Arbusers

Hi,

I am wondering where everyone is getting their data? I want to build a model (to start and tune the model I want to focus on the Dutch Eredivisie as a test), but I need loads of past data to calculate statisical true odds. Now there are a lot of individual websites with all kinds of outcomes from all kinds of leagues. Is there a know website that tracks all matches of a certain/all sports?

Are there any recommendations for a statistical model?

Thanks!

DaanK wrote: ↑

Thu Mar 02, 2023 6:37 pm

Hi,

I am wondering where everyone is getting their data? I want to build a model (to start and tune the model I want to focus on the Dutch Eredivisie as a test), but I need loads of past data to calculate statisical true odds. Now there are a lot of individual websites with all kinds of outcomes from all kinds of leagues. Is there a know website that tracks all matches of a certain/all sports?

Are there any recommendations for a statistical model?

Thanks!

Hey DaanK,

There are tons of data sites out there, but it is not the most important thing. Congrats on your intention to create your own proprietary model!

web-for-inplay-stats-t8637/93048#p93048 My opinion here for in-play Stats sources is also valid for all data sources. In other words, the quality of the processing model you develop and its constant tuning is far more important than the input source you choose.

Later, you would do your own high-speed fundamental and technical processing. For example, these are some of my recent projects related to Dutch soccer:
- The "Goal-appearing model" comparison between Eredivisie vs Eerste Divisie, later own Handicap computing, BTTS, 1X2, and so on;
- The "Eredivisie vs KNVB Beker" discrepancies - frequencies and probabilities of occurrence;
- "Feyenoord Model 2022/2023" vs "Feyenoord Model Last 5 Seasons";
- The characteristics, probabilities, and additional calculations for attacking the relevant markets of the "Ajax 2022/23 mid-season crisis model";
- Ruud van Nistelrooy's tactical influence and the probabilities computing regarding PSV Eindhoven's possibility of a title attack.
- AZ Alkmaar's arbitrage positions and Pascal Jansen's tactical influence.
And so on.


Via Google you'll choose the sources of data suited you, but the vision of your own creative process is your responsibility.
I wish you success! Return here later.

This should help you get started: football-data.co.uk/netherlandsm.php

arb12 wrote: ↑

Thu Mar 02, 2023 11:10 pm

DaanK wrote: ↑

Thu Mar 02, 2023 6:37 pm

Hi,

I am wondering where everyone is getting their data? I want to build a model (to start and tune the model I want to focus on the Dutch Eredivisie as a test), but I need loads of past data to calculate statisical true odds. Now there are a lot of individual websites with all kinds of outcomes from all kinds of leagues. Is there a know website that tracks all matches of a certain/all sports?

Are there any recommendations for a statistical model?

Thanks!

Hey DaanK,

There are tons of data sites out there, but it is not the most important thing. Congrats on your intention to create your own proprietary model!

web-for-inplay-stats-t8637/93048#p93048 My opinion here for in-play Stats sources is also valid for all data sources. In other words, the quality of the processing model you develop and its constant tuning is far more important than the input source you choose.

Later, you would do your own high-speed fundamental and technical processing. For example, these are some of my recent projects related to Dutch soccer:
- The "Goal-appearing model" comparison between Eredivisie vs Eerste Divisie, later own Handicap computing, BTTS, 1X2, and so on;
- The "Eredivisie vs KNVB Beker" discrepancies - frequencies and probabilities of occurrence;
- "Feyenoord Model 2022/2023" vs "Feyenoord Model Last 5 Seasons";
- The characteristics, probabilities, and additional calculations for attacking the relevant markets of the "Ajax 2022/23 mid-season crisis model";
- Ruud van Nistelrooy's tactical influence and the probabilities computing regarding PSV Eindhoven's possibility of a title attack.
- AZ Alkmaar's arbitrage positions and Pascal Jansen's tactical influence.
And so on.


Via Google you'll choose the sources of data suited you, but the vision of your own creative process is your responsibility.
I wish you success! Return here later.

Thank you for your reply Arb12!

I indeed saw a load of websites offering data, I didn't even bother to try in-play sources haha! Are you Dutch aswell, you seem to know quite a lot of edges within this small league!

I am wondering what software/programming language you are using to get the data and to process this. I am currently trying to clean the data (struggling with it because I never really processed this kind of data. So a great learning oppertunity here!) in excel and or PowerBi. Note that I first want to see if I can calculate all the pre-match full time results before I start to dive into the halftime/in-play game. So I want to learn on the job.

Thanks in advance!

RB-Simon wrote: ↑

Fri Mar 03, 2023 9:07 am

This should help you get started: football-data.co.uk/netherlandsm.php

Thank you very much! This is a great start

What kind of data do you plan to use in your model? If it’s simple events such as attacks, shots, corners, goals (what you would find in football-data.co.uk) - then the whole history of any single league isn’t “loads of data” and will not be very helpful in building a model that will work going forward.

DaanK wrote: ↑

Fri Mar 03, 2023 10:47 pm

arb12 wrote: ↑

Thu Mar 02, 2023 11:10 pm

DaanK wrote: ↑

Thu Mar 02, 2023 6:37 pm

Hi,

I am wondering where everyone is getting their data? I want to build a model (to start and tune the model I want to focus on the Dutch Eredivisie as a test), but I need loads of past data to calculate statisical true odds. Now there are a lot of individual websites with all kinds of outcomes from all kinds of leagues. Is there a know website that tracks all matches of a certain/all sports?

Are there any recommendations for a statistical model?

Thanks!

Hey DaanK,

There are tons of data sites out there, but it is not the most important thing. Congrats on your intention to create your own proprietary model!

web-for-inplay-stats-t8637/93048#p93048 My opinion here for in-play Stats sources is also valid for all data sources. In other words, the quality of the processing model you develop and its constant tuning is far more important than the input source you choose.

Later, you would do your own high-speed fundamental and technical processing. For example, these are some of my recent projects related to Dutch soccer:
- The "Goal-appearing model" comparison between Eredivisie vs Eerste Divisie, later own Handicap computing, BTTS, 1X2, and so on;
- The "Eredivisie vs KNVB Beker" discrepancies - frequencies and probabilities of occurrence;
- "Feyenoord Model 2022/2023" vs "Feyenoord Model Last 5 Seasons";
- The characteristics, probabilities, and additional calculations for attacking the relevant markets of the "Ajax 2022/23 mid-season crisis model";
- Ruud van Nistelrooy's tactical influence and the probabilities computing regarding PSV Eindhoven's possibility of a title attack.
- AZ Alkmaar's arbitrage positions and Pascal Jansen's tactical influence.
And so on.


Via Google you'll choose the sources of data suited you, but the vision of your own creative process is your responsibility.
I wish you success! Return here later.

Thank you for your reply Arb12!

I indeed saw a load of websites offering data, I didn't even bother to try in-play sources haha! Are you Dutch aswell, you seem to know quite a lot of edges within this small league!

I am wondering what software/programming language you are using to get the data and to process this. I am currently trying to clean the data (struggling with it because I never really processed this kind of data. So a great learning oppertunity here!) in excel and or PowerBi. Note that I first want to see if I can calculate all the pre-match full time results before I start to dive into the halftime/in-play game. So I want to learn on the job.

Thanks in advance!

Why is my interest in the Netherlands' soccer? In my business work, the analysis of Dutch football is very important. The legacy of Hiddink, Cruyff, Michels, De Haan, Beenhakker, and Advocaat and their styles and influences on modern football is mind-boggling. Never forget the following:
- The Dutch team in the EURO '88 Final under Michels (Very Great Respect to Rinus Michels and Valeriy Lobanovskyi here!!! The 11 Dutch vs the 11 rebranded Ukrainians made an unforgettable tactical battle!);
- The AC Milan team in the Sacchi-Rijkaard-Van Basten-Gulit era;
- The Real Madrid team under Beenhakker's governance;
- The Ajax team under Van Gaal's management;
- The Barcelona team under Cruyff's mastery coaching.
Did you recognize some elements of Guardiola's style of coaching here? If you know in detail the coaching style of Carlo Mazzone and Johan Cruyff, I think you will get 2/3rd of Guardiola's coaching decisions. You need to build an additional matrix later. Therefore it is now very easy to attack Man City's markets.
- And many more, and many others.
This is in aid of fundamental analysis - both the origin and the modern. Therefore, Martin Jol and Ten Hag's styles could be decomposed to some extent in our analysis. And so on.

The most important thing is not the tool (programming language, software), but the Algorithm and Model developed by you. That's your task only.
I hear some use Python, R, and even Excel (Open Office's Calc and Spreadsheet), etc. But again, this isn't the most important thing, but the Core of the real Model that powers your business. Warren Buffett only uses a pen, paper, and a simple calculator. The pattern is in his Mind, not in the software tool.

Wish you great success!