Fooball bet optimizer for Betfair

[Porcupine]

Hi forum. Not a big contributor, but a big reader of this forum, and today I want to share the story of my love for sports and math, and what came out of that. I am a mathematician myself, with degree in Combinatorics & Optimization, and at the same time a big football fan. That led me to discover and start to admire Betfair, a sports exchange that most punters already know. With time, I've realized the huge potential that lays behind it, where simply with your smarts you can make money. I've started to play around with different mathematical models that I've build for horseracing and football. That led me to different "discoveries" and realizations about Betfair markets. One interesting realization was the fact that football betting has a vast, and unknown to most, potential for optimizing bets across multiple markets... That's where it might get a bit complicated, but you are a smart crowd, and I am sure you can follow my train of thoughts here. An average football game has dozens of different markets. Examples of the market are: Match Odds, Over/Under x Goals, Correct Score, Both Teams to Score, and so on. Every one of those is a separate market on which you can bet independently. Popular games can have more than 60 different markets. What many do not realize is that many of those markets overlap. By that I mean, the outcome on one market will imply certain outcome on another market. For example, betting that there will be Under 0.5 Goals will imply that Both Teams will NOT Score, it will also imply a Correct Score of 0-0. And this is very basic example of one overlap, and there are hundreds of such overlaps, some of which are very complex and involve many different markets. But if different bets on different markets represent the same outcome, which one is more profitable? Which odds are better? For example, betting that there will be Under 1.5 goals scored, is the same as backing scores 0-0, 1-0 and 0-1 on Correct Score market... Most would simply bet on Under\Over market, but does it have better odds than backing 3 separate Correct Scores? You can take a calculator, or open Excel and calculate it in few minutes, but that's not a proper way to do it, and this is only one overlap out of hundreds. There is no way for human to calculate in timely manner all possible overlaps on all possible markets to find a best possible way to bet. That's when I realized that this is a perfect opportunity to apply mathematical optimization to. For few months, I've been building and polishing a mathematical model that would incorporate all possible football markets and overlaps between them. The idea was to have a model that would analyze all possible markets and their prices and come up with the most optimized way to place bets (think "should I bet on Under 1.5 goals or bet on 0-0, 1-0, 0-1? Which one is better?" but hundreds of times more complex). The result was a completely new way of football betting on Betfair. My model would wrap all separate Betfair markets into one huge market, and create a new layer over it, with one simple input method: you choose a bet amount, and pick scores that you want to win, break-even and lose on, and the model will calculate and place bets across all available markets achieving the maximum possible win. I know that this part might be especially confusing, since it presents a completely new way to bet on football. Simple example: Man U playing against Everton in Manchester. I want to win on scores 1-0, 2-0, 2-1, 3-1, and 1-1. I want to break-even (get my money back) on scores 0-0, 1-1, 0-1, and I am willing to lose 20 pounds on all other scores. My model would analyze all markets and my input and would tell me that my maximum win would be let's say 14 pounds (on all winning scores that I have picked), and to achieve it, 13 bets will be placed on 6 different markets. If I were happy with those numbers, it would place those bets for me. Then I made it a bit more flexible, where I could distribute the weight between different scores, for example if I want to win as much as possible (100%) on score 1-0, but I am ok with winning half of that on score 3-1 (50%), or even 10% of that max win. Or what if I am ok with losing 20 pounds on score 0-4, but I only want to lose 5 pounds on score 1-2. That made it even more customizable, but still kept it under one simple input interface. The result was a surprisingly fun way of betting on football. No more going through different markets, thinking where I should bet... Just analyze the match, think about which scores could probably happen, pick winning and break-even scores and optimize. I later added a lot more math on top of it, to automate and analyze markets even more, but I will not go there this time. Now this is the only tool that I use to bet on football. I later realized that there is no harm for others to know and use the same system, because the key ingredient here is your football knowledge, which in this case is your input (winning and break-even scores), and the model itself only optimizes your input to guarantee the placement of best possible bets. And you cannot dilute the winning system by going public with this system, because the system doesn't provide you one, your football knowledge and your input is the winning (or losing) system here. So I decided to build a Betfair application that anyone can use. I hired a developer and got official Betfair Vendor certification. Now I am trying to share it with everyone, because I truly believe this system provides the smartest way to bet on football. Being an honest guy, I can say that this post has twofold purpose. First, I believe the system that I developed will be beneficial for anyone betting on football, it does something no other system is currently doing. Secondly, I will not lie, there is a self-interest in it. Developing a Betfair certified application takes time, money and effort. Even though the system is free to try, it is still a paid system, but I honestly believe that if used properly (and smartly), the surplus amount that the system will optimize for you will exceed the cost of using the system. Due to the second point, and knowing the rules of the forum, I will refrain from posting a link, or even mentioning the name of the application. Even though, I've tried to contact admins of the forum in regards to conditions under which I can create such a thread with a link in it... I haven't heard anything back. Not sure why. So to stay in line with the topic of this forum (Systems & Strategy), I would like to discuss with you the proposed system, approach and any recommendations you might have to improve it. I understand that you might have questions about the system, so if needed, I am willing to explain more, to start a meaningful conversation. My name is Greg by the way. Let’s bet smart

[Porcupine]

Re: Fooball bet optimizer for Betfair Holy moly, I haven't realized how long the post came out to be... I'll have to learn some mastery of briefness :)