Arbitrage betting backs a selection at a high bookmaker price and lays it lower on the Betfair Exchange, locking profit either way. After 2% commission you need roughly a 3% back-over-lay gap to clear a real edge. It's legal, the Exchange allows it, but bookmakers restrict arbers fast.
This page contains affiliate links — if you open an account through them we may earn a commission at no cost to you. It never changes our verdict.
Arbitrage betting is the closest thing to a free lunch in betting — and the Betfair Exchange is what puts it on the table. Because the Exchange gives you a lay price that's sharper than any bookmaker, you can back a selection at a generous book price, lay it on Betfair, and pocket the difference no matter which way the event goes. This guide sits under our arbitrage and value pillar and walks the whole thing end to end: what an arb is, the exact lay-stake maths, a real worked example from my own account, the risks the YouTube videos skip, and whether arbing is actually worth doing in 2026.
Quick distinction up front, because people conflate them: an arb exploits naturally mispriced odds; matched betting exploits a bookmaker bonus. The mechanics are identical — back at the book, lay on the Exchange — but the source of the edge differs.
What an Arb Actually Is
An arbitrage bet (an "arb" or "surebet") is a set of bets across two or more prices that locks in profit no matter which outcome wins. It exists because different operators disagree about the price of the same event. When a bookmaker's back price is higher than the Betfair Exchange lay price for the same selection, you can back at the book, lay on the Exchange, and bank the gap. No prediction required — the edge is in the price discrepancy, not in being right.
Betfair sits at the centre of nearly every arb because the Exchange is the sharpest, most liquid price in the market. It is both the place you lay off bookmaker bets and, occasionally, one half of an exchange-to-exchange arb against a rival exchange. If you are still learning the back/lay mechanics, start with the arbitrage and value pillar and the Exchange beginner's guide before chasing live arbs.
The Three Arb Types You Will Actually See
In practice, almost every arb a UK or Irish bettor finds falls into one of three buckets, and Betfair is involved in two of them.
Back-bookmaker / lay-exchange
The workhorse. Back a selection at a generous bookmaker price, lay the same selection on the Betfair Exchange at a lower price. The gap between the two, after commission, is your locked profit. This is structurally identical to matched betting — the difference is that an arb uses naturally mispriced odds rather than a bonus.
Exchange-to-exchange
Rare but real: back on one exchange and lay on another when their prices drift apart. Liquidity on the smaller exchange is the limiting factor, and the windows are short. Most traders ignore these because the size you can get on is tiny.
Cross-bookmaker (no exchange)
Back the home win at one book and the away/draw at others so the combined prices guarantee profit. Betfair is not needed here, but the Exchange is usually the cleanest place to complete or hedge the position, and the same overround maths applies.
The Maths, Done Once Properly
The whole thing comes down to one comparison: is the back price bigger than the lay price by more than commission eats? Here is the lay-stake formula every arber uses to balance a back-bookmaker / lay-exchange position:
Lay stake = (back odds × back stake) ÷ (lay odds − commission)
Work an example. You back £100 at 4.0 with a bookmaker (potential return £400). The Exchange lay price is 3.9 and your commission is 2%. Lay stake = (4.0 × 100) ÷ (3.9 − 0.02) = 400 ÷ 3.88 = £103.09. Your lay liability is 103.09 × (3.9 − 1) = £299. Whichever way the event goes, you end up a few pounds ahead. The bigger the back-over-lay gap, the bigger the locked profit. Below roughly a 2–3% gap, commission and slippage swallow the edge entirely.
| Back price (book) | Lay price (exchange) | Approx. arb % after 2% comm. | Worth it? |
|---|---|---|---|
| 4.00 | 3.90 | ~1.8% | Marginal |
| 4.00 | 3.80 | ~3.4% | Yes |
| 6.00 | 5.60 | ~4.6% | Yes, if liquidity holds |
| 2.10 | 2.08 | ~0.6% | No — commission kills it |
Finding Arbs
Arbs appear when a bookmaker is slow to move a price the Exchange has already corrected — after team news, a big in-play swing, or a price boost. You find them three ways: an oddsmatcher / arb-scanner that compares book prices to the Exchange in real time, manual monitoring of a handful of markets you know well, or by working bookmaker promotions (the most reliable source for beginners). The scanner route is fastest but crowded; the manual route is slower but the arbs last longer because fewer people see them. Software options are compared in value betting software for Betfair.
Speed matters more than people expect. A genuine arb on a liquid market is usually gone within seconds to a minute, because the bookmaker corrects or the Exchange price moves to meet it. By the time a free scanner shows you a 4% arb, the back price has often already shortened. This is why the reliable, repeatable version of arbing for most people is offer-driven, not scanner-driven.
From the Desk: A Real Back-Lay Arb
This one came from a price boost, which is where I find most of my clean arbs.
Market: Premier League, a midfielder to score anytime, boosted by a bookmaker from 4.5 to 6.0.
Back: £40 at 6.0 with the book (potential return £240).
Lay on Betfair Exchange: price available 5.4, plenty matched. Lay stake = (6.0 × 40) ÷ (5.4 − 0.02) = 240 ÷ 5.38 = £44.61; liability £196.28.
Outcome both ways: if he scores, the book pays £200 profit and the lay loses £196.28 → net +£3.72. If he doesn't, I lose the £40 back stake and keep the lay winnings of £44.61 less 2% commission (£43.72) → net +£3.72.
Locked profit: £3.72 on £40, win or lose. Small, but risk-free, and it took ninety seconds. Stack twenty of these a week and it adds up without ever needing to predict a result.
Notice the profit is identical on both outcomes — that is the signature of a true arb. If your two-way numbers don't match, you have a value bet or a mistake, not an arb. The distinction matters and is covered in finding value bets on the Betfair Exchange.
The Risks Nobody Advertises
Arbing is called "risk-free" but the risk is operational, not mathematical. The real dangers:
- Price movement between legs. If the Exchange price moves after you've placed the bookmaker bet but before you lay, your locked profit can turn into a small loss. Always place the leg most likely to move first.
- Bookmaker restrictions. Arbers get gubbed faster than anyone. Consistent arbing flags your account and your stakes get cut or your bonuses pulled. The Exchange never restricts you — the bookmaker side is the fragile half.
- Bet cancellation / palpable error. Books void obviously mispriced bets, leaving you with an unhedged lay liability on the Exchange.
- Liquidity gaps. If the Exchange lay side is thin, you may only get part of your lay matched, leaving you partially exposed.
"Risk-free" describes the maths, not the practice. Execution errors, restrictions and voided bets are the ways arbers actually lose money. Most casual arbers make small, slow profits; a minority get gubbed and quit. Treat it as a grind, not a money machine.
Is It Allowed?
Arbing is legal everywhere it is legal to bet, and the Betfair Exchange has no problem with it because the Exchange profits from commission either way. Bookmakers dislike it and respond with restrictions, but that is a commercial decision, not a legal one. The full picture — what Betfair permits, what gets you gubbed, and where matched betting sits — is in the sibling is arbitrage betting legal on Betfair?
Is It Worth It?
For most people, pure scanner arbing is a poor use of time: the arbs are small, fleeting, and crowded, and your bookmaker accounts degrade quickly. The version that actually works long-term is offer-driven — using bookmaker promotions and price boosts as your arb source, with the Betfair Exchange as your lay engine. That is steadier, the edges are bigger, and it dovetails with matched betting. If your real goal is a repeatable edge, learning to read and trade the Exchange directly — covered across our case studies and value-finding pages — scales far better than chasing two-second surebets.
Arbing is a fine way to learn back/lay mechanics with low variance. Master the lay-stake maths first, start with offer-driven arbs, and keep your stakes small while your accounts are young.
Arbitrage & Value Pillar Open Betfair Account →Arbing vs Matched Betting vs Value Betting
These three get muddled constantly, yet they sit on a clear spectrum of certainty and ceiling. Knowing which you're doing tells you how much you can stake and how long the edge lasts.
| Approach | Source of edge | Per-bet outcome | Long-term ceiling |
|---|---|---|---|
| Matched betting | Bookmaker bonus/free bet | Locked profit | Limited by offers available |
| Arbitrage | Price discrepancy | Locked profit (small) | Limited by account life + scanner crowding |
| Value betting | Mispriced true probability | Variable — wins and losses | High, but needs an edge and a stomach for variance |
Matched betting and arbing both give you a known result on every bet; value betting does not. The trade-off is that value betting scales far higher because it isn't capped by the supply of offers or the lifespan of your bookmaker accounts. Most people who stick with the Exchange start with matched betting, dabble in arbs, and gradually move toward value and trading as their reading of the market improves. The value-finding method is the natural next step once arbs feel mechanical.
Bankroll and Staking for Arbs
Arbing is liability-heavy, not stake-heavy. The number that constrains you is your Exchange balance, because the lay liability ties up funds until the market settles. On the £40 example earlier, the lay liability was nearly £200 — so a £40 back position actually needs £200 sitting idle on the Exchange. Run several arbs at once and your Exchange bank is the bottleneck, not your bookmaker funds.
Practical sizing for a beginner: keep individual back stakes at £10–£50 while your bookmaker accounts are new, and hold at least 5× your typical back stake as free Exchange liability. Scale up only once you've executed fifty arbs cleanly without a balancing error. The discipline is identical to the staking rules in our trading material — never deploy money you might need to balance another position.
The Execution Mistakes That Cost Real Money
Almost every losing arb I've seen traces back to one of these, not to the maths:
- Laying first when the bookmaker leg is the volatile one. Place the leg most likely to move first — usually the bookmaker boost — then lay into the stable Exchange price.
- Forgetting commission in the lay-stake formula. Skip it and you'll under-lay, leaving a small unhedged exposure on every arb.
- Chasing a 0.5% arb. The edge is below your slippage and commission. You are working for free and risking an unbalanced position for nothing.
- Partial matches on thin markets. If only half your lay matches, you're half-exposed. Check available liquidity before you commit the back bet.
- Rounding stakes by eye. Use a calculator. The whole point of an arb is precision; a 50p rounding error compounds across hundreds of bets.
Run your numbers through the site's free Betfair calculator until the lay-stake maths is second nature. The calculator removes the single most common source of arbing losses: arithmetic done in a hurry while a price is moving.
Tools That Help (and the One That Doesn't)
You need three things: a price source (a scanner or oddsmatcher), a calculator, and fast Exchange execution. Dedicated trading software like Bet Angel or Geeks Toy gives you one-click laying that beats the slow native Betfair grid, which matters when an arb is closing. The tool that doesn't help is any product promising "guaranteed daily arb profits" for a subscription — the genuinely repeatable edge is offer-driven, and the scanner-only arbs are too crowded to subscribe your way to wealth. A grounded comparison of paid value tools is in value betting software for Betfair.
What a Realistic Week Looks Like
Forget the screenshots of four-figure days. For a part-timer working offer-driven arbs and the odd scanner opportunity around a busy Saturday, a realistic week is a string of small, certain returns rather than one big hit. The point of arbing is consistency and low variance, not size.
| Activity | Typical count/week | Avg locked profit each | Weekly total |
|---|---|---|---|
| Price-boost arbs | 8–15 | £3–£6 | ~£40–£75 |
| Reload-offer arbs | 3–6 | £8–£20 | ~£35–£90 |
| Scanner surebets | 2–8 | £1–£4 | ~£10–£25 |
That lands most disciplined part-timers somewhere around £80–£180 a week before their accounts start getting restricted — which they will. The honest arc of an arbing career is: strong early returns while your bookmaker accounts are healthy, then a tapering as restrictions bite, at which point the smart players have already migrated to Exchange trading where there are no restrictions and no ceiling. For the longer view of where that road leads, the income reality is mapped in realistic Betfair trading income, and real account journeys are documented in the case studies.
One last principle worth internalising: an arb is a discipline, not a windfall. The traders who do well with it are methodical — they place the volatile leg first, factor commission every time, size to their Exchange liability rather than their ego, and walk away the moment a price gap closes below their threshold. Treat it as paid practice in back-lay execution and bankroll control, and it becomes the cleanest on-ramp to the wider skill set the rest of this site is built around. Master the boring version first; the money follows the discipline, not the other way round.
FAQ
Is arbitrage betting really risk-free? The maths is risk-free, but the practice is not. Price movement between legs, bookmaker bet cancellations, account restrictions and thin Exchange liquidity are real ways arbers lose money. Call it low-risk, not no-risk.
Does Betfair allow arbitrage betting? Yes. The Exchange earns commission whichever way your bet lands, so it has no reason to stop arbers. Bookmakers are the ones that restrict arbing accounts, not Betfair.
How much commission do I pay on the lay side? Standard Betfair Exchange commission, typically 2 to 5 percent on net market winnings depending on your market base rate. You must factor it into the lay-stake formula or your arb maths will be wrong.
How big does an arb need to be to be worth it? After commission and slippage, aim for at least a 3 percent back-over-lay gap. Below 2 percent the edge is usually eaten by price movement and commission before you complete both legs.
Will I get my bookmaker account restricted for arbing? Very likely over time. Arbers are gubbed faster than almost any other customer type. The Exchange side is safe; the bookmaker side is the fragile half of every arb.
Is arbing better than learning to trade the Exchange? For a long-term, scalable edge, learning to trade the Exchange usually wins. Arbing is excellent for learning back/lay mechanics with low variance, but the arbs are small, fleeting and crowded.