need anyway.

I could see using advanced stats to uncover overlooked gems, guys that are buried on 3rd/4th lines but have potential for more. I don't think it would be that useful for 1st and 2nd liners due to the nature of the junior game.
So if such a hypothetical system exists and works, it should suggest that Mallet was much more than just a 4th liner for Rimouski before breaking out, correct? Even if he never got the ice time to produce much before.

I thought it might be a potentially useful tool for finding underrated players, but I think you also make a good point in using it first by looking at guys on the same team - provided there's a way to figure out the other metrics - i.e guys who stand out in terms of possession but don't have big numbers to show because of limited opportunity.

So then the problem of comparing players across different teams (and leagues) still remains, when it comes to compiling your overall rankings. I also wonder how one adjusts for league difficulties, team qualities, coaching/system factors. How do you even decide what values to use for these? Interesting stuff nonetheless.

Btw, how do you figure out a player's Quality of Teammates if, say, one teammate is a 16 year-old playing his first game in the WHL? I guess this also goes for NHL rookies too.

Everyone says drafting is such a crapshoot. If VAN can figure out a way to reduce that uncertainty even just a little bit with these metrics - by spending money that isn't even capped to begin with! - and remove some bias and subjectivity out of the equation, we might gain an advantage at the draft.


Last edited by vanuck: 11-09-2012 at 02:16 AM.
vanuck is online now  
Old
11-09-2012, 03:16 PM
  #23
mbarker
Registered User
 
Join Date: Nov 2009
Posts: 32
vCash: 500
http://www.subwaysuperseries.ca/sche...how/game/64149

Did Corrado not suit up?

mbarker is offline  
Old
11-09-2012, 03:21 PM
  #24
arsmaster
Registered User
 
Join Date: Mar 2009
Posts: 17,914
vCash: 500
Quote:
Originally Posted by mbarker View Post
Not all prospects dress for both games.

I believe he's on the roster for the game in Sarnia.

arsmaster is offline  
Old
11-09-2012, 04:09 PM
  #25
Tiranis
Registered User
 
Join Date: Jun 2009
Location: Toronto, ON
Country: Czech_ Republic
Posts: 20,975
vCash: 500
Quote:
Originally Posted by arsmaster View Post
Not all prospects dress for both games.

I believe he's on the roster for the game in Sarnia.
Correct. That's the way it was announced back then too:

Quote:
Corrado will play for Team OHL in Sarnia when the two-game Ontario series wraps up on Nov. 12. Leivo will play for Team OHL in Guelph on Nov. 8 and will also join Corrado in Sarnia on Nov. 12.
http://www.northernlife.ca/news/spor...s-sudbury.aspx

Tiranis is offline  
Closed Thread

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 08:18 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.