sports-betting
sports-betting is a set of tools for building, testing and using sports betting models. You can use it from Python, from the command line, or through an AI agent.
There are two main parts. A dataloader downloads the data and shapes it for modelling. You build it from a statistics source and an odds source, so you always know where your data came from:
from sportsbet.dataloaders import DataLoader
from sportsbet.sources import FootballDataOdds, FootballDataStats
dataloader = DataLoader(
param_grid={'league': ['England'], 'division': [1], 'year': [2023, 2024]},
stats=FootballDataStats(),
odds=FootballDataOdds(),
)
X, Y, O = dataloader.extract_train_data(odds_type='market_maximum')
print(X.shape)(760, 29)That’s two seasons of the English top division: 760 matches, each with 29 columns.
The second part is the bettor. It wraps any scikit-learn estimator, backtests a strategy on the data, and predicts the value bets of upcoming matches.
The library can also place a bet for you. Be careful with this, because it spends real money. Some bookmakers have an API, and for the rest an agent drives their website on your account, which breaches most bookmakers’ terms of service and puts your account at risk.
Install it with pip install sports-betting.