HFBoards

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

The Avs Advanced Stats Thread [Now with a Fresh, "New Title" look!]

Reply
 
Thread Tools
Old
10-02-2014, 05:46 PM
  #26
BrooksBets
Registered User
 
BrooksBets's Avatar
 
Join Date: Apr 2010
Posts: 1,269
vCash: 500
There are countless articles about the Avs inevitable "regression" this year, but this one by Travis Yost does the best job of breaking it down....

http://www.tsn.ca/yost-a-mountain-of...lanche-1.97188

Regardless of how it shakes out, I think it will be a successful year if they can improve their posession numbers and get another year of playoff experience.

BrooksBets is offline   Reply With Quote
Old
10-02-2014, 06:58 PM
  #27
Freudian
Clearly deranged
 
Freudian's Avatar
 
Join Date: Jul 2003
Country: Sweden
Posts: 40,753
vCash: 50
Quote:
Originally Posted by BrooksBets View Post
There are countless articles about the Avs inevitable "regression" this year, but this one by Travis Yost does the best job of breaking it down....

http://www.tsn.ca/yost-a-mountain-of...lanche-1.97188

Regardless of how it shakes out, I think it will be a successful year if they can improve their posession numbers and get another year of playoff experience.
What a sloppy article.

Using minuscule samples with big ranges and dividing those results by 10 and counting that average as a prediction of where Avs will end up when it comes to shooting and saves next year.

Shooting for these ten teams in year two varied from 7.26-9.76%. Save percentages varied from .913 to .931. Those are gigantic ranges so to "expect Colorado to shoot around 7.89 per cent for next year at evens and stop around 92.47 per cent of the shots." is a bizarre conclusion, unless you believe in regression to the mean being a real force inevitably pulling a team there.

The "Further, it's difficult to project improved possession numbers when the Avalanche brain trust doesn't seem inclined to dig into possession-based analytics." sentence made me laugh. Does he seriously think that just because teams count shot attempts they automatically improve their puck possession?

He's also wrong about Avs not "digging into possession-based analytics". Avs track zone entries, type of zone entries, if different type of zone entries result in shots or not, zone time and scoring chances. Those are much better tools for evaluating successful puck possession than corsi/fenwick are.

Everyone expect Avs to regress in some way this season. They had a 112 point season with a 95 point caliber team. Sakic and Roy expects Avs to finish with fewer points this coming season too. They've more or less said so publicly this off-season. So predicting Avs regressing using advanced stats doesn't mean more than predicting it via the eye test.

Freudian is offline   Reply With Quote
Old
10-03-2014, 08:53 AM
  #28
InjuredChoker
Registered User
 
InjuredChoker's Avatar
 
Join Date: Dec 2011
Location: LTIR or golf course
Posts: 30,321
vCash: 925
dellow had better article on his site.

InjuredChoker is offline   Reply With Quote
Old
10-03-2014, 09:38 AM
  #29
Freudian
Clearly deranged
 
Freudian's Avatar
 
Join Date: Jul 2003
Country: Sweden
Posts: 40,753
vCash: 50
If we look at the ES save percentages and shooting percentages for the league the last seven seasons we find that there isn't any true average that teams gravitate towards. Some teams are more successful at shooting the puck than other teams. Some teams are more successful at stopping the puck than other teams. This is the reason why the idea that every team will gravitate towards 1000 PDO is silly.



We don't know what the true mean is for this Avs team when it comes to shooting. We don't know what the true mean is for this Avs team when it comes to save percentage. We don't know where they are regressing to, if they are performing better than expected.

Of course Avs are bound to do worse in some areas this year. If we didn't believe this, we would believe they would get 112 points again and no one does. But we don't know if it's shooting or save percentages going down (it's likely) and if it does we don't know how much. We don't know if they will take more or less shots or allow more or less shots. We don't know if they'll be more successful on the PP/PK or not. We don't know if they'll be more or less successful in OT/SO or not.

As for Yost and his sloppy analysis. Even if the nine teams he picked were perfect predictors for Avs in 2014-15 (which they aren't since all the players and the systems are different) at least he could perform a basic statistical analysis.

Based on his sample and his assumption that those teams are perfect predictors for Avs shooting percentage he can predict with 95% confidence that Avs ES shooting next year will be between 7.47-8.43%. With 99% confidence the range is 7.32-8.58%.

Based on his sample and his assumption that those teams are perfect predictors for Avs save percentage he can predict with 95% confidence that Avs ES save percentage next year will be between .919 and .927. With 99% confidence the range is .918 to .928.

Quite different than the Nostradamusesque prediction that we can "expect Colorado to shoot around 7.89 per cent for next year at evens and stop around 92.47 per cent of the shots".

This is the most annoying aspect of the advanced stats community. They are overselling their results, drawing conclusions that aren't supported by the data and then other parts of the community are using those claims as proof of the validity of advanced stats.


Last edited by Freudian: 10-03-2014 at 09:52 AM.
Freudian is offline   Reply With Quote
Old
10-03-2014, 10:38 AM
  #30
The Mars Volchenkov
Registered User
 
Join Date: Mar 2002
Location: Colorado
Country: United States
Posts: 47,025
vCash: 127
Send a message via MSN to The Mars Volchenkov
Quote:
Originally Posted by InjuredChoker View Post
dellow had better article on his site.
I'm sure he thought that as well, and told everyone that it's the greatest.

The Mars Volchenkov is offline   Reply With Quote
Old
10-03-2014, 10:42 AM
  #31
InjuredChoker
Registered User
 
InjuredChoker's Avatar
 
Join Date: Dec 2011
Location: LTIR or golf course
Posts: 30,321
vCash: 925
Quote:
Originally Posted by The Mars Volchenkov View Post
I'm sure he thought that as well, and told everyone that it's the greatest.
haha, he's a [MOD] but smart guy.

not that i agreed with all of his conclusions, esp. at the time.


Last edited by Frenchy: 10-03-2014 at 05:44 PM.
InjuredChoker is offline   Reply With Quote
Old
10-03-2014, 11:01 AM
  #32
Flanagan
Registered Loser
 
Flanagan's Avatar
 
Join Date: Jul 2011
Location: St. John's
Country: Canada
Posts: 4,339
vCash: 653
Since this is a thread dedicated to advanced stats, can someone explain to me exactly how QoC is calculated? I've looked it up a few times and the explanations I've seen were pretty meaningless.

Flanagan is offline   Reply With Quote
Old
10-03-2014, 11:06 AM
  #33
henchman24
Mr. Meeseeks
 
henchman24's Avatar
 
Join Date: Feb 2012
Posts: 26,671
vCash: 500
Okay... I think people are missing the point of this thread for the Avs. This is for tracking various stats to see their impact, and possibly using different methods than just Corsi/Fenwick.

henchman24 is offline   Reply With Quote
Old
10-03-2014, 11:10 AM
  #34
InjuredChoker
Registered User
 
InjuredChoker's Avatar
 
Join Date: Dec 2011
Location: LTIR or golf course
Posts: 30,321
vCash: 925
Quote:
Originally Posted by Flanagan View Post
Since this is a thread dedicated to advanced stats, can someone explain to me exactly how QoC is calculated? I've looked it up a few times and the explanations I've seen were pretty meaningless.
there are different methods to calculate it.

+/-, corsi, ice time and so on. the more you play against better players, the higher the rating will be, and the sites will use different ways to record it.

i don't think we can accurately measure how hard it's to play against other teams 4th line or 1st line. meaning the current QoC metrics tell more about how coach uses the player than how good he is.

stars usually play against stars and goons against goons. so many teams and shifts so it evens out.

InjuredChoker is offline   Reply With Quote
Old
10-03-2014, 12:40 PM
  #35
Mystic MacK
Foppa2118
 
Mystic MacK's Avatar
 
Join Date: Oct 2003
Country: United States
Posts: 28,322
vCash: 500
Quote:
Originally Posted by henchman24 View Post
Okay... I think people are missing the point of this thread for the Avs. This is for tracking various stats to see their impact, and possibly using different methods than just Corsi/Fenwick.
When people see Avalanche advanced stats thread it sounds like corsi/fenwick.

Mystic MacK is offline   Reply With Quote
Old
10-03-2014, 12:49 PM
  #36
henchman24
Mr. Meeseeks
 
henchman24's Avatar
 
Join Date: Feb 2012
Posts: 26,671
vCash: 500
Quote:
Originally Posted by Foppa2118 View Post
When people see Avalanche advanced stats thread it sounds like corsi/fenwick.
Well there is the OP....

henchman24 is offline   Reply With Quote
Old
10-03-2014, 12:50 PM
  #37
Mystic MacK
Foppa2118
 
Mystic MacK's Avatar
 
Join Date: Oct 2003
Country: United States
Posts: 28,322
vCash: 500
Quote:
Originally Posted by henchman24 View Post
Well there is the OP....
Yea but I bet people rarely read the OP after a certain point. Especially if it's more than one or two sentences.

Mystic MacK is offline   Reply With Quote
Old
10-03-2014, 12:51 PM
  #38
henchman24
Mr. Meeseeks
 
henchman24's Avatar
 
Join Date: Feb 2012
Posts: 26,671
vCash: 500
Quote:
Originally Posted by Foppa2118 View Post
Yea but I bet people rarely read the OP after a certain point. Especially if it's more than one or two sentences.
But....


henchman24 is offline   Reply With Quote
Old
10-03-2014, 02:35 PM
  #39
Flanagan
Registered Loser
 
Flanagan's Avatar
 
Join Date: Jul 2011
Location: St. John's
Country: Canada
Posts: 4,339
vCash: 653
Quote:
Originally Posted by InjuredChoker View Post
there are different methods to calculate it.

+/-, corsi, ice time and so on. the more you play against better players, the higher the rating will be, and the sites will use different ways to record it.

i don't think we can accurately measure how hard it's to play against other teams 4th line or 1st line. meaning the current QoC metrics tell more about how coach uses the player than how good he is.

stars usually play against stars and goons against goons. so many teams and shifts so it evens out.
I know the basics of QoC and what it's used for. I'm interested in the exact formulas used to calculate it. Let's say using +/-.

According to this site, "It's measured by the player's relative plus/minus (plus/minus of when the player is on the ice / [plus/minus when the player is on the ice + plus/minus when the player is off it]) and it's weighted based on TOI versus the player."

Looking at the denominator of the equation first, it's "plus/minus when the player is on the ice + plus/minus when the player is off it". Is that just goal differential for the team? (excluding PP goals against).

Then if you actually consider the very nature of the +/- stat (can be positive or negative), it's obvious that the formula above makes no sense. If a player is +1 and his overall team is -1, then his relative +/- is -1. If he's a +3 and his team is +1 then he gets a +3. Both those cases have a player whose personal goal differential is +2 better than his team as a whole. Yet their relative +/- stats are incomparable. And what if the team has an even goal differential? The whole formula goes to crap unless you figure out how to divide by zero.

I've posted this question here a few times with no luck. I feel like I'm completely misunderstanding that equation, otherwise it appears to be pure garbage.


Last edited by Flanagan: 10-03-2014 at 02:43 PM.
Flanagan is offline   Reply With Quote
Old
10-03-2014, 02:45 PM
  #40
InjuredChoker
Registered User
 
InjuredChoker's Avatar
 
Join Date: Dec 2011
Location: LTIR or golf course
Posts: 30,321
vCash: 925
maybe this helps?

or this?

InjuredChoker is offline   Reply With Quote
Old
10-03-2014, 02:55 PM
  #41
Flanagan
Registered Loser
 
Flanagan's Avatar
 
Join Date: Jul 2011
Location: St. John's
Country: Canada
Posts: 4,339
vCash: 653
Quote:
Originally Posted by InjuredChoker View Post
I've seen that first article before. The link to relative +/- is broken. I guess I should rephrase and say that the stat I'm asking about is rel. plus minus, not QoC.

Flanagan is offline   Reply With Quote
Old
10-03-2014, 03:05 PM
  #42
InjuredChoker
Registered User
 
InjuredChoker's Avatar
 
Join Date: Dec 2011
Location: LTIR or golf course
Posts: 30,321
vCash: 925
Quote:
Originally Posted by Flanagan View Post
I've seen that first article before. The link to relative +/- is broken. I guess I should rephrase and say that the stat I'm asking about is rel. plus minus, not QoC.
http://www.arcticicehockey.com/2009/...d-question-1-a

InjuredChoker is offline   Reply With Quote
Old
10-03-2014, 05:47 PM
  #43
Flanagan
Registered Loser
 
Flanagan's Avatar
 
Join Date: Jul 2011
Location: St. John's
Country: Canada
Posts: 4,339
vCash: 653
Quote:
Originally Posted by InjuredChoker View Post
"We can make a small improvement on +/- by subtracting the +/- when a player is off the ice from it."

Ok, so that source I was originally quoting is indeed completely wrong. I knew dividing made little sense.

Thanks for clearing that up.

Flanagan is offline   Reply With Quote
Old
10-06-2014, 06:40 PM
  #44
The Mars Volchenkov
Registered User
 
Join Date: Mar 2002
Location: Colorado
Country: United States
Posts: 47,025
vCash: 127
Send a message via MSN to The Mars Volchenkov
NHL Network just had a review of the Avs and the analytics. The Avs were 5th in the league in offensive zone possession at 5 on 5, and were 12th overall in puck possession overall. They were also near the bottom of the league in offensive zone turnovers, despite spending more time in the offensive zone than most teams. Thought it was interesting. I hope NHL.com puts the video up later.

The Mars Volchenkov is offline   Reply With Quote
Old
10-06-2014, 07:56 PM
  #45
The Mars Volchenkov
Registered User
 
Join Date: Mar 2002
Location: Colorado
Country: United States
Posts: 47,025
vCash: 127
Send a message via MSN to The Mars Volchenkov
Here's the video with Avs possession stuff. Not sure how to embed it.

http://video.nhl.com/videocenter/con...d=nhl:topheads

The Mars Volchenkov is offline   Reply With Quote
Old
10-06-2014, 08:03 PM
  #46
StayAtHomeAv
Registered User
 
Join Date: May 2014
Posts: 5,554
vCash: 500
Quote:
Originally Posted by The Mars Volchenkov View Post
NHL Network just had a review of the Avs and the analytics. The Avs were 5th in the league in offensive zone possession at 5 on 5, and were 12th overall in puck possession overall. They were also near the bottom of the league in offensive zone turnovers, despite spending more time in the offensive zone than most teams. Thought it was interesting. I hope NHL.com puts the video up later.
But, but, but, corsi........

I'm just glad someone had the actual possession times. Corsi is such a joke. Why are these times not talked about more? This week is the first I have heard them used and it's about to be a new season.

Well, henchman, hopefully this is one thing you guys won't need to track of.

StayAtHomeAv is online now   Reply With Quote
Old
10-06-2014, 08:08 PM
  #47
Freudian
Clearly deranged
 
Freudian's Avatar
 
Join Date: Jul 2003
Country: Sweden
Posts: 40,753
vCash: 50

Freudian is offline   Reply With Quote
Old
10-06-2014, 08:18 PM
  #48
Avs_19
Peter the Great
 
Avs_19's Avatar
 
Join Date: Jun 2007
Country: Canada
Posts: 59,028
vCash: 500
Good stuff. I laughed at that Duchene quote. That was Sacco's gameplan for the longest time, cross the blueline and just fire.

Avs_19 is offline   Reply With Quote
Old
10-06-2014, 10:01 PM
  #49
BrickAHL
Terry Balboa
 
BrickAHL's Avatar
 
Join Date: Mar 2011
Location: Vancouver
Country: Canada
Posts: 13,592
vCash: 50
Amazing video. So we are a good possession team. I never understood the whole shots for thing counting as possession. I'm glad we actually have numbers that show we actually have the puck on our sticks longer than a lot of the league.

BrickAHL is offline   Reply With Quote
Old
10-06-2014, 11:50 PM
  #50
ProgOg
Registered User
 
Join Date: Aug 2014
Posts: 2,563
vCash: 500
They still allowed the 4th most shot attempts against , only Toronto, Buffalo and Edmonton allowed more.

Is anyone surprised that the skill is there? Problem isn't with the forwards.


Last edited by ProgOg: 10-06-2014 at 11:57 PM.
ProgOg 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 08:30 PM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"

vBulletin Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. @2017 All Rights Reserved.