Find sports arbitrage opportunities across books and markets

A sports arbitrage scanner finds games where two venues disagree on price by enough that you could bet both sides and come out ahead whichever side wins. It does that by converting every price to an implied probability and flagging any pair whose probabilities add up to less than 100 percent. The quant+ arbitrage scanner runs that check across sportsbooks and prediction markets on one board.

Arbitrage is not the same thing as no risk. Prices move while you're placing the second bet, books cancel or limit, and fees eat thin margins. This page covers the math and every one of those catches.

How sports arbitrage works

Each side of a bet has an implied probability. On a normal market the two sides add up to more than 100 percent; that excess is the book's margin. When you take the best price for each side from two different venues, the total can drop below 100. The gap is the arbitrage margin.

The check is one line:

1 / decimal odds (side A) + 1 / decimal odds (side B) < 1 means an arbitrage exists

A worked example with the stake split

Book A has Team X at +110. Book B has Team Y at -105. Here is the arithmetic (an illustration, not a result).

	Team X at Book A	Team Y at Book B
American odds	+110	-105
Decimal odds	2.100	1.952
Implied probability	47.62%	51.22%
Sum	98.84%
Arbitrage margin	1.16%
Stake split of 100	48.18	51.82
Total returned if this side wins	101.18	101.18

Each stake equals the total times that side's implied probability, divided by the sum. Bet 48.18 on X and 51.82 on Y, and either result returns about 101.18 on 100 staked. That's a 1.16 percent margin, before anything goes wrong.

Why arbitrage windows exist at all

If markets were perfectly in sync, they wouldn't. They aren't, because sportsbooks don't price like an exchange. Economist Steven Levitt found that books announce a price and then adjust it in small, infrequent steps. When one book moves after news and another hasn't yet, the two prices can straddle each other for a short time.

The other source is venues that price differently. Prediction markets quote contracts in cents and set prices by trading, while books set prices by their own traders. They don't always agree, which is why the quant+ scanner puts both on the same board.

The catches, in order of how often they bite

1. The price moves before you get the second bet down. Arbitrage windows are often short. If the second book moves first, you're left holding one side at a price you only took because of the other. Place the harder bet (smaller limit, faster-moving book) first.

2. Fees on prediction markets. Polymarket charges takers a fee on sports markets at a rate of 0.05 that peaks around a 50 percent price, and makers pay none. Kalshi charges a fee on the expected earnings of each contract. A margin of about 1 percent can turn negative after fees, so the scanner's numbers should always be read with the venue's fees in mind.

3. Account limits. Books watch for arbitrage patterns. In one documented case, researchers betting a market-based strategy saw bookmakers restrict their stakes within a few months, to as little as $1.25 on a $50 bet, and some bets were held for manual review. Betting both sides of the same game at different books is exactly the kind of pattern books look for.

4. Rules that don't match. Two venues can settle the "same" bet differently: overtime included or not, a player prop voided if the player doesn't start, a game moved to another day. Read both venues' rules before treating a pair as a true arbitrage.

5. Voided or palpable-error bets. If one book decides its price was an obvious error and voids your bet, the other side is now a plain wager.

How to read the quant+ arbitrage scanner

Each row shows the game, the market, the two venues and their prices, and the arbitrage margin. Work the stake split with the formula above before you bet. Next to it, you can open:

• Line movement, to see which side moved most recently. That's usually the side that will close the window.
• The odds board, to check every other venue's price on the same market. If a third book agrees with one side, the outlier is the one to act on first.
• The fair price, built from sharp-book prices with the margin removed. When one leg of an arbitrage is also +EV against the fair price, some bettors prefer to take just that leg. That's a judgement call, and it's no longer arbitrage.

The fair price uses a margin-removal method; a 2017 study in the American Journal of Sports Science found the power method outperforms simple normalization, which matters most on lopsided lines.

What the scanner won't do

quant+ never places bets, holds funds or connects to your sportsbook accounts. It shows the pair and the math. Speed of execution, account health and reading the rules are on you. Keep a record in the bet journal so you can see what share of flagged arbitrages you actually got down at the quoted prices.

Check that sports betting, and prediction markets if you use them, are legal where you are. Read the risk disclosure before you start.

Start small

Run the scanner for a week without betting. Note how long each window stays open and how often the second price moves before you could have acted. That tells you whether arbitrage suits your books and your speed.

Start free to open the scanner, see the board on the home page, or read the quant+ guides first.

Quant+ is analytics for informational purposes. 21+. If you or someone you know has a gambling problem, call 1-800-GAMBLER.