HFBoards

Go Back   HFBoards > NHL Eastern Conference > Atlantic Division > Montreal Canadiens
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
Notices

Sabremetrics/Microstats in Hockey

Reply
 
Thread Tools
Old
06-05-2012, 02:14 PM
  #226
Fozz
Registered User
 
Fozz's Avatar
 
Join Date: Aug 2002
Posts: 6,390
vCash: 500
Quote:
Originally Posted by SaskRinkRat View Post
Are you against using information to inform decisions?
I'm sure they much prefer analyzing players by what they see, not what numbers tell them. I'm also certain that if some stats are indeed usefull, they are already very well known about and used accordingly.

Fozz is offline   Reply With Quote
Old
06-05-2012, 03:47 PM
  #227
SaskRinkRat
Registered User
 
Join Date: Apr 2010
Posts: 396
vCash: 500
Quote:
Originally Posted by Fozz View Post
I'm sure they much prefer analyzing players by what they see, not what numbers tell them. I'm also certain that if some stats are indeed usefull, they are already very well known about and used accordingly.
Not that I want to restart this debate, but this certainly wasn't the case in baseball prior to Oakland's experiments in the early 2000s. Why would it be the case in the NHL today?

SaskRinkRat is offline   Reply With Quote
Old
06-05-2012, 04:26 PM
  #228
Talks to Goalposts
Registered User
 
Join Date: Apr 2011
Location: Edmonton
Country: Canada
Posts: 3,610
vCash: 500
Quote:
Originally Posted by SaskRinkRat View Post
Not that I want to restart this debate, but this certainly wasn't the case in baseball prior to Oakland's experiments in the early 2000s. Why would it be the case in the NHL today?
Or to look at it another way, NHL executives generally don't have the kind of background that goes with being saavy to the value and use of basic applied mathematics.

Talks to Goalposts is offline   Reply With Quote
Old
06-06-2012, 05:44 AM
  #229
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 3,733
vCash: 500
Quote:
Originally Posted by SaskRinkRat View Post
Not that I want to restart this debate, but this certainly wasn't the case in baseball prior to Oakland's experiments in the early 2000s. Why would it be the case in the NHL today?
I tend to be on the side that Oaklnad's experiment has been completely over exaggerated. A couple good years followed by mostly sucking.

Whatever they have been doing by in large hasn't worked. I realize others caught on yada yada yada and started employing some of the same analysis, but I just don't see the ground breaking change of events that they created.

Billy Beane has been mostly garbage. Just my opinion.

habsfanatics is offline   Reply With Quote
Old
06-06-2012, 12:42 PM
  #230
SaskRinkRat
Registered User
 
Join Date: Apr 2010
Posts: 396
vCash: 500
Quote:
Originally Posted by habsfanatics View Post
I tend to be on the side that Oaklnad's experiment has been completely over exaggerated. A couple good years followed by mostly sucking.

Whatever they have been doing by in large hasn't worked. I realize others caught on yada yada yada and started employing some of the same analysis, but I just don't see the ground breaking change of events that they created.

Billy Beane has been mostly garbage. Just my opinion.
I think this depends how you define "hasn't worked". If you define it strictly by World Series wins and losses then sure, it hasn't worked. But if you factor in other important things (like how much of a payroll deficit the team continues to operate under) then it's hard to argue that their approach "hasn't worked".

The other thing this goes to show is just how much randomness there is in any sort of system. If you only define success as championships, and you shift your strategy every time you don't win one, you'll probably never win one. You could be employing the absolute perfect strategy and still lose on account of the randomness in the process.

It's why the best team in the NHL almost never wins the Stanley Cup.

SaskRinkRat is offline   Reply With Quote
Old
06-06-2012, 03:16 PM
  #231
Roke
Registered User
 
Roke's Avatar
 
Join Date: Jul 2003
Location: Winnipeg
Country: Canada
Posts: 1,889
vCash: 500
If you're looking at statistical analysis in baseball you have to look beyond what happened in 2002.

Look what Boston did (up until this season) to help close the gap on the Yankees, look at what Tampa Bay is doing to fill out their roster, look at what the Blue Jays are doing under Anthopoulos.

Roke is offline   Reply With Quote
Old
06-06-2012, 03:31 PM
  #232
Fozz
Registered User
 
Fozz's Avatar
 
Join Date: Aug 2002
Posts: 6,390
vCash: 500
Quote:
Originally Posted by Roke View Post
If you're looking at statistical analysis in baseball you have to look beyond what happened in 2002.

Look what Boston did (up until this season) to help close the gap on the Yankees, look at what Tampa Bay is doing to fill out their roster, look at what the Blue Jays are doing under Anthopoulos.
That's all nice and dandy but irrelevant in regards to hockey. Trying to build a statistical analysis of hockey has way too many holes to be counted upon to make decisions. It may be used to support some specific elements but those who think a 'moneyball' type model is remotely possible in hockey are just kidding themselves. Let's face it, baseball is an individual sport in over 95% of it's action, while hockey is the complete opposite.

Fozz is offline   Reply With Quote
Old
06-06-2012, 03:36 PM
  #233
Roke
Registered User
 
Roke's Avatar
 
Join Date: Jul 2003
Location: Winnipeg
Country: Canada
Posts: 1,889
vCash: 500
Quote:
Originally Posted by Fozz View Post
That's all nice and dandy but irrelevant in regards to hockey. Trying to build a statistical analysis of hockey has way too many holes to be counted upon to make decisions. It may be used to support some specific elements but those who think a 'moneyball' type model is remotely possible in hockey are just kidding themselves. Let's face it, baseball is an individual sport in over 95% of it's action, while hockey is the complete opposite.
Real gains have already been made in the statistical analysis of hockey. There's a lot of statistical analysis in soccer which involves a tonne of teamwork and collective action

I view the difficulty of getting good data as a good thing for an organization rather than a bad thing. It means that if you put in the time and effort you can come up with something ahead of what other organizations can do, whether it be the NBA route of collecting proprietary data or putting together your own systems.

The goal shouldn't to invent an all-encompassing system that will tell you a player's value down to a nickel (not even baseball has that yet). The should be to build tools to complement your scouting and give you greater insight to the contributions players and teams make on the ice.

Roke is offline   Reply With Quote
Old
06-06-2012, 07:06 PM
  #234
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 3,733
vCash: 500
Quote:
Originally Posted by SaskRinkRat View Post
I think this depends how you define "hasn't worked". If you define it strictly by World Series wins and losses then sure, it hasn't worked. But if you factor in other important things (like how much of a payroll deficit the team continues to operate under) then it's hard to argue that their approach "hasn't worked".

The other thing this goes to show is just how much randomness there is in any sort of system. If you only define success as championships, and you shift your strategy every time you don't win one, you'll probably never win one. You could be employing the absolute perfect strategy and still lose on account of the randomness in the process.

It's why the best team in the NHL almost never wins the Stanley Cup.
I'm not basing it solely on championships, the A's haven't been playing .500 ball in years. They stink.

Yeah they had a reign at the top or were one of the best teams for a 4-5 year stretch, but many teams go through these stretches.

Obviously it's a valuable tool, I suggest every team employ some sort of statistician ect, but it will never trump pro scouting, the two together I have no problem. Anything to get an edge, but I find the usefulness vastly overstated around here at times.

Some talk as if reading a scorecard offers more than visible subjective opinions from long standing pros and I simply disagree. If all the top point scorers or possession players in the minors were drafted based solely on this there would be some very poor drafting records.

habsfanatics is offline   Reply With Quote
Old
06-06-2012, 07:08 PM
  #235
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 3,733
vCash: 500
Quote:
Originally Posted by Roke View Post
If you're looking at statistical analysis in baseball you have to look beyond what happened in 2002.

Look what Boston did (up until this season) to help close the gap on the Yankees, look at what Tampa Bay is doing to fill out their roster, look at what the Blue Jays are doing under Anthopoulos.
What is anthopoulos doing? He is rebuilding. I don't see much moneyball going on here.

habsfanatics is offline   Reply With Quote
Old
06-06-2012, 07:26 PM
  #236
Roke
Registered User
 
Roke's Avatar
 
Join Date: Jul 2003
Location: Winnipeg
Country: Canada
Posts: 1,889
vCash: 500
Quote:
Originally Posted by habsfanatics View Post
What is anthopoulos doing? He is rebuilding. I don't see much moneyball going on here.
The Blue Jays run a tight ship so we don't know much. But, exploiting the old Type A/Type B free agency compensation to garner extra draft picks (Miguel Olivo in particular) is something minor though he was merely taking that to the next level rather than innovating. Not worrying about players' reputations and going out and trading them for their ability (Escobar, Rasmus, Lawrie) is something the Habs need to learn. In terms of data collection (most relevant to hockey because most of the NHL's data is crap) the Blue Jays have volunteers charting amateur pitchers' pitching performance.


It's not as if the Habs would be the first team in hockey to do this. San Jose's assistant GM had a couple of interviews where he mentioned they use statistical analysis, Vancouver does it, Stan Bowman in Chicago has talked about them using some sort of rating system, Chiarelli mentioned that Boston uses it (Burke in the same panel laughed off this stuff mind you). Dave Tippett had an interview where he talked about his time in the AHL where they had a "defensive defenseman" and an "offensive defenseman" where the offensive guy had a reputation for being poor defensively but they collected data, developed some sort of system and the numbers showed the offensive dman was better defensively because he spent a lot less time in the defensive zone. The work's being done.


Last edited by Roke: 06-06-2012 at 07:35 PM.
Roke is offline   Reply With Quote
Old
06-06-2012, 07:35 PM
  #237
Player 61
#Winning
 
Player 61's Avatar
 
Join Date: Aug 2007
Location: West Island
Country: Canada
Posts: 6,162
vCash: 500
Send a message via AIM to Player 61
Quote:
Originally Posted by habsfanatics View Post
What is anthopoulos doing? He is rebuilding. I don't see much moneyball going on here.
Toronto isn't exactly in a Moneyball situation, Unlike the Oakland's & the Tampa Bay's. Toronto's grown as large as Chicago! Besides it's owned by Rogers it's own Network, they almost don't mind losing some money for TV filler, where they just make it back.

You can see some smart acquisitions though in Rasmus & Santos. I hope Alex gets his 5 years, I think he's on the right track.

Player 61 is offline   Reply With Quote
Old
06-06-2012, 07:39 PM
  #238
Roke
Registered User
 
Roke's Avatar
 
Join Date: Jul 2003
Location: Winnipeg
Country: Canada
Posts: 1,889
vCash: 500
Quote:
Originally Posted by MaxPac67 View Post
Toronto isn't exactly in a Moneyball situation, Unlike the Oakland's & the Tampa Bay's. Toronto's grown as large as Chicago! Besides it's owned by Rogers it's own Network, they almost don't mind losing some money for TV filler, where they just make it back.

You can see some smart acquisitions though in Rasmus & Santos. I hope Alex gets his 5 years, I think he's on the right track.
I don't view "Moneyball" as a small-market team winning or doing weird stuff. That's how Michael Lewis portrayed Oakland in his book but the Bill James' chapters are more informative of the overall philosophy of asking questions, forming hypotheses, collecting the data to test those hypotheses, and so on.

It's more about innovation and attempted innovation than it is the size of the market.

Roke is offline   Reply With Quote
Old
06-06-2012, 07:41 PM
  #239
Player 61
#Winning
 
Player 61's Avatar
 
Join Date: Aug 2007
Location: West Island
Country: Canada
Posts: 6,162
vCash: 500
Send a message via AIM to Player 61
Quote:
Originally Posted by Roke View Post
I don't view "Moneyball" as a small-market team winning or doing weird stuff. That's how Michael Lewis portrayed Oakland in his book but the Bill James' chapters are more informative of the overall philosophy of asking questions, forming hypotheses, collecting the data to test those hypotheses, and so on.

It's more about innovation and attempted innovation than it is the size of the market.
I can assure you, the Jays are innovative. Might even be something new like Jaysball. ;-)

Player 61 is offline   Reply With Quote
Old
06-07-2012, 04:59 PM
  #240
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 3,733
vCash: 500
Quote:
Originally Posted by MaxPac67 View Post
Toronto isn't exactly in a Moneyball situation, Unlike the Oakland's & the Tampa Bay's. Toronto's grown as large as Chicago! Besides it's owned by Rogers it's own Network, they almost don't mind losing some money for TV filler, where they just make it back.

You can see some smart acquisitions though in Rasmus & Santos. I hope Alex gets his 5 years, I think he's on the right track.
Most definitely. I think he's the best executive in the league, but his approach hasn't been Billy Beane's approach, not even close.

I think the jury is still out on both Santos and Rasmus, however, I loved both moves. I loved Morrow for League. His best move yet imo.

habsfanatics is offline   Reply With Quote
Old
06-07-2012, 05:10 PM
  #241
Talks to Goalposts
Registered User
 
Join Date: Apr 2011
Location: Edmonton
Country: Canada
Posts: 3,610
vCash: 500
Quote:
Originally Posted by habsfanatics View Post
Most definitely. I think he's the best executive in the league, but his approach hasn't been Billy Beane's approach, not even close.
Not the same in methodology but very close in spirit. Its all about looking for inefficiencies in the market to exploit and innovating better methods to evaluate players. Evidence based reasoning is the major and essential component of both. The big reason innovation is necessary is because everyone else already knows about the old ways of doing things and even if they work there's little competitive advantage in them.

You can't pull a Moneyball with SABR stats anymore because everyone knows about them and the players they like get valued accordingly.

Stat work in Hockey isn't widespread so there is likely still an advantage to be exploited by using them.

Talks to Goalposts is offline   Reply With Quote
Old
06-07-2012, 06:06 PM
  #242
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 3,733
vCash: 500
Quote:
Originally Posted by Talks to Goalposts View Post
Not the same in methodology but very close in spirit. Its all about looking for inefficiencies in the market to exploit and innovating better methods to evaluate players. Evidence based reasoning is the major and essential component of both. The big reason innovation is necessary is because everyone else already knows about the old ways of doing things and even if they work there's little competitive advantage in them.

You can't pull a Moneyball with SABR stats anymore because everyone knows about them and the players they like get valued accordingly.

Stat work in Hockey isn't widespread so there is likely still an advantage to be exploited by using them.
I agree that they should be implemented. How much of an advantage they may give someone is a guess. I'm not particularly convinced it would help all that much, but I'm not against exploring it.

habsfanatics is offline   Reply With Quote
Old
06-07-2012, 06:16 PM
  #243
Talks to Goalposts
Registered User
 
Join Date: Apr 2011
Location: Edmonton
Country: Canada
Posts: 3,610
vCash: 500
Quote:
Originally Posted by habsfanatics View Post
I agree that they should be implemented. How much of an advantage they may give someone is a guess. I'm not particularly convinced it would help all that much, but I'm not against exploring it.
If anything they are pretty good at catching obvious mistakes that look decent at the time. Largely due to the obvious flaws that standard stats like goals, points and plus minus have for evaluating players.

However, I think the biggest opportunity for improvement is if media types knew more about them to prevent some of the many obviously wrong and false things they come up with. But that might be wishful thinking.

Talks to Goalposts is offline   Reply With Quote
Old
06-26-2012, 12:10 PM
  #244
Roke
Registered User
 
Roke's Avatar
 
Join Date: Jul 2003
Location: Winnipeg
Country: Canada
Posts: 1,889
vCash: 500
Read a really interesting article at lunch so I thought I'd drag this topic back up. It's about the correlation between scoring chances and (unblocked) shot-attempts)... basically trying to find effects of shot quality. A really good read:

http://nhlnumbers.com/2012/6/26/shot...nd-shot-totals

Basically, with the data available generating large numbers of quality chances goes hand-in-hand with generating large volumes of shots. teams that out-chance their opponents by scoring chances do so by out-shooting them.

Roke is offline   Reply With Quote
Old
06-26-2012, 02:10 PM
  #245
Miller Time
Registered User
 
Miller Time's Avatar
 
Join Date: Sep 2004
Posts: 8,184
vCash: 500
Quote:
Originally Posted by Talks to Goalposts View Post
If anything they are pretty good at catching obvious mistakes that look decent at the time. Largely due to the obvious flaws that standard stats like goals, points and plus minus have for evaluating players.

However, I think the biggest opportunity for improvement is if media types knew more about them to prevent some of the many obviously wrong and false things they come up with. But that might be wishful thinking.
it's not in their interest to do so...

media pundits aren't first and foremost concerned with accuracy, they need to tell a story. more important to tell a convincing emotional story, than to be accurate.

That's why Cherry is who he is.

there was a great clip the other day from ESPN where Mark Cuban basically made Skip Bayless look like a fool... and yet it didn't bother Skip one bit (nor get him to concede an inch).

Pro sports is an entertainment business first and foremost to pretty much anyone not tied directly to the day-to-day business of playing/competing (including training staff).

likewise, the average fan is more swayed by the emotional/entertaining angle, than he/she is by the more accurate one (and really, even the most advanced statistical breakdowns leave subjective gaps that can be interpreted different ways... though at least opinions formed from that basis of knowledge are generally more accurate than the "he's soft because he's a euro" or "they won because they wanted it more" basis.

Miller Time 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 11:30 AM.

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.