What is arbitrage?
When two bookmakers disagree enough about the same game, every outcome can be backed at a profit. Here is how that happens, and why it is rarer and touchier than it sounds.
An arbitrage, or arb, is a set of bets that covers every possible result of a market at prices good enough that you finish in front no matter what happens. Nobody has to predict anything. You are not betting on the game at all, you are betting that two bookmakers have priced it inconsistently.
Why a market usually cannot be arbed
Take a two way market where both teams are genuinely even. Fair odds would be 2.00 each, because each side has a 50% chance. No bookmaker offers that. They offer something like 1.90 and 1.90, and the missing bit is their margin.
The quick way to see it is to turn each price into a percentage by dividing 1 by the odds. At 1.90 and 1.90 you get 52.6% and 52.6%, adding to 105.2%. That extra 5.2% is the overround, the built in edge that makes bookmaking a business. Back both sides at one book and you lose about 5% of your money every time, reliably.
Where the gap comes from
You are never arbing one bookmaker against itself. You are taking the best price on each side from different books, and occasionally the two best prices together add to less than 100%.
That happens because bookmakers are not one market. One book has taken a pile of money on the home team and shortens it to balance. Another is running a promotion on the away team. A third has not updated since a team news announcement. For a few minutes the top of each side sits at a different book, and the sum drops below the line.
A realistic example
Sportsbet has the home side at 2.05. TAB has the away side at 2.00. Converted to percentages that is 48.78% and 50.00%, which adds to 98.78%. Under 100%, so there is 1.23% in it.
Split $100 in proportion to those percentages and you get $49.38 on the home side and $50.62 on the away side.
| Sportsbet returns $49.38 × 2.05 | +$101.23 |
| TAB bet loses | $0.00 |
| Profit on $100 staked | +$1.23 |
| TAB returns $50.62 × 2.00 | +$101.24 |
| Sportsbet bet loses | $0.00 |
| Profit on $100 staked | +$1.24 |
About $1.23 either way. That is what a real arb looks like. Not the 8% screenshots that circulate on social media, which are almost always a palpable error that gets voided, a stale price that no longer exists, or two different lines being compared as though they were the same market.
The profit only exists if both bets get on at the prices you saw. Get the first leg matched and miss the second and you are not arbing, you are holding a straight bet on a football game with money you never intended to gamble. Place the leg most likely to move first, and have the second window already open and filled in.
How this site surfaces arbs
Every scan you run through the Bonus Basher looks at the same odds twice. Once for bonus conversions, and once for cross book arbitrage in the markets it just downloaded. If it finds any, a capsule appears in the top bar next to the logo.
The capsule reads something like Arb +1.23% - Sydney vs Melbourne with the legs underneath expressed per $100, so $49.38 on Sydney at 2.05 (Sportsbet) tells you the split without you working it out. If there is more than one, faint cards stack behind the front one and clicking cycles through them. The banner sticks around for fifteen minutes as you move between pages, then clears itself so old prices cannot mislead you.
Turn arb detection on
- Arb scanning is a Tracker feature, so the Tracker has to be unlocked in this browser first. Without it the scanners skip arbitrage entirely.
- In Settings, set Min arbitrage profit. The default of 2% is a sensible starting point. Drop it to 1% and you will see far more, most of which will be gone before you get there.
- Run a scan from the Bonus Basher. Anything at or above your threshold raises the capsule.
- Turn on Sound or Desktop notification in Settings if you want to hear about it while working in another tab. Quiet hours and a repeat cooldown stop the same opportunity nagging you.
- Your Dashboard keeps a card of the arbs from your most recent scan, filtered to your threshold, so you can look again without burning another scan.
The engine throws away anything under 0.3% before it gets to you. At that size the number is usually just the two prices drifting apart for a moment rather than a real gap, and chasing it costs more in mistakes than it pays.
The Low Loss scanner shows the same thing from the other direction. It ranks matches by how much it costs to hedge a cash bet, and when that cost goes below zero the row is an arbitrage. It is the better tool when your goal is moving a specific balance rather than hunting for profit.
What actually goes wrong
- The price disappears while you are typing. The most common outcome by a distance. Prices move constantly, and a scan is a photograph, not a live feed. Check both prices in the bookmakers' own windows before confirming.
- Palpable errors get voided. If a price was obviously wrong, the bookmaker will void that bet under their terms. Your other leg stays live, and now you have an unhedged bet.
- You get a partial fill. Exchanges only match what somebody else is offering. Half a lay leg is half a hedge.
- The two legs are not the same market. Over 45.5 points and Under 46.5 points look similar and are not opposites. Anything landing between them loses both. The scanners group by exact line for this reason, but if you are eyeballing prices yourself it is an easy mistake.
- Commission is forgotten. A 1.2% edge does not survive 6% commission on a winning lay leg. Any calculation involving an exchange has to include it, which is why the commission field exists on both scanners.
- Stake limits. A book that will happily take $500 on the favourite may cap you at $20 on the side you need, and a $20 leg against a $500 leg is not a hedge.
Bookmakers watch for it, and they are good at it. Repeatedly taking the top price seconds after it appears, in odd markets, at odd stake sizes, is a recognisable pattern and it does not take many before the maximum bet on your account quietly drops to a few dollars. A promo account that lasts a year is usually worth more than a handful of 1% arbs. If you plan to do this regularly, read the account protection page first.
Is it worth doing?
Honestly, it depends on what you are after. As a way of turning a large bankroll into a steady wage it is hard work, because the margins are thin, the opportunities are brief, and the accounts wear out. As a way to occasionally squeeze a bit more out of accounts you are already using for promos, it is a nice extra.
Most people using this site find the arb capsule pays for itself simply because it appears during scans they were already running. You are not going out of your way. When one shows up at a size worth taking, it is there.
Whatever you take, log it. Paste the legs into the Tracker as two rows and mark the result when the game settles, so your realised profit reflects what happened rather than what the scan promised.
Where to go next
Everything on this page is general information about how these tools work, not betting advice, and no return is ever guaranteed. Bookmakers do not offer promotions to lose money on them. If betting has stopped being fun, free and confidential help is at Gambling Help Online or 1800 858 858, and you can block yourself from every Australian online wagering site through BetStop. 18+ only.