live-odds-checker-with-arbitrage-betting

A python script that scrapes betting websites APIs in real time to gather live events data and find arbitrage betting opportunities

Supported websites

1. Sisal
2. AllInBet
3. Better
4. Eurobet
5. VinciTu
6. PlayMatika

Supported sports and bets:

• football: 1X2, DC, GG/NG
• basketball: T/T, P/D
• tennis: T/T

Data scraped about matches:

• odds
• sport (eg: "football")
• match name (eg: "Inter - Juventus")
• tournament (eg: "Serie A")
• start date and time
• running time
• score

Functions and purposes:

1. scraping sport betting websites' APIs to gather events data
2. match events between different providers, creating also a new set of odds which are the highest ones found for each event
3. check for arbitrage opportunities

This whole process run infinitely in a loop

Output:

The results of each iteration of this loop will be written in the output folder, in particular the files you'll find are:

1. events.json --> all the raw events dictionaries, which contains info, odds and highest_odds for each singular event
2. arbs.json --> all the arbs found (if there are)
3. odds.json --> a dictionary where the key is events bet_radar_ids and the value is their odds
4. info.json --> a dictionary where the key is events bet_radar_ids and the value is their info
5. events_by_sport.json --> a dictionary where the key is the sport name and the value is a list of all the events for that sport

I uploaded both the files in the output folder and the script logs in files/logs so that you can have a look at an example run.

How to:

1. clone the github repo
2. create a venv & activate it
3. run pip install -r requirements.txt
4. set script behavior by changing config.py
5. run python main.py
6. OPTIONAL: if you are on a Linux system install python package uvloop, as it has better performances compared to asyncio

Disclaimer

Since i didn't implement a proxy you will be sending requests with your own IP, pay attention to not having the value of config.sleep_after_run set too low or running the script for too long. As websites could temporarily limit your IP (you can see that in the scrapers logs in files/logs/scrapers/, it will tell you that the response return a 403 code, look at eurobet.log for example)