[Journey] Predicting Sports Results with Artificial Intelligence

My man, if you can beat the closing odds on Asian bookmakers, mainly Pinnacle, over the long term, after thousands of bets, you can sit back and enjoy the money... I wish you the best!
 
Which tool/framework do you use?
Keras, Tensorflow, and Theano
My man, if you can beat the closing odds on Asian bookmakers, mainly Pinnacle, over the long term, after thousands of bets, you can sit back and enjoy the money... I wish you the best!
Thanks!

I've lost all bets so far on the weekend so I'll analyze everything after the weekend is over and all games are played. I don't know whether I'm going to place bets for the next weekend or not. I need to optimize my tool further but currently lacking the knowledge and time to do what I want to do. Over the next weeks/month, I'll continue learning and update my tool accordingly.
 
Any possibility to share your code and your dataset? Or to set up a private group to contribute? I ve read some papers on this. I really doubt we can beat betting sites to win money, but it would be interesting.
 
Good luck @issorc !!!
Thank you! :)
Any possibility to share your code and your dataset? Or to set up a private group to contribute? I ve read some papers on this. I really doubt we can beat betting sites to win money, but it would be interesting.

I'll consider this in the future. Right now I'll not work on the project for two weeks most likely (exams + vacation) and after that I'll get together with a mate who does his soccer analysis without a neural network but brings in good statistical knowledge + knowledge of soccer in general. He had good results so far without using deep learning so we'll try to merge our models and bring up a better version with extended statistical analysis + deep learning. I've just started development so I'm currently don't want to release anything because the code is messy and I've written many different scripts to get the data, convert the data, run the neural network, analyse results etc. All these scripts need to be run separately so I guess I'd even take me a day to write a documentation how that shit works lol. There's still a lot of manual work involved and as long as I don't have all modules integrated and mainly automated I actually wouldn't want to release anything. But your input is still appreciated and as soon as I'll consider to open doors for other programmers or a small group of people you'll hear from me.
 
Midtjylland sadly won 3:0 instead of the predicted 1:0 otherwise my profit would be much greater. In total I won 6/28 games so far with arsenal still to play (while counted in the 28). My money turned from 15€ at the beginning of this journey to 48,58€ so far. Will post a huge update tomorrow, I'm really glad how things turned out so far. Tripled my money within 7 days. Nevertheless I've been extremely lucky. 5 Minutes before I cashed a bet out at 16€ and one minute later one team scored and crashed the bet. Perfect timing.
 
just a thought: share your daily updates in the mornings with the predicted scores for the day, that we we can piggyback on your journey - i love sports betting too, so 3x investment in 7 days sounds okay to me - yes i know it is unsafe and dangerous and all, i still do it, and it may be time for AI to get shit done :)
 
@issorc this is fantastic! Good work mate, following, hopefully we will watch you start to rake in the $$$
Thank you! :) I'll keep you guys updated about the figures :)
just a thought: share your daily updates in the mornings with the predicted scores for the day, that we we can piggyback on your journey - i love sports betting too, so 3x investment in 7 days sounds okay to me - yes i know it is unsafe and dangerous and all, i still do it, and it may be time for AI to get shit done :)

I posted every bet (except one I forgot but it was lost anyway) since the creation of the thread :) Second, Third and Fourth English League to play today so I'll do my analysis later and in case that I find something interesting I'll post it here :) Sports betting always has been fun to me too but it's even more fun if you win something :D I wouldn't care anyway if I lose all my money again. I started with a 15€ bankroll and do not rely on any earnings from that in any way so I guess I'm good to go. Wish you good luck in your further sports betting adventures! :p


Arsenal has won the 2:0 prediction yesterday too so I'm at 7/28 bets (25% win rate) and 50,01€
All bet predictions from the weekend will be analyzed with income and earnings when I'll post bets for today :)
 
good thing i commented on the post than so i can receive mails for the updates, can't wait to se todays predictions :) oh and good luck with the AI, do you plan on making a website/service out of it, or you will just use it for yourself?
 
good thing i commented on the post than so i can receive mails for the updates, can't wait to se todays predictions :) oh and good luck with the AI, do you plan on making a website/service out of it, or you will just use it for yourself?
There's even an option at the top right corner of each thread to watch the thread and receive alerts and email notifications ;)
Thanks and I've only betted on 28 games so far and I've been testing this tool for two weeks, so far I don't have any plans for a website or service. This will need a lot more testing and development to maybe one day turn into something really serious :)
 
If this will turn into something really powerful and effective you can expect gambling sites contacting you and offering millions of $$ for that tool :)
Good luck!
 
If this will turn into something really powerful and effective you can expect gambling sites contacting you and offering millions of $$ for that tool :)
Good luck!
Thank you! :)

Here quick today's tips, sadly ain't got time for more:

Barnsley 0:0 QPr
Brentford 0:0 Derby
Bristol 2:0 Bolton
Cardiff 1:1 Leeds
Hull 1:0 Preston
Nottingham Forest 1:1 Fulham
Birmingham 1:0 Sheffield Wednesday
 
The model is predicting national leagues only? 1/6, not bad!
Yes, I'd need to figure out a factor to weight the different leagues. A team that always wins 3:0 in the premier league is most likely better than one that always wins 3:0 in League 1 :) Therefore I didn't do any predictions for CL or EL yet. And so far the bet that was right had odds of 7.00 so I exactly made the money back for the seven bets. Still placed a system bet on top of that where I lost 4,2€ (if birmingham does not win 1:0 this evening) but I don't care too much about that.
Good Luck kind sir.
I really appreciate your kind words Sir dog.
 
Also you can try betting a system with final results to see if that has a bigger ROI.
 
Also you can try betting a system with final results to see if that has a bigger ROI.
I'll try that out in the future but currently sadly don't have time to implement that. Actually, I wouldn't care if I'm not able to make money with this in the future because many people have tried already without success. But I'll use this project to implement stuff I'll be learning over the next few months and wouldn't be sad either if it turns out that I can make some money with it :D
 
Wow such diferent journey,
But there a proverb in my country say "ball is round" so basicly anything can happen.
Best of luck op
 
It is more than extremely interesting.

I've been thinking about going into that field for a while (I love betting, odds and numbers - and I love AI), but my hands are full at the moment.


You should definitely consider including the odds in your data. If you ever create a facebook/skype group, count me in. I have a lot to share :)
 
Back
Top