Examples · Python

Build a Sports Arbitrage Finder in Python

Arbitrage happens when two books disagree enough that betting both sides guarantees a profit. If one book pays -114 on a player going over 56.5 rushing yards and another pays +127 on the under, splitting $500 the right way returns about $514 whatever happens.

This tutorial builds a Python script that lists current arbitrage opportunities and tells you exactly how much to put on each side.

Cost: 1 credit per run. The free plan includes 1,000 credits a month. Full project on GitHub.

1. Get arbitrage opportunities

One call, ml.arbitrage(), returns every opportunity the API has found, with the profit percent and the stake for each leg on a $1,000 total.

The API checks that the two legs really are opposite sides of the same bet. DFS apps are left out, because their prices aren't bets you can place at those odds, and exchange prices only count when they're close to the sportsbook consensus.

2. Size the stakes for your bankroll

The API sizes stakes for $1,000. The stakes_for function scales each leg to the bankroll you pass with --bankroll, so both sides return the same amount.

3. Place both bets fast

Arbitrage windows often close within minutes, and books can limit accounts that bet this way. Confirm both prices at the books before placing either bet, and place the harder-to-get leg first.

The code

arbitrage-finder/arbitrage_finder.py, straight from the repository.

arbitrage_finder.py
"""Find sports betting arbitrage (guaranteed-profit) opportunities with MoneyLine Sports API.

Usage:
    export MONEYLINE_API_KEY=your-key
    python arbitrage_finder.py --bankroll 500 --min-profit 1
"""
import argparse

from moneyline_sports_api import MoneyLine


def fmt_odds(price):
    return f"+{price}" if price > 0 else str(price)


def find_arbs(ml, league=None, min_profit=0.5):
    """Return arbitrage opportunities at or above min_profit percent, best first."""
    arbs = ml.arbitrage(league=league, minProfit=min_profit, limit=50)  # 1 credit
    return sorted(arbs, key=lambda a: a["arbitrage"]["profitPct"], reverse=True)


def stakes_for(arb, bankroll):
    """Scale the API's per-$1,000 stakes to your bankroll."""
    scale = bankroll / arb["arbitrage"]["totalStake"]
    return [(leg, round(leg["stake"] * scale, 2)) for leg in arb["arbitrage"]["books"]]


def main():
    parser = argparse.ArgumentParser(description="Find arbitrage opportunities across US sportsbooks.")
    parser.add_argument("--league", help="nfl, nba, mlb, nhl, ncaa_football, ... (default: all)")
    parser.add_argument("--min-profit", type=float, default=0.5, help="minimum guaranteed profit in percent (default: 0.5)")
    parser.add_argument("--bankroll", type=float, default=100, help="total amount to split across both bets (default: 100)")
    args = parser.parse_args()

    ml = MoneyLine()  # reads MONEYLINE_API_KEY
    arbs = find_arbs(ml, args.league, args.min_profit)
    if not arbs:
        print(f"No arbitrage at {args.min_profit}% or better right now. Prices move fast; try again soon.")
        return

    for arb in arbs:
        profit = arb["arbitrage"]["profitPct"]
        print(f"\n{profit:.2f}% profit  {arb['leagueId']}  {arb['market']}  {arb['outcome']}")
        for leg, stake in stakes_for(arb, args.bankroll):
            print(f"    bet ${stake:>8.2f} on {leg['selection']:<40} at {fmt_odds(leg['odds']):>6}  {leg['bookmaker']}")
        print(f"    returns about ${args.bankroll * (1 + profit / 100):.2f} whichever side wins")
    print("\nCheck each price at the book before betting: arbitrage windows often close within minutes.")


if __name__ == "__main__":
    main()

Run it

Get a free API key, clone the example, then run:

Terminal
pip install moneyline-sports-api
export MONEYLINE_API_KEY=your-key
python arbitrage_finder.py --bankroll 500 --min-profit 1

More examples