Betting Strategy and Model Validation

1. Summary

Betting Strategy and Model Validation, I analyse the staking model of a sportsbook agency which follow bets from consultancy firm A.

The rebates value inside the dataset doesn't count in Return and PL since it is not includes in win/loss profit but only awarded upon hit a certain amount of stakes. Rebates is a marketing strategy for bankers and also agent to fight for revenue and occupy the market shares.

• Normally rebates only offers by credit market sportsbook makers.
• The maximum stakes per bet of cash market normally only up to HKD10,000 while credit market can up to millions HKD1,000,000.
• Normally only high volume agents, small bookmakers and also high volume sportbook consultancy firms will able to get the rebates.

2. Case Study

2.1 Section A

• Natural Language Analysis : Applied NLP to filter team name.

2.2 Section B

• Betting Strategy and Model Validation - Part I (Sep-2016), (Alternative Link) : Analyse the staking breakdown.

2.3 Section C

• Betting Strategy and Model Validation - Part II (Notes : The latest Rmd files lost but only html file available)

  • Kindly click on regressionApps to use the ShinyApp
  • 
  • Kindly click on KellyApps to use the ShinyApp
  • <iframe width="560" height="315" src="https://www.youtube.com/embed/42NOxuYjOQo" frameborder="0" allowfullscreen></iframe>
  • <iframe width="560" height="315" src="https://www.youtube.com/embed/NDUkg4jHmiA" frameborder="0" allowfullscreen></iframe>
  • Kindly refer to [English Soccer League] for analysis on only English soccer which is pretest on my personal research.
  • You are feel free to browse over the summary of Kelly investment fund BRSum (BRSum.Op, BRSum.Hi, BRSum.Lo, BRSum.Cl, BRSum.Vo, BRSum.Ad).
  • Extention : summarise the comparison of Kelly fund due to the BSMV II heavily loading...

3. Old File

I splited below file to be two part as list above since there are a lot of gragh and high volume dataset. The Return and PL columns suddenly noticed wrong figure and cause the profit of below files are not accurate. Secondly, the old files are not completed as it only stop at section 3 which just a breakdown display and summary of staking data but section 4 and so fort analyse the dataset.

Kindly refer to new files in section [2. Case Study] to know the profit and return of firm A from agent A.

• Betting Strategy and Model Validation (Old) (Remarks: Initially one file, but due to file size and finally seperate to few sections and files.)
• Betting Strategy and Model Validation - Part I (Aug-2016)
• Miscellaneous : Introduction of R Packages
• Miscellaneous : Introduction of R Packages (Shiny)
• Miscellaneous : Unable display highchart at html format
• Betting Strategy and Model Validation - Part II (Sep-2017), (Alternative Link)
• Due to KellyApps shiny app heavily loading, below video is the original file.

<iframe width="560" height="315" src="https://www.youtube.com/embed/TEhmN-Of--Y" frameborder="0" allowfullscreen></iframe>

4. Future Work

• Part II in Application of Kelly Criterion model in Sportsbook Investment which apply shinyApp.
• Algorithmic scrapping team name and filtering, matching team name.
• Writing a website to apply the staking model which is similar with http://www.matchodds.org/.

5. Reference

1. Odds Modelling and Testing Inefficiency of Sports Bookmakers : Rmodel by ®γσ, Eng Lian Hu (2016)
2. Apply Kelly-Criterion on English Soccer 2011/12 to 2012/13 by ®γσ, Eng Lian Hu (2014)
3. Creating a Profitable Betting Strategy for Football by Using Statistical Modelling by Niko Marttinen (2006)

---

Powered by - Copyright© Intellectual Property Rights of Scibrokes®