HFBoards

Go Back   HFBoards > NHL Eastern Conference > Metropolitan Division > New York Rangers
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
Notices

The Advanced Stats Thread Episode IV: A New Hope For Advanced Stats

Reply
 
Thread Tools
Old
12-09-2016, 09:03 PM
  #26
eperry
Registered User
 
Join Date: Jun 2016
Posts: 49
vCash: 500
Quote:
Originally Posted by Greg02 View Post
Gave a real read of the PSO article. Don't have hard evidence, but based on what was shared it's worthless.

Disclaimer: ML isn't really my area of CS, and stats really isn't my area of math.

That said, one of the reasons I dislike the ML field is because of how much buzzwordy it is. People go "oh, ML is the cool new thing, so let's just blindly use some standard library to solve all our problems" and then go and come up with things like this. It's not that I don't think ML can be applied in effective ways to hockey, because I do. It's just that it requires some careful thought.

For instance, if you're throwing in four hundred dimensions into your model, it is almost certain that some of them are going to be correlated (and in fact, I guarantee you that a number of these dimensions are pairwise correlated in extremely trivial ways). Now you're solving an optimization problem where the solution space isn't unique. So given different initial locations and velocities, you'll likely get different weights. Notice that calculating the weights was only ran once. But because the dimensions are linearly dependent, there are multiple "correct" convergences. There are a few interesting pieces of data that aren't included in the blog post; one might be to show the actual weights of each dimension. Then run it multiple times and compare. The multi-colinearity can be addressed by PCA or something, but honestly, better off with aggressive feature selection. So many features just seems rife for overfitting to me.

The model is then evaluated on its training data, which is a worthless judgment. Obviously it's going to converge so that the training set is fulfilled; it took those exact pairwise comparisons as the reason to weigh the different features in the first place! It's circular. The test dataset has to be different from the training dataset.
Glad you liked it.

eperry is offline   Reply With Quote
Old
12-10-2016, 06:40 AM
  #27
Greg02
Registered User
 
Greg02's Avatar
 
Join Date: Jun 2009
Posts: 2,123
vCash: 500
Quote:
Originally Posted by eperry View Post
Glad you liked it.
I'm sorry. That was obviously inappropriately framed for direct feedback and was unfair for me to project my frustrations with the ML "industry" as a whole on you.

Greg02 is online now   Reply With Quote
Old
12-10-2016, 12:12 PM
  #28
chosen
Registered User
 
chosen's Avatar
 
Join Date: Aug 2005
Posts: 6,735
vCash: 500
Quote:
Originally Posted by Mac n Gs View Post
I wonder if Manny isn't getting hired because of his outspoken views regarding other topics, which quite frankly, is a load of a horse crap if that's what's holding him back from getting an NHL job. Guy's a really, really good mathematician/statistician.
What outspoken views?

chosen is offline   Reply With Quote
Old
12-10-2016, 01:18 PM
  #29
Raspewtin
Free Kat Dennings
 
Raspewtin's Avatar
 
Join Date: May 2013
Country: France
Posts: 30,917
vCash: 50
Quote:
Originally Posted by chosen View Post
What outspoken views?
On twitter (which is his main platform) he speaks on politics a lot.

__________________
"It's going to be a zoo no matter what, Staalmania, Staals everywhre, omg Staal on Staal action, three brothers hockeying! Staal returning to Carolina omg what news! All the feels he is feel..." - Ail on Eric Staal's return to Carolina.
Raspewtin is offline   Reply With Quote
Old
12-10-2016, 03:22 PM
  #30
PlamsUnlimited
PUMP IT UP PUEMPEL
 
PlamsUnlimited's Avatar
 
Join Date: May 2010
Location: Lancaster, NY
Country: United States
Posts: 23,211
vCash: 919
I'm probably stupid but I was thinking about something. Is there a way to track shots in certain areas? By this I mean tracking "danger level". Like for example, someone shoots a low danger shot but inside of a high danger zone. If anyone gets what I mean. Or ripping a howitzer shot from a low danger area.

PlamsUnlimited is offline   Reply With Quote
Old
12-10-2016, 07:10 PM
  #31
Machinehead
Our Bayley is better
 
Machinehead's Avatar
 
Join Date: Jan 2011
Location: NYC
Country: United States
Posts: 63,632
vCash: 50
Quote:
Originally Posted by Raspewtin View Post
On twitter (which is his main platform) he speaks on politics a lot.
Tim Thomas got a job in the NHL

__________________
Machinehead is offline   Reply With Quote
Old
12-10-2016, 07:32 PM
  #32
eyjee
the basement
 
eyjee's Avatar
 
Join Date: Jun 2008
Location: Alphabetrium
Country: United States
Posts: 27,963
vCash: 500
Quote:
Originally Posted by Machinehead View Post
Tim Thomas got a job in the NHL
It's wrong, but I've heard from multiple people in the show that you can't be a dick on Twitter. Not saying anyone is or isn't a dick on Twitter, but you can't be one. They said that all the guys who you think don't have Twitter accounts have them, and they are looking. Those guys being GMs, other FO people.

I went back a full calendar year and had to delete like 200 tweets criticizing the Rangers

EDIT - It's wrong meaning how much weight that seemingly goes into being a "good person" on Twitter in terms of if you want an NHL job.

eyjee is offline   Reply With Quote
Old
12-10-2016, 07:43 PM
  #33
Mac n Gs
Gorton plz
 
Mac n Gs's Avatar
 
Join Date: Jan 2014
Country: United States
Posts: 10,160
vCash: 500
Pretty much, which again sucks, because Manny's a really smart guy and provides a lot of useful information.

Now to rear us back on topic, Cody Ceci's being mentioned again. Why? He's bad both via the eye test and stats.

Mac n Gs is offline   Reply With Quote
Old
12-10-2016, 07:46 PM
  #34
eyjee
the basement
 
eyjee's Avatar
 
Join Date: Jun 2008
Location: Alphabetrium
Country: United States
Posts: 27,963
vCash: 500
Quote:
Originally Posted by Mac n Gs View Post
Pretty much, which again sucks, because Manny's a really smart guy and provides a lot of useful information.
I believe quite firmly that there are probably over 50 hockey people I follow on Twitter that I'd take over anyone in the Rangers FO right now.

Quote:
Originally Posted by Mac n Gs View Post
Now to rear us back on topic, Cody Ceci's being mentioned again. Why? He's bad both via the eye test and stats.
See what I mean?

eyjee is offline   Reply With Quote
Old
12-10-2016, 08:29 PM
  #35
Mac n Gs
Gorton plz
 
Mac n Gs's Avatar
 
Join Date: Jan 2014
Country: United States
Posts: 10,160
vCash: 500
Who would be your top choice to hire?

Mac n Gs is offline   Reply With Quote
Old
12-10-2016, 08:35 PM
  #36
Irishguy42
Registered User
 
Irishguy42's Avatar
 
Join Date: Sep 2015
Location: Bayside
Country: United States
Posts: 8,515
vCash: 500
Quote:
Originally Posted by Mac n Gs View Post
Who would be your top choice to hire?
Manny

Irishguy42 is offline   Reply With Quote
Old
12-10-2016, 08:56 PM
  #37
Machinehead
Our Bayley is better
 
Machinehead's Avatar
 
Join Date: Jan 2011
Location: NYC
Country: United States
Posts: 63,632
vCash: 50
Manny's article is great but I'm confused by his use of expert agreement. Isn't the "expert" hivemind something we're trying to avoid?

30 out of 30 GM's would tell you Can Fowler is a great player. Every regular in this thread and Manny himself would tell you the opposite.

Machinehead is offline   Reply With Quote
Old
12-10-2016, 09:03 PM
  #38
eyjee
the basement
 
eyjee's Avatar
 
Join Date: Jun 2008
Location: Alphabetrium
Country: United States
Posts: 27,963
vCash: 500
Quote:
Originally Posted by Mac n Gs View Post
Who would be your top choice to hire?
Quote:
Originally Posted by Irishguy42 View Post
Manny
Manny, Micah, DTM About Heart (I forget his real name), Tierney, Carolyn Wilke, Jack Han.

Some really smart people doing some really great work. I'm forgetting so many. I'd also really love to steal Wiseman from the NWHL. That guy DEFINITELY rebuilt the Riveters this off-season with analytics in mind. Their defense is 6 big-bodied puck movers.

I'd take Fratkin off the Rivs and put her next to McDonagh right now.

eyjee is offline   Reply With Quote
Old
12-10-2016, 09:07 PM
  #39
Machinehead
Our Bayley is better
 
Machinehead's Avatar
 
Join Date: Jan 2011
Location: NYC
Country: United States
Posts: 63,632
vCash: 50
Machinehead ^^

Machinehead is offline   Reply With Quote
Old
12-10-2016, 09:25 PM
  #40
an Girar i
Registered User
 
Join Date: Oct 2014
Posts: 2,840
vCash: 500
Erik Karlsson is a 46.8% possession player and he's dead even in terms of corsiRel. So they're basically the same team without him on the ice. He's also an anchor to Marc Methot.

Methot without: 53.5
Karlsson without: 47.3

What does one Draw from that? Just a bad stretch? Numbers misleading?

an Girar i is offline   Reply With Quote
Old
12-10-2016, 09:35 PM
  #41
eyjee
the basement
 
eyjee's Avatar
 
Join Date: Jun 2008
Location: Alphabetrium
Country: United States
Posts: 27,963
vCash: 500
Quote:
Originally Posted by 100Gs View Post
Erik Karlsson is a 46.8% possession player and he's dead even in terms of corsiRel. So they're basically the same team without him on the ice. He's also an anchor to Marc Methot.

Methot without: 53.5
Karlsson without: 47.3

What does one Draw from that? Just a bad stretch? Numbers misleading?
Sample size

Let's trade for Zach Redmond.


Last edited by eyjee: 12-10-2016 at 10:17 PM.
eyjee is offline   Reply With Quote
Old
12-10-2016, 10:19 PM
  #42
an Girar i
Registered User
 
Join Date: Oct 2014
Posts: 2,840
vCash: 500
Quote:
Originally Posted by eyjee View Post
Sample size

Let's trade for Zach Redmond.
lol. come on man...

Redmond's sample size: 32 events
Karlsson's sample size: 1100+ events...

your comparing apples to avocados there

I agree with you about the Methot anchor part. Methot's only spent 60 minutes away from Karl. And, Karl is a much better player. But, it's just hard to justify those raw and rel numbers 1/3 of the seasons over already.

an Girar i is offline   Reply With Quote
Old
12-10-2016, 10:25 PM
  #43
eyjee
the basement
 
eyjee's Avatar
 
Join Date: Jun 2008
Location: Alphabetrium
Country: United States
Posts: 27,963
vCash: 500
Quote:
Originally Posted by 100Gs View Post
lol. come on man...

Redmond's sample size: 32 events
Karlsson's sample size: 1100+ events...

your comparing apples to avocados there

I agree with you about the Methot anchor part. Methot's only spent 60 minutes away from Karl. And, Karl is a much better player. But, it's just hard to justify those raw and rel numbers 1/3 of the seasons over already.
Wait, what?

The 'Sample Size' and the 'Let's trade for Redmond' were two very different thoughts in the same post. Karlsson/Redmond have nothing to do with one another

eyjee is offline   Reply With Quote
Old
12-10-2016, 10:41 PM
  #44
an Girar i
Registered User
 
Join Date: Oct 2014
Posts: 2,840
vCash: 500
my fault then I thought u were trying to use Redmond as an example for sample size problem like Karlsson, obviously not much of a comparison as i pointed out

an Girar i is offline   Reply With Quote
Old
12-11-2016, 01:52 PM
  #45
Mac n Gs
Gorton plz
 
Mac n Gs's Avatar
 
Join Date: Jan 2014
Country: United States
Posts: 10,160
vCash: 500
Incoming tweet storm from Sportlogiq. Interpret this however you want:

Mac n Gs is offline   Reply With Quote
Old
12-11-2016, 04:44 PM
  #46
Hunter Gathers
Electrified Angels
 
Hunter Gathers's Avatar
 
Join Date: Feb 2002
Location: NJ/NYC
Country: United States
Posts: 99,663
vCash: 500
Send a message via AIM to Hunter Gathers
Quote:
Originally Posted by eyjee View Post
It's wrong, but I've heard from multiple people in the show that you can't be a dick on Twitter. Not saying anyone is or isn't a dick on Twitter, but you can't be one. They said that all the guys who you think don't have Twitter accounts have them, and they are looking. Those guys being GMs, other FO people.

I went back a full calendar year and had to delete like 200 tweets criticizing the Rangers

EDIT - It's wrong meaning how much weight that seemingly goes into being a "good person" on Twitter in terms of if you want an NHL job.
Why is it wrong?

__________________
"Of course giving Sather cap space is like giving teenagers whiskey and car keys." - SBOB
"Watching Sather build a team is like watching a blind man with no fingers trying to put together an elaborate puzzle." - Shadowtron
My MS Excel crashed so I couldn't watch the game. - ReggieDunlop68
Hunter Gathers is offline   Reply With Quote
Old
12-11-2016, 04:58 PM
  #47
eyjee
the basement
 
eyjee's Avatar
 
Join Date: Jun 2008
Location: Alphabetrium
Country: United States
Posts: 27,963
vCash: 500
Quote:
Originally Posted by Hunter Gathers View Post
Why is it wrong?
I don't think you should be potentially disqualified from job positions because you post political beliefs or have conversations with other people on Twitter where you... prove them wrong(? - not exactly sure how to word that).

I get why people are turned off by that, and I get why tweeting things like: "Dan Girardi is a piece of **** and sucks at hockey" shouldn't be tweeted, but qualified people are being passed over for these reasons.

eyjee is offline   Reply With Quote
Old
12-11-2016, 05:00 PM
  #48
Hunter Gathers
Electrified Angels
 
Hunter Gathers's Avatar
 
Join Date: Feb 2002
Location: NJ/NYC
Country: United States
Posts: 99,663
vCash: 500
Send a message via AIM to Hunter Gathers
Quote:
Originally Posted by eyjee View Post
I don't think you should be potentially disqualified from job positions because you post political beliefs or have conversations with other people on Twitter where you... prove them wrong(? - not exactly sure how to word that).

I get why people are turned off by that, and I get why tweeting things like: "Dan Girardi is a piece of **** and sucks at hockey" shouldn't be tweeted, but qualified people are being passed over for these reasons.
I think that if you make it known that you are an ******* and use your on name out there in showing that you're an ******* that it is disqualifying to a number of employers. Saying someone is a "piece of ****" on Twitter is a great way to make yourself look like a "piece of ****".

If you want to say obnoxious things about folks, then do it anonymously. There's a reason I don't use my Twitter, LinkedIn, or Facebook to start trouble. They are heavily monitored by employers, industry folks, and everyone else under the sun.

Hunter Gathers is offline   Reply With Quote
Old
12-11-2016, 05:03 PM
  #49
eyjee
the basement
 
eyjee's Avatar
 
Join Date: Jun 2008
Location: Alphabetrium
Country: United States
Posts: 27,963
vCash: 500
Quote:
Originally Posted by Hunter Gathers View Post
I think that if you make it known that you are an ******* and use your on name out there in showing that you're an ******* that it is disqualifying to a number of employers. Saying someone is a "piece of ****" on Twitter is a great way to make yourself look like a "piece of ****".

If you want to say obnoxious things about folks, then do it anonymously. There's a reason I don't use my Twitter, LinkedIn, or Facebook to start trouble. They are heavily monitored by employers, industry folks, and everyone else under the sun.
Right, that was just an extreme example. In the past, I've tweeted things like: Girardi is bad, here's a reason why. After conversations with people in the show, I was told that tweets like that are frowned upon. If I'm looking for a job with the Rangers, and ******** on the Rangers, well, that's not going to be looked upon highly.

The example in the previous post was extreme. The one in this post is reality.

And I get it, I do, but I don't agree with it.

eyjee is offline   Reply With Quote
Old
12-11-2016, 05:08 PM
  #50
Hunter Gathers
Electrified Angels
 
Hunter Gathers's Avatar
 
Join Date: Feb 2002
Location: NJ/NYC
Country: United States
Posts: 99,663
vCash: 500
Send a message via AIM to Hunter Gathers
Quote:
Originally Posted by eyjee View Post
Right, that was just an extreme example. In the past, I've tweeted things like: Girardi is bad, here's a reason why. After conversations with people in the show, I was told that tweets like that are frowned upon. If I'm looking for a job with the Rangers, and ******** on the Rangers, well, that's not going to be looked upon highly.

The example in the previous post was extreme. The one in this post is reality.

And I get it, I do, but I don't agree with it.
Yeah, but Manny curses all over his Twitter and he's generally kind of a dick. I don't get how he thinks he looks presentable or professional at all. He looks like an angry kid who needs to put down his phone now and then. I like what he presents, but that is the exact kind of stuff that is a HUGE turnoff to any potential employer.

Hunter Gathers 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 06:59 AM.

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.