This project is a Python-based tool that scrapes betting odds from various bookmaker websites. It aims to identify the best odds for each team playing on a given day, allowing users to quickly compare odds, helping make more informed decisions as well as be alerted when an arbitrage betting opportunity arises. Currently, the script is designed for NBA games only. Please also note that due to region restrictions, this tool is designed solely for use in Australia.
Arbitrage betting is a betting strategy where you place bets on all possible outcomes of an event with different bookmakers, taking advantage of differing odds to guarantee a profit regardless of the event's result.
We can take advantage of the differences in odds offered by various bookmakers to secure a guaranteed profit. This strategy, known as arbitrage betting, is based on the principle that bookmakers set odds to reflect the expected probabilities of the outcomes of an event. For example, consider a basketball game between the NY Knicks and Orlando Magic, where Sportsbet offers odds of 2.00 for the Knicks and 1.85 for the Magic. These odds imply that Sportsbet estimates a 50% chance of the Knicks winning (calculated as 1 divided by the odds, or 1/2.00) and a 54.05% chance for the Magic (1/1.85).
You'll notice that the combined probability exceeds 100% (50% + 54.05% = 104.05%), which is typical in betting markets. This essentially ensures that the bookmaker profits regardless of the outcome if bets are evenly distributed. However, arbitrage opportunities arise when the combined implied probabilities of all possible outcomes are less than 100%. For instance, if another bookmaker offers odds of 2.10 for the Magic, the combined probability becomes 1/2.00 + 1/2.10, which equals 97.62%. This under 100% total indicates an arbitrage opportunity, with an ROI of 2.38% (calculated by 100% - combined probability).
We also need to calculate how much to bet on each outcome to ensure we make the same return regardless of the game's actual outcome. To determine the bet sizing, we take the expected return and divide through by the odds offered on each outcome. In this example, let's suppose I'm betting with $100. Since I know my ROI is 2.38%, I can expect a return of $102.38. This means that I'll bet $51.19 on the Knicks and $38.75 on the Magic to ensure a profit is made. These bets were calculated by taking my expected return of $102.38, and dividing it by each team's odds to find the bet sizing. In this case, $102.38/2.00 = $51.19 and $102.38/2.10 = $38.75 giving me the amount to bet on the Knicks and Magic respectively.
The email alert includes the following information:
A sample email alert can be seen here: https://imgur.com/a/6oP44W7
.env File.env FileTo securely handle your email credentials for alert notifications:
.env file in your project root directory.EMAIL_ADDRESS=your_email@example.com
EMAIL_PASSWORD=your_password.env is listed in your .gitignore to keep your credentials secure.Use the dotenv package to load and access these variables in your Python scripts:
from dotenv import load_dotenv
import os
load_dotenv()
email = os.getenv('EMAIL_ADDRESS')
password = os.getenv('EMAIL_PASSWORD')This is done in src/email_alert.py so adding to the .env file is sufficient.
This is a common issue even if you've inputted the correct email and password. Instead, create an app specific password (requires you to have 2FA turned on) and instead use the app specific password you've created. This will fix this issue
Similarly, use the .env file to set the recipient's email. Add to the file
your recipient email in the format:
RECIPIENT_EMAIL=your_email@example.comThis will automatically change your recipient email in the script.
This tool is provided for educational and informational purposes only. It is important to note that gambling involves significant financial risk and may not be suitable for everyone. I do not promote or endorse gambling. Users are advised to exercise caution and bet within their means. Additionally, the accuracy of the betting odds data extracted by this tool cannot be guaranteed, as bookmakers frequently update odds. Users are responsible for verifying the odds with the bookmaker before placing any bets. This tool is intended for personal use, and the user must ensure compliance with all local laws and regulations regarding online betting in their jurisdiction.
pip install selenium, pip install tabulate and pip install python-dotenvgit clone git@github.com:mxl1503/arbitrage-betting-bot.git.To run the scraper:
python3 src/main.pyCurrently the script uses the following bookmakers to find arbitrage opportunities:
More to be added soon!
Project inspired by the following sources: