The Hotel End
March 29, 2024, 07:33:51 am
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News:
 
  Home Help Search Arcade Downloads Gallery Links Staff List Calendar Login Register Chat  

Data Based Approach

Pages: [1] 2 3 4   Go Down
  Print  
Author Topic: Data Based Approach  (Read 4306 times)
0 Members and 1 Guest are viewing this topic.
Jim Hall Fixit
Hero Member
*****
Offline Offline

Posts: 633



View Profile
Badges: (View All)
Tenth year Anniversary 500 Posts Avatar
« on: May 18, 2021, 05:31:16 am »

https://www.northamptonchron.co.uk/sport/football/cobblers-player-ratings-for-the-season-who-had-the-highest-average-rating-and-who-won-the-most-star-man-awards-3239266

So the club say they want to adopt a data based approach to player development and procurement??

This simple list, although not very objectively based, shows the man that for a long time was the first name on the team sheet at rock bottom with an average (yes average!) of 6 out of 10.

I’m all for data driven but if everything is pointing to a conclusion you really ignore it at your peril.
Report Spam   Logged

"My three children were born in Northamptonshire so I like to think I'm an adopted Northamptonian and it's the same for Rico and Sammo. It means a lot to us because this is our county.

Share on Facebook Share on Twitter

Deepcut Cobbler
Administrator
*****
Offline Offline

Posts: 14652



View Profile
Badges: (View All)
Search Level 6 Windows User
« Reply #1 on: May 18, 2021, 05:51:22 am »

Which one are you talking about?
Report Spam   Logged

“They shall grow not old as we that are left grow old:
Age shall not weary them, nor the years condemn.
At the going down of the sun and in the morning
We will remember them.” Laurence Binyon

The Hotelend Grand National Sweepstake Champion 2009
Winslow Lee
Hero Member
*****
Online Online

Posts: 3150


View Profile
Badges: (View All)
Search Apple User Mobile User
« Reply #2 on: May 18, 2021, 06:22:37 am »


This simple list, although not very objectively based, shows the man that for a long time was the first name on the team sheet at rock bottom with an average (yes average!) of 6 out of 10.


One of KC’s top targets too! in Edmondson, though Rose another seemingly perpetual starter only a few places above him.
Report Spam   Logged
Clarity
Hero Member
*****
Offline Offline

Posts: 1020



View Profile
Badges: (View All)
1000 Posts Fifth year Anniversary Search
« Reply #3 on: May 18, 2021, 06:27:38 am »

https://www.northamptonchron.co.uk/sport/football/cobblers-player-ratings-for-the-season-who-had-the-highest-average-rating-and-who-won-the-most-star-man-awards-3239266

So the club say they want to adopt a data based approach to player development and procurement??

This simple list, although not very objectively based, shows the man that for a long time was the first name on the team sheet at rock bottom with an average (yes average!) of 6 out of 10.

I’m all for data driven but if everything is pointing to a conclusion you really ignore it at your peril.
That’s not data in the Chron that’s peoples opinions. Completely different
Report Spam   Logged
Deepcut Cobbler
Administrator
*****
Offline Offline

Posts: 14652



View Profile
Badges: (View All)
Search Level 6 Windows User
« Reply #4 on: May 18, 2021, 06:41:10 am »

That’s not data in the Chron that’s peoples opinions. Completely different

It's not even people's opinions, it's James Henaghans opinion from his post match subjective assessments.  Wink
Report Spam   Logged

“They shall grow not old as we that are left grow old:
Age shall not weary them, nor the years condemn.
At the going down of the sun and in the morning
We will remember them.” Laurence Binyon

The Hotelend Grand National Sweepstake Champion 2009
Jim Hall Fixit
Hero Member
*****
Offline Offline

Posts: 633



View Profile
Badges: (View All)
Tenth year Anniversary 500 Posts Avatar
« Reply #5 on: May 18, 2021, 06:57:30 am »

Which one are you talking about?

The one at 25th with a score of 6 - Edmundson.
Report Spam   Logged

"My three children were born in Northamptonshire so I like to think I'm an adopted Northamptonian and it's the same for Rico and Sammo. It means a lot to us because this is our county.
Jim Hall Fixit
Hero Member
*****
Offline Offline

Posts: 633



View Profile
Badges: (View All)
Tenth year Anniversary 500 Posts Avatar
« Reply #6 on: May 18, 2021, 07:01:10 am »

It's not even people's opinions, it's James Henaghans opinion from his post match subjective assessments.  Wink

Fair point, but give the lad credit, he can’t be that far out can he - week in week out? Generally speaking I pretty much agreed with most of his scorings. My point, not very well made, is unless you consider Mr H is on a different planet every week ... what was the data that put Edmundson in the team week after week ?
Report Spam   Logged

"My three children were born in Northamptonshire so I like to think I'm an adopted Northamptonian and it's the same for Rico and Sammo. It means a lot to us because this is our county.
threeinabed
Hero Member
*****
Offline Offline

Posts: 3026


View Profile
Badges: (View All)
Linux User Mobile User Windows User
« Reply #7 on: May 18, 2021, 07:05:49 am »

They didnt use data last year - they are going to use it more moving forwards.

the team wasnt picked based on C&E ratings!

it doesnt tell us anything we didnt know, as a team we werent good enough, as individuals that majority of them werent / arent good enough
Report Spam   Logged
Deepcut Cobbler
Administrator
*****
Offline Offline

Posts: 14652



View Profile
Badges: (View All)
Search Level 6 Windows User
« Reply #8 on: May 18, 2021, 07:10:38 am »

Fair point, but give the lad credit, he can’t be that far out can he - week in week out? Generally speaking I pretty much agreed with most of his scorings. My point, not very well made, is unless you consider Mr H is on a different planet every week ... what was the data that put Edmundson in the team week after week ?

Leeds will sell/give Edmundson to P@@h and he'll score a hatfull of goals before being sold for £10m+.  We all know it's going to happen.  Grin
Report Spam   Logged

“They shall grow not old as we that are left grow old:
Age shall not weary them, nor the years condemn.
At the going down of the sun and in the morning
We will remember them.” Laurence Binyon

The Hotelend Grand National Sweepstake Champion 2009
Manwork04
Hero Member
*****
Offline Offline

Posts: 9314



View Profile
Badges: (View All)
Windows User Mobile User Spammer 25 Posts in one day
« Reply #9 on: May 18, 2021, 08:13:22 am »

Leeds will sell/give Edmundson to P@@h and he'll score a hatfull of goals before being sold for Ł10m+.  We all know it's going to happen.  Grin
Grin You just know that’s going to happen Again!
Report Spam   Logged

Rule Britannia
Smoking Boots
Jr. Member
**
Offline Offline

Posts: 60


View Profile
Badges: (View All)
« Reply #10 on: May 18, 2021, 10:13:57 am »

https://www.bbc.co.uk/news/newsbeat-55816277

Data Base on the cheep.
Report Spam   Logged
manbearpig
Newbie
*
Offline Offline

Posts: 2


View Profile
Badges: (View All)
« Reply #11 on: May 18, 2021, 11:37:43 am »



I have the (mis)fortune of working with databases all day and providing reporting/stats/data to .

Which means I have heard business's claim to aim to be, or actively be data driven in their decision making. So, when that remark was made i have to admit i chuckled somewhat. There's usually a huge gulf between what is said and what actually is done. And sadly thats usually a leadership issue, you could have the best analyst's in the world with the cleanest data, but if the top table don't utilise it then it's utterly worthless.

Now whilst the club would be foolish to give away what they are trying to do datawise, as far as i can tell that we have but a single analyst, and setting up and sorting something like this out from scratch is usually more than one individual can take on top of there day job. So without seeing anything about hiring more bodies to really interrogate data and build reporting and algorithms I am tempted to say this is a bit of a placating statement to sound modern.

Barnsley have proved this approach can work very well in football, admittedly they happen to have Billy Beane himself as a part owner so that's not entirely a surprise. But done well this could really help the club punch above it's weight whilst keeping the finances in good shape.

I genuinely hope my skepticism is misplaced as i'd really like to see us follow through on this as it really could benefit the club in the long term.
Report Spam   Logged
Manwork04
Hero Member
*****
Offline Offline

Posts: 9314



View Profile
Badges: (View All)
Windows User Mobile User Spammer 25 Posts in one day
« Reply #12 on: May 18, 2021, 11:38:33 am »

Like a budgerigar ?
Report Spam   Logged

Rule Britannia
Manwork04
Hero Member
*****
Offline Offline

Posts: 9314



View Profile
Badges: (View All)
Windows User Mobile User Spammer 25 Posts in one day
« Reply #13 on: May 18, 2021, 11:42:06 am »


I have the (mis)fortune of working with databases all day and providing reporting/stats/data to .

Which means I have heard business's claim to aim to be, or actively be data driven in their decision making. So, when that remark was made i have to admit i chuckled somewhat. There's usually a huge gulf between what is said and what actually is done. And sadly thats usually a leadership issue, you could have the best analyst's in the world with the cleanest data, but if the top table don't utilise it then it's utterly worthless.

Now whilst the club would be foolish to give away what they are trying to do datawise, as far as i can tell that we have but a single analyst, and setting up and sorting something like this out from scratch is usually more than one individual can take on top of there day job. So without seeing anything about hiring more bodies to really interrogate data and build reporting and algorithms I am tempted to say this is a bit of a placating statement to sound modern.

Barnsley have proved this approach can work very well in football, admittedly they happen to have Billy Beane himself as a part owner so that's not entirely a surprise. But done well this could really help the club punch above it's weight whilst keeping the finances in good shape.

I genuinely hope my skepticism is misplaced as i'd really like to see us follow through on this as it really could benefit the club in the long term.
Have you heard of the following, Ai, machine learning, anomaly detection?
These things stop the need for data analysis which is subjective.
Report Spam   Logged

Rule Britannia
manbearpig
Newbie
*
Offline Offline

Posts: 2


View Profile
Badges: (View All)
« Reply #14 on: May 18, 2021, 11:57:25 am »

AI and Machine Learning both need a humans to create, train and maintain. Infact there is inherant bias in machine learning if training data for a model not carefully selected for the appropriate task. Plus all models are wrong, some just happen to be useful.
Humans need to be continually involved as nothing is static, and the people with those exact skills are not cheap. AI and machine learning are incredibly frustrating as whilst they have some incredible uses thet are far from perfect. Just throwing out buzz words means nothing. They're far from simple things to create well.
Report Spam   Logged
Gen.Disorda
Hero Member
*****
Offline Offline

Posts: 2924



View Profile
Badges: (View All)
Level 6 Linux User Mobile User
« Reply #15 on: May 18, 2021, 11:57:56 am »

Data is obtained from companies such as opta stats. The club would simply be agreeing KPIs for particular positions and using the results to vet transfer targets.
Report Spam   Logged

BackOfTheNet
Hero Member
*****
Offline Offline

Posts: 5884


View Profile
Badges: (View All)
Search Level 6 Combination
« Reply #16 on: May 18, 2021, 13:20:42 pm »

Data is obtained from companies such as opta stats. The club would simply be agreeing KPIs for particular positions and using the results to vet transfer targets.

I would expect this to be the case. A subscription to an existing service would be far more sensible than trying to build your own model from scratch, as that's a huge undertaking that would, frankly, be beyond our means. Of course, that doesn't afford you a competitive advantage as you only have access to the same pool of data as every other subscriber, but that's the trade-off you always have to make in a buy vs build selection process.
« Last Edit: May 18, 2021, 13:42:06 pm by BackOfTheNet » Report Spam   Logged

The Hotelend Grand National* Sweepstake Champion 2020
OCoole
Hero Member
*****
Offline Offline

Posts: 548



View Profile
Badges: (View All)
Windows User Sixth year Anniversary Level 5
« Reply #17 on: May 18, 2021, 13:27:04 pm »

Have you heard of the following, Ai, machine learning, anomaly detection?
These things stop the need for data analysis which is subjective.

Forgive me for saying this but this is a remarkably arrogant question to ask someone in this field given that you seemingly have back-of-the-fag-packet knowledge of the topics listed?

Like Manbearpig I also work in a data focused role - including using machine learning techniques to build models for either predictions or decision making - and their most recent reply is a very good explanation for you about what machine learning and AI are actually used for. As Gen Disorda has also said it is highly unlikely that a football club (certainly in l2) would actually have need for someone to build statistical models, rather it would be someone using analytical tools such as Wyscout, Opta, Statsbomb etc to carry out decriptive statistics and then present it to the head of recruitment or the first team manager.

As for "These things stop the need for data analysis which is subjective"... Well, often a data analyst / scientist / machine learning engineer will carry out exploratory data analysis first before model building but yes it is not a necessity. However to suggest that machine learning is not subjective is wrong. 'Machine learning' is just another way of saying supervised learning - because it requires someone to build, monitor and prune a model. Depending on how complex and overarching model can be it can become incredibly difficult to know if your model is accurate or not as there is no set way of knowing.

If you want to learn more about it in a simple way I suggest googling 'overfitting' in regards to classification and regression models. If you want to learn why it can be so complicated then try googling 'naive bayes' or 'multiple linear regression'  Grin
Report Spam   Logged
Manwork04
Hero Member
*****
Offline Offline

Posts: 9314



View Profile
Badges: (View All)
Windows User Mobile User Spammer 25 Posts in one day
« Reply #18 on: May 18, 2021, 13:33:22 pm »

Here’s one I made earlier the last time in was in Walldorf.

https://www.sap.com/uk/products/sports-one.html

Fag packet indeed. 😂
Report Spam   Logged

Rule Britannia
OCoole
Hero Member
*****
Offline Offline

Posts: 548



View Profile
Badges: (View All)
Windows User Sixth year Anniversary Level 5
« Reply #19 on: May 18, 2021, 13:37:52 pm »

Here’s one I made earlier the last time in was in Walldorf.

https://www.sap.com/uk/products/sports-one.html

Fag packet indeed. 😂

Well colour me shocked. Again you'll have to forgive me for saying this, but your initial reply to Manbearpig hardly demonstrated any expertise.
Report Spam   Logged
Pages: [1] 2 3 4   Go Up
  Print  
 
Jump to:  

Powered by EzPortal
Parental guidance is urged as this messageboard may not be suitable for all persons especially those under the age of 16 as the forums may contain words, phrases and expressions not considered appropriate for a younger audience so please express caution. If any posts in the forums offend you, please let us know and we will look at them and if we agree with your complaint, we will remove them. You are personally responsible and potentially liable for the contents of your posting and may be sued should your posting contain content of a defamatory or other illegal nature. Every message posted leaves a traceable IP number. We check the forums at various times of the day and remove offending posts. Other supporters are welcome but abusive or silly posts will be removed and the offenders potentially barred from future access to the site. We advise that you never reveal any personal information about yourself or anyone else (for example: telephone number, home address or email address), and please do not include postal addresses of any kind. This messageboard is not endorsed or in any way affiliated with Northampton Town FC. All postings on this board become copyright of The Hotel End & may not be reproduced without the permission of the board administrator. By signing up to this message board you agree to this. The Hotel End cannot be held liable for the actions or postings of its members. The Hotel End reserve the right to edit, delete, move or close any thread for any reason. The Hotel End may disclose user information to government authorities at their discretion or when required by law. The Hotel End may also disclose user information when The Hotel End has reason to believe that someone is causing injury to or interference with its rights or property, other The Hotel End users, or anyone else that could be harmed by such activities. By registering for The Hotel End, you agree to indemnify The Hotel End its representatives, and agents, and hold them harmless from any and all claims (including claims for legal fees) which may arise from your participation on the The Hotel End. You also agree that The Hotel End is not responsible for the materials posted by users of The Hotel End. In addition, you grant The Hotel End and its affiliates, worldwide, royalty-free perpetual, irrevocable, non-exclusive right and license to use, reproduce, modify, adapt, publish, translate, create derivative works from, distribute, perform and display any message or content posted on The Hotel End and/or e-mail sent by you to The Hotel End (in whole or in part). The Hotel End reserves the right to make the rules up as it goes along. Thank you - The Hotel End I love Quidco
Bookmark this site!
Powered by SMF | SMF © 2016, Simple Machines
Privacy Policy