Arbitrage

Arbitrage (or arbing) is when you bet both sides of a market across different sportsbooks at prices that guarantee profit no matter which side wins. It’s the closest thing to a free lunch in sports betting — but the books actively fight it, and the limits are tight.

How arbitrage works

Sportsbooks set lines independently. Sometimes Book A’s price on Side X plus Book B’s price on Side Y sum to less than 100% implied probability. When that happens, you can size bets at both books such that you profit either way.

Quick example

Book A: Side X +120 (decimal 2.20, implied 45.5%) Book B: Side Y +110 (decimal 2.10, implied 47.6%)

Sum of implied: 45.5% + 47.6% = 93.1% — that’s less than 100%, so an arbitrage exists.

To split a $1,000 stake:

• Bet on Side X: $1,000 / 2.20 / 0.931 ≈ $488 at Book A
• Bet on Side Y: $1,000 / 2.10 / 0.931 ≈ $512 at Book B

If Side X wins: $488 × 2.20 = $1,073.60. Net = $1,073.60 − $1,000 = +$73.60 If Side Y wins: $512 × 2.10 = $1,075.20. Net = $1,075.20 − $1,000 = +$75.20

Either way, profit. That’s a 7.4% arbitrage — very unusual; most arbs are 1-3%.

How arbs appear

Arbs exist because:

• Different books use different models — they price independently.
• Liquidity differences — small books are slower to update lines.
• Promo / boosted markets — books occasionally inflate one side as a marketing play.
• News/injury lag — one book updates faster than another.

Most arbs are 1-2% windows that close in minutes once both books see the action. Arbitrage hunters use specialized tools that scan all major books in real-time.

Why books hate arbers

Arbers are unprofitable customers. They generate volume but never lose to the vig — they’re effectively earning the gap between two books’ models. Books track betting patterns and limit or close accounts that consistently grab the best price across markets.

This is why pro arbers maintain dozens of accounts (often using “beards” — proxy accounts in other people’s names), bet small to stay under the radar, and rotate books constantly. It’s a real cat-and-mouse game.

Retail arbitrage

You don’t have to be a pro arber to occasionally catch one. Retail bettors with 3-4 books can occasionally spot a 1-3% arb — usually around major news (injury announcements, weather), promos, or live betting (where book models lag the action).

Tools that help:

• The Vig Remover — quickly checks if two prices on opposite sides imply less than 100% summed probability
• Line-shopping browser extensions
• Real-time arb alerting services (paid)

Risks

• Limits — books cap your bets, so you can’t always get the size you want on both sides.
• Voided bets — one side gets graded as a push or void, leaving you with naked exposure on the other side.
• Account closures — books close arber accounts.
• Slippage — line moves after you bet one side but before you bet the other.

The “free money” pitch glosses over these. Real arbing is operationally heavy and the win rate (capturing arbs successfully) is lower than the headline number suggests.

Related: middling

A middle is a different concept — betting both sides of a market at different point spreads to potentially win both bets if the final score lands between the two spreads. See middling for the math.