A bookmaker builds its margin into the price. An exchange charges it separately, as commission, and shows you prices set by other bettors. The prices look better than a bookmaker's because they are — before commission. This guide explains the two commission models, gives the formulas for effective odds and break-even, and shows how to compare an exchange price honestly with a bookmaker's.
Two models
Commission on net winnings is the common model. When a bet wins, the exchange takes a percentage of the profit; when it loses, nothing extra. Betfair, Smarkets and Betdaq use it, at rates that vary by venue, country and activity tier. The rates used on this site are placeholders; check your own account.
Commission on turnover is charged on the stake whether the bet wins or loses. It is rarer, and it behaves differently: it makes both sides of the bet worse rather than only the winning side.
The distinction matters because a net-winnings charge can be folded into a single "effective price", and a turnover charge cannot.
Effective odds
For commission on net winnings at rate c, the profit on a winning bet is multiplied by (1 − c). The effective odds are the bookmaker price that would pay the same.
effective = 1 + (odds − 1) × (1 − c) 2.10 at 5%: 1 + 1.10 × 0.95 = 2.045 2.10 at 2%: 1 + 1.10 ×
0.98 = 2.078 5.00 at 5%: 1 + 4.00 × 0.95 = 4.80
Notice how the cost scales with the price. At 2.10 a 5% commission costs 0.055 of a decimal point; at 5.00 it costs 0.20. The longer the price, the bigger the profit, and the bigger the slice the exchange takes. Backers of favourites feel commission much less than backers of outsiders. The commission calculator shows a whole table of rates for any price.
Break-even
The break-even win rate is the implied probability of the effective odds.
break-even = 1 / effective 2.10 at 5%: 1 / 2.045 = 48.90% 2.10 at 0%: 1 / 2.10 = 47.62%
Commission raised the break-even rate from 47.62% to 48.90%: you need to win 1.3 percentage points more often to stand still. Put another way, a bookmaker offering 2.05 or better on the same outcome beats the exchange at 2.10 with 5% commission.
For turnover commission at rate c, the win profit is (odds − 1 − c) per unit and the loss is (1 + c) per unit, and the break-even rate is the loss divided by the sum:
break-even = (1 + c) / odds 2.10 at 2% turnover: 1.02 / 2.10 = 48.57%
Lay bets
Laying is the exchange's distinctive feature: you take the bookmaker's side of a bet. When you lay at odds L against a backer's stake B, you risk a liability of B × (L − 1) and win B if the outcome loses. Commission is charged on your winnings, which are the backer's stake.
Laying 100 at 2.10 with 5% commission
Liability: 100 × 1.10 = 110. If the outcome loses, you win 100 minus 5% commission = 95. If it wins, you pay 110. Break-even (the probability of the outcome at which you neither gain nor lose): 95 / (95 + 110) = 46.3%. Equivalently, laying at 2.10 is backing "not this outcome" at 2.10/1.10 = 1.909, which after 5% commission is 1.864; 1/1.864 = 53.7% is the break-even for "not this outcome", and 100% − 53.7% = 46.3% for the outcome itself. The
commission calculator in lay mode
shows liability, profit and break-even together.
The lay betting guide goes deeper into the conversion between lay odds and back odds, which is what makes bookmaker-versus-exchange hedges solvable.
Comparing an exchange with a bookmaker
The honest comparison is: effective exchange price versus bookmaker price. Three cases:
- Backing. Exchange at 2.10 with 5% commission is 2.045 effective. A bookmaker at 2.05 or more is better; at 2.00 it is worse.
- Laying to hedge a bookmaker bet. Convert the lay to its equivalent back price on the other outcome, apply commission, and run the arbitrage test. The arbitrage guide has the worked example.
- Reading fair odds. Exchange prices, after the small overround between back and lay, are close to the fair line. But if you are going to bet on the exchange, the fair line you should compare against is the one after commission.
Commission and arbitrage
Commission raises every implied probability on the exchange side, so it narrows arbitrages. A two-way market at 2.10 and 2.10 across two exchanges with 5% commission on both is 2.045 / 2.045 effective, which sums to 97.8%: still an arbitrage, at 2.25% instead of 5%. At 10% commission each side the effective prices are 1.99 and the arbitrage is gone. The arbitrage calculator applies commission per leg, so you can mix a bookmaker leg with an exchange leg or two exchanges at different rates.
Commission per market, not per bet
Most exchanges charge commission on your net winnings across a market, not on each bet separately. If you back and lay the same outcome in the same market and end up with a small net profit, commission applies to that net profit only. This is friendlier than the per-bet model the calculators assume, so a real position can be slightly better than the calculator shows when several bets share a market. The difference is small unless you trade in and out repeatedly.
Discounts, tiers and promotions
Base rates are not what most regular users pay. Exchanges reduce commission with activity-based discounts, charge different rates in different countries, and run promotional periods of very low or zero commission on selected markets. The rate that matters is the one shown in your account for the market you are betting on. Every calculator on this site lets you type a custom rate for that reason, and the preset rates for named exchanges are placeholders that need checking against the current schedule.
Commission across a whole position
The per-market rule changes the arithmetic when you hold several bets in the same market. Suppose you back a team at 2.50 for 100 and, after the price shortens, lay it at 2.00 for 125. If the team wins you make 150 on the back and lose 125 on the lay: net +25. If it loses you lose 100 on the back and win 125 on the lay: net +25. Under a per-bet model, commission at 5% would be charged on 150 in the first case and on 125 in the second, giving 142.5 − 125 = 17.5 and −100 + 118.75 = 18.75. Under the per-market rule it is charged once on the net 25, leaving 23.75 either way. The difference is nearly six units on a position that nets 25: trading in and out of a market is far cheaper than the per-bet model suggests, which is why exchange traders can work on small edges.
The calculators on this site apply commission per bet, so a multi-bet position in one market is slightly better than they show. For a single bet, or for bets in different markets, the two models agree.
Choosing between exchanges at different rates
When the same price is available at two exchanges, the lower commission wins, but the difference is smaller than it looks at short prices and larger at long ones. Two exchanges quoting 1.50, one at 2% and one at 5%, are worth 1.49 and 1.475: a gap of a hundredth and a half. At 6.00 the same two rates give 5.90 and 5.75: a gap of fifteen hundredths. A bettor who backs outsiders should weight commission heavily in choosing a venue; a bettor who backs favourites can go where the liquidity is. Liquidity matters because a better rate on a price you can only get for a fraction of your stake is worth less than a worse rate with depth behind it.
Quick reference
| Price | 0% | 2% | 5% | 8% |
|---|---|---|---|---|
| 1.50 | 1.500 | 1.490 | 1.475 | 1.460 |
| 2.00 | 2.000 | 1.980 | 1.950 | 1.920 |
| 3.00 | 3.000 | 2.960 | 2.900 | 2.840 |
| 5.00 | 5.000 | 4.920 | 4.800 | 4.680 |
| 10.00 | 10.000 | 9.820 | 9.550 | 9.280 |
Each cell is the effective decimal price after commission on net winnings. The rate columns are placeholders for whatever your account actually charges; the pattern, not the exact figure, is the point.
Reading an exchange screen
An exchange shows three back prices and three lay prices per outcome, with the money available at each. The best back price is the highest on the back side; the best lay price is the lowest on the lay side, and the gap between them is the spread. On a liquid market the gap is one tick; on a thin one it can be several. Backing at the best available price takes money someone has offered to lay; offering a better price than that and waiting for a backer is the same as laying at the price you ask. Everything on the screen is quoted before commission, so the back price you see is worth 1 + (price − 1)(1 − c) once matched, and the lay price you see costs you commission on the backer's stake if the outcome loses. The commission calculator is meant to be used alongside the screen: type the visible price, and read off what it is really worth before deciding whether it beats the bookmaker.
Summary
- Commission on net winnings: effective odds = 1 + (odds − 1)(1 − c); break-even = 1 / effective.
- Commission on turnover: no single effective price; break-even = (1 + c) / odds.
- Commission costs more at longer prices.
- Lays are backs on the complement at L / (L − 1), staked with the liability.
- Always compare after commission, and use the rate from your own account.