HFBoards

Go Back   HFBoards > General Hockey Discussion > By The Numbers
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
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.

Oilers, Leafs, Flames near the bottom, Sens near the top - Analytics report

Reply
 
Thread Tools
Old
08-07-2013, 09:39 AM
  #101
Master_Of_Districts
Registered User
 
Master_Of_Districts's Avatar
 
Join Date: Apr 2007
Location: Black Ruthenia
Country: Belarus
Posts: 1,745
vCash: 500
Quote:
Originally Posted by Trainspotter View Post
So, basically, despite copious hand waving your answer is "no". You can't really argue:

A) that an r^2 value of .3 shows a strong relationship between two variables.
B) Fenwick shows a strong relationship to regulation points in a single season.
C) Fenwick is a good way to predict future regulation points either within a season or season-to-season.

0.5 is not a strong relationship. You may find that a stat allowing one to predict less than half of future results is "relatively impressive", but I'm not so easily sold.
It is relatively impressive - when appropriately contextualized. 47% of the variance in team results over a half season sample is explained by randomness. So there's a fairly sharp theoretical limit on how well any particular metric can predict results from one half of the season to the other.

It's not exceptionally impressive. More nuanced models - like the one MacDonald created - do a better job. Which is what we would expect. If I were predicting future results, I would use a model like MacDonald's. Or a model based on Bayesian inference.

But it is relatively impressive. Most people regard a team's record as a reasonably good proxy for its underlying ability. So for a metric to predict future results better than actual record - especially a very simple one like Fenwick - is relatively impressive.


Last edited by Trebek: 08-07-2013 at 10:33 AM. Reason: Flaming
Master_Of_Districts is offline   Reply With Quote
Old
08-07-2013, 09:44 AM
  #102
Master_Of_Districts
Registered User
 
Master_Of_Districts's Avatar
 
Join Date: Apr 2007
Location: Black Ruthenia
Country: Belarus
Posts: 1,745
vCash: 500
Quote:
Originally Posted by Trainspotter View Post
The problem is that if the Leafs aren't in the playoffs confirmation bias will be rampant while the predictive validity of Fenwick remains weak. In fact, at least an equal proportion of any team's success or lack thereof will be accounted for by some other factor or combination of factors and The Whims of the Man in the Moon...ahem..."random variance" or whatever Bafflegab translation for "we don't have a clue how that happens" is fashionable at that point. But they'll still say, "we told you so".
For clarity, the random variation component of team results is literally that - random variation. It's not a label for the proportion of variation that can't be explained.

Master_Of_Districts is offline   Reply With Quote
Old
08-07-2013, 12:58 PM
  #103
AuraSphere
Registered User
 
AuraSphere's Avatar
 
Join Date: Jun 2012
Location: Toronto, Canada
Country: Canada
Posts: 1,288
vCash: 500
I'd be tremendously happy if the leafs tank the 2014-2015 season and come last place.

AuraSphere is online now   Reply With Quote
Old
08-07-2013, 06:47 PM
  #104
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 4,623
vCash: 500
I agree 100% with the analysis, I expect those 4 teams to be around lottery positions, there will be a few other teams in the running, but neither of the oilers, flames, leafs or sabres will make the playoffs this year IMO.

habsfanatics is offline   Reply With Quote
Old
08-07-2013, 07:05 PM
  #105
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 4,623
vCash: 500
Quote:
Originally Posted by BraveCanadian View Post
Bingo.

Even if the personnel didn't change, they have all aged a year and practiced another year etc.
It seems most of hard time understanding probability, what is meant by these articles is that if the play continues the way it has for the past 48 games then results will most likely be this. There's always a possibility that leafs, flames, oilers, sabres improve their possession games, but if they don't, this is the closest educated guess you can make.

If everything stays the same, or even close to the same as far as possession, he'll be close to bang on IMO.

If the newly acquired players do nothing to improve the teams underlying metrics, ie increase possession, then they'll have little or nothing to do with the prediction.

habsfanatics is offline   Reply With Quote
Old
08-07-2013, 08:17 PM
  #106
Dr Quincy
Registered User
 
Dr Quincy's Avatar
 
Join Date: Jun 2005
Country: United States
Posts: 16,864
vCash: 500
Quote:
Originally Posted by habsfanatics View Post
It seems most of hard time understanding probability, what is meant by these articles is that if the play continues the way it has for the past 48 games then results will most likely be this. There's always a possibility that leafs, flames, oilers, sabres improve their possession games, but if they don't, this is the closest educated guess you can make.

If everything stays the same, or even close to the same as far as possession, he'll be close to bang on IMO.

If the newly acquired players do nothing to improve the teams underlying metrics, ie increase possession, then they'll have little or nothing to do with the prediction.
The problem is, the prediction would have said the same thing after 20 games last year, and 30 and 40, and if after another 30 games this year, if those teams still are being "lucky" and winning more than this prediction expects, the prediction will expect them to start losing. When the prediction is wrong it's just "Oh that's random variance or luck, and the team will regress to the mean" and that can be said at any point in time. Yes as soon as one of those teams has a bad record I'm sure some proponents will say, "See, the model is right!"

It would be great if every time I looked like I was wrong I could just say, "Yeah but eventually I'll be right." And then at any moment I look like I'm right I can just say, "Yup, I'm right."

I'm not saying all or most people espousing advanced stats do this (and I'm not even against the advanced stats) but it's something I have seen by some people who have become slaves to certain statistics.

Dr Quincy is online now   Reply With Quote
Old
08-07-2013, 09:52 PM
  #107
Samsquanch
Creep
 
Samsquanch's Avatar
 
Join Date: Nov 2008
Location: Sudbury
Country: Canada
Posts: 2,283
vCash: 500
Quote:
Originally Posted by AuraSphere View Post
I'd be tremendously happy if the leafs tank the 2014-2015 season and come last place.
You do realize that this article is about the upcoming 2013-2014 season, right? Or myabe you do, and just like stating random facts, as Im sure all teams would love to get McDavid.

Samsquanch is online now   Reply With Quote
Old
08-07-2013, 11:07 PM
  #108
TheDevilMadeMe
Global Moderator
 
TheDevilMadeMe's Avatar
 
Join Date: Aug 2006
Location: Brooklyn
Country: United States
Posts: 42,411
vCash: 500
Quote:
Originally Posted by habsfanatics View Post
It seems most of hard time understanding probability, what is meant by these articles is that if the play continues the way it has for the past 48 games then results will most likely be this. There's always a possibility that leafs, flames, oilers, sabres improve their possession games, but if they don't, this is the closest educated guess you can make.

If everything stays the same, or even close to the same as far as possession, he'll be close to bang on IMO.

If the newly acquired players do nothing to improve the teams underlying metrics, ie increase possession, then they'll have little or nothing to do with the prediction.
Well, the Leafs just got David Clarkson, one of the best players in the league at getting weak shots on goal, so that should improve their Corsi/Fenwick metrics quite a bit.

TheDevilMadeMe is online now   Reply With Quote
Old
08-08-2013, 05:11 AM
  #109
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 4,623
vCash: 500
Quote:
Originally Posted by TheDevilMadeMe View Post
Well, the Leafs just got David Clarkson, one of the best players in the league at getting weak shots on goal, so that should improve their Corsi/Fenwick metrics quite a bit.
Entirely subjective that David clarkson generates substantially more weak shots than any1 else. He does generate shots, but nowhere close to enough to make them go from atrocious to good.

habsfanatics is offline   Reply With Quote
Old
08-08-2013, 10:38 PM
  #110
BraveCanadian
Registered User
 
BraveCanadian's Avatar
 
Join Date: Jun 2010
Country: Canada
Posts: 9,264
vCash: 500
Quote:
Originally Posted by habsfanatics View Post
If the newly acquired players do nothing to improve the teams underlying metrics, ie increase possession, then they'll have little or nothing to do with the prediction.

So if we totally ignore my point that it is in fact a different team playing this year, and then assume everything stays the same, then the prediction will be more likely to be true?

I'm glad we have you around to tell us about these things.


Last edited by Trebek: 08-09-2013 at 10:58 AM. Reason: Flaming.
BraveCanadian is online now   Reply With Quote
Old
08-08-2013, 11:10 PM
  #111
The Korean*
 
Join Date: Jul 2007
Location: New York
Country: South Korea
Posts: 7,635
vCash: 500
Send a message via AIM to The Korean* Send a message via Yahoo to The Korean*
Quote:
Originally Posted by Beastly Bozak View Post
Yeah most analysts nailed the leafs being a lottery team last year, so I guess we might as well not even try this season
Half a season screwed it up.

The Korean* is offline   Reply With Quote
Old
08-09-2013, 02:04 AM
  #112
Kershaw
 
Kershaw's Avatar
 
Join Date: Jun 2011
Country:
Posts: 25,519
vCash: 500
If the Sens continue at a .935 sv% pace, which I believe is attainable due to their ability to control possession, I suspect they'll challenge for the President's Trophy.

Kershaw is offline   Reply With Quote
Old
08-09-2013, 05:11 PM
  #113
habsfanatics
Registered User
 
Join Date: May 2012
Posts: 4,623
vCash: 500
Quote:
Originally Posted by BraveCanadian View Post
So if we totally ignore my point that it is in fact a different team playing this year, and then assume everything stays the same, then the prediction will be more likely to be true?

I'm glad we have you around to tell us about these things.
The majority of the core players are the same. Not much should be expected to change. Had they continued their play last year for a full 82 games they most likely would have missed the playoffs again.

habsfanatics is offline   Reply With Quote
Old
08-09-2013, 05:24 PM
  #114
Inverted
Registered User
 
Inverted's Avatar
 
Join Date: Oct 2010
Location: Melbourne, Australia
Country: Australia
Posts: 2,728
vCash: 500
These predictions are meaningless. A lot happens through out the season; injuries, players with hot/cold streaks, trades whether its a shorten one or a full 82 games. Last year the difference between 5th in the east and 9th was only 6 points. Likewise in the west the difference between 5 and 9th was merely 4 points. The only teams you can bet as locks are probably the Penguins, Bruins, Blackhawks and probably the Canucks. Besides now with the realignment the standings would be a lot different to previous seasons anyway.


Last edited by Inverted: 08-09-2013 at 05:30 PM.
Inverted is offline   Reply With Quote
Old
08-09-2013, 05:59 PM
  #115
BLONG7
Registered User
 
BLONG7's Avatar
 
Join Date: Oct 2002
Location: Nova Scotia
Posts: 13,374
vCash: 500
Quote:
Originally Posted by Kershaw View Post
If the Sens continue at a .935 sv% pace, which I believe is attainable due to their ability to control possession, I suspect they'll challenge for the President's Trophy.
I suspect you may well be disappointed...

BLONG7 is offline   Reply With Quote
Old
08-09-2013, 06:20 PM
  #116
sharks9
Registered User
 
sharks9's Avatar
 
Join Date: Jan 2012
Location: Canada
Posts: 10,411
vCash: 500
Quote:
Originally Posted by Kershaw View Post
If the Sens continue at a .935 sv% pace, which I believe is attainable due to their ability to control possession, I suspect they'll challenge for the President's Trophy.
There is no chance they continue at a .935 SV% pace

sharks9 is offline   Reply With Quote
Old
08-09-2013, 08:47 PM
  #117
Volcanologist
Used Register
 
Volcanologist's Avatar
 
Join Date: Jan 2003
Location: Cosmodrome
Country: Germany
Posts: 20,864
vCash: 500
Quote:
Originally Posted by habsfanatics View Post
The majority of the core players are the same. Not much should be expected to change. Had they continued their play last year for a full 82 games they most likely would have missed the playoffs again.
Based on what?

Volcanologist is online now   Reply With Quote
Old
08-09-2013, 09:52 PM
  #118
Samsquanch
Creep
 
Samsquanch's Avatar
 
Join Date: Nov 2008
Location: Sudbury
Country: Canada
Posts: 2,283
vCash: 500
Quote:
Originally Posted by Kershaw View Post
If the Sens continue at a .935 sv% pace, which I believe is attainable due to their ability to control possession, I suspect they'll challenge for the President's Trophy.
Its encouraging to hear that kind of praise coming from other fan bases

While I dont feel this is a presidents trophy caliber team yet, there is very good reason for optimism among Sens fans. An incredible amount of things went wrong for them last year, but the upside to that was that there was a chance for rookies to solidify themselves as NHLers, and for the most part there was tremendous growth across the board from just about everyone. Not to mention we added the top line winger we've been on for years this offseason.

((Growth) + (health) + (good offseason additions))*(better luck) = an excited Sens fan base.

Samsquanch is online now   Reply With Quote
Old
08-10-2013, 04:40 PM
  #119
Kershaw
 
Kershaw's Avatar
 
Join Date: Jun 2011
Country:
Posts: 25,519
vCash: 500
Quote:
Originally Posted by sharks9 View Post
There is no chance they continue at a .935 SV% pace
Probably not, but I think Anderson is a borderline elite tender. The Sens SA/G will drastically go down if they aren't decimated with injuries like they were last season.

Kershaw is offline   Reply With Quote
Old
08-23-2013, 01:53 AM
  #120
DayTripper
Registered User
 
DayTripper's Avatar
 
Join Date: Jul 2013
Location: Earth
Country: North Korea
Posts: 601
vCash: 500
Quote:
Originally Posted by swiftwin View Post
lol, analytics is not the same as analysts.
Hmmm, indeed...I just checked, and the advanced stats guys predicted the Leafs finishing 11th in the league last season, not anywhere close to 30th/lottery territory. But apparently they were wrong anyway, since the Leafs only got there by luck, apparently, and would have been a lottery team if not for that. :p


Last edited by DayTripper: 08-23-2013 at 01:59 AM.
DayTripper 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 08:31 PM.

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

vBulletin Copyright ©2000 - 2015, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. 2015 All Rights Reserved.