actfap

Lay Bet & Matched Betting Calculator

Get the exchange lay stake, liability and locked-in result for a qualifying bet or a free bet.

This sizes the exchange side of a bet you have already decided to place and prices what the pair locks in, so it answers what a qualifying bet costs or what a token converts to, and says nothing about the offer that prompted it.

Exchange slip

Lay stake

$9.43

Liability, before the bet is matched

$20.75

This has to already be sitting in your exchange account. It is not an extra loss on top of the result below, it is what the exchange holds while the bet is open.

Qualifying loss

−$0.75

If selection wins

Back bet wins$20.00Lay liability paid$20.75You finish−$0.75

If selection loses

Lay stake kept, after commission$9.25Back stake lost$10.00You finish−$0.75

You lock in this result whichever way the selection goes, the cost of qualifying for an offer. Closer lay odds and lower commission shrink it toward zero.

Your two bets

Odds gap

6.7%

Every price field reads a decimal price (2.50). Switch the format above to enter prices the way your book quotes them, or open the odds converter to see one price in all of them.

Retention across back odds

1.50back odds20.00
  • Free (SNR)32% to 89%
  • Free (SR)94.5% to 93.4%
  • Qualifying, selected-5.5% to -6.6%
  • Exchange liability, Qualifying line

This curve holds the lay price at 5% above the back price and commission at 2%, so it describes odds in general rather than the two prices you entered above. A stake-not-returned token climbs from 32% to 89% across this range; a stake-returned token barely moves, 94.5% to 93.4%; a qualifying bet stays flat, -5.5% to -6.6%. The dotted line is exchange liability for the Qualifying line (right scale, not drawn to numbers).

Even a qualifying bet ties up exchange liability while it is open, from $5.55 to $190.66 across this range, which is what the dotted line above prices here too. Switch to a free-bet mode to see the bigger trade: retention climbs steeply for a stake-not-returned token and barely moves for a stake-returned one.

Written and reviewed by Alexander Korsager · last reviewed October 1, 2026 · corrections

The idea

What this tool shows

Laying a bet on an exchange means betting on it not to happen, which cancels out a back bet you placed with a bookmaker. Size the lay correctly and your net result is the same whether the selection wins or loses, that is the mechanic behind matched betting and cash-out.

  1. For a normal back bet the lay stake is the back return divided by the lay odds minus commission, so that the exchange pays you enough on a loss to match your bookmaker profit on a win. The result is usually a small, known 'qualifying loss', the price of unlocking an offer. For a free bet where the stake is not returned, the lay stake is based on the potential winnings alone, and the locked-in figure is a guaranteed profit instead: around 70–80% of the free bet's face value at odds of 4.0 to 6.0, holding commission at 2% and the lay price within about 5% of the back price. Both of those matter to the answer, and the band is not a peak. Retention keeps climbing past 6.0; what stops most people going there is the liability the exchange holds while the bet is open, not the maths.

  2. The numbers assume both bets get matched at the odds you enter and that commission applies only to net exchange winnings.In practice, watch that lay liability (it has to be sitting in your exchange account before the bet is matched) and that the back and lay odds haven't drifted apart, which is what turns a tidy qualifying loss into a real one.

FAQ

Questions players ask

What lay odds should I look for?

As close to the back odds as possible. The tighter the gap between back and lay, the smaller the qualifying loss and the more of a free bet you keep. A gap of more than a few percent usually means waiting for the market to settle or picking a more liquid one.

Why is my qualifying loss bigger than expected?

Either the lay odds are well above the back odds, or commission is high, or both. At 5% commission and a 10% odds gap the loss on a $10 qualifier can be over $1. Low-commission exchanges and matching odds keep it near zero.

Stake not returned vs stake returned, which is my free bet?

Almost all bookmaker free bets are stake-not-returned: if it wins you get the winnings but not the token back. Stake-returned free bets and risk-free bet refunds behave differently and are worth more, so pick the right mode.

Methodology

How this is calculated

Every figure in the panel comes from one equation. Call R what the bookmaker hands back if the selection wins: stake plus winnings for a qualifying bet or a stake-returned token, and the winnings alone for a stake-not-returned token, since the book keeps that one.

The lay stake is R divided by the lay odds less commission, which is the size that makes the exchange pay on a loss exactly what the bookmaker pays on a win.

Liability, the cash the exchange holds while the bet is open, is that stake times the lay odds less one, and it is the number that decides whether you can place the bet at all. The two branch figures are then arithmetic. If the selection wins you collect R, less the back stake if you paid one, less the liability.

If it loses you keep the lay stake less commission, less the back stake if you paid one. Those two are equal for every input, which is what 'locked in' means here, and the panel prints both rather than one averaged figure so that a difference would be visible. Retention is that locked-in figure over the face value you entered.

Common misreadings3
  • 'Locked in' describes the arithmetic, not the afternoon. The certainty is that two settled bets pay the same amount either way, and the ways the exercise actually goes wrong are execution: a lay that never matched, a bet that got voided, an account that got restricted, or the wrong one of the three modes selected.
  • A qualifying loss is a real loss and you pay it every time. The page shows it in the loss color deliberately. It is only ever recovered if the offer it qualifies for returns more than the sum of the qualifying losses paid to reach it, and this page prices one leg rather than that sum.
  • Retention is a conversion rate, not a return on your money. Keeping 75.7% of a stake-not-returned token is 75.7% of something that was never cash and could not be withdrawn. The money genuinely at stake in the exercise is the liability the exchange holds plus the qualifying losses already paid, and none of that is what the percentage is measured against.
Assumptions7
  • Both bets are matched at exactly the prices you type. A lay is an order rather than a purchase: it fills only if someone takes the other side at that price and for that size. Nothing here checks that the money is there, and on a thin market the price that fills is not the one that was on screen when you looked.
  • Commission is charged on this one lay, and only when it wins. That is the simple case rather than the general one. Betfair's own Automation Hub documentation (betfair-datascientists.github.io/gettingStarted/commission, re-verified 2026-10-01) states that commission is calculated on net winnings across a whole market and is cumulative within it, so a loss elsewhere in the same market reduces the profit the rate is applied to. If this lay is your only position in that market the figure is exact; if it is not, the commission you actually pay is at most this and often less.
  • The back stake is cash and a token is not. A qualifying bet costs you the stake whichever way it goes, and both free-bet modes assume the token cost you nothing, so it never appears in the losing branch. If the token did cost something, a deposit you had to place or a stake you had to risk, neither free-bet mode prices that and the locked-in figure is too high by it.
  • Every price field reads whichever format the odds format control is set to, and only that one. The control exists, rather than a field that takes anything, because 150 is a real decimal price as well as the American price +150: a field reading both would have to guess which you meant on exactly the input where guessing wrong is invisible, since a wrong price of that kind still prints a complete and correctly formatted panel. A price that does not read in the selected format shows nothing rather than a figure, and the page names the format it would read in when there is one. Switching the format rewrites the prices already entered, so a field never holds a number in a format the page is no longer reading, and a fractional price is rewritten to the closest fraction with a denominator of 100 or less.
  • The odds gap is measured on the decimal price, not on implied probability. The panel shows the lay price minus the back price over the back price, so a 10% gap at 2.0 means a lay of 2.2.
  • One lay, at one price, placed once. No partial fill, no laying half now and half later at a different price, and no allowance for the market moving in the interval between placing the back bet and placing the lay.
  • The retention and liability curve on the page holds the lay price at a fixed 5% longer than the back price and commission at whatever you have it set to. It prices a modeled lay spread across a range of back odds, not a lay price you could get quoted today at any one price, which is what the panel above the curve is for. Some published matched-betting formulas skip the lay price entirely (retention equals the back odds minus one, times one minus commission, over the back odds, a figure with no lay-odds term at all): that frictionless version returns 78.4% at back odds 5.00 and 2% commission, where this page's curve, pricing a real 5% lay spread, returns 75.0% for the same inputs.
Limits5
  • Retention rises with the back odds and does not stop, so there is no odds band that keeps the most value. At 2% commission with the lay matching the back, a stake-not-returned token returns 65.8% at 3.0, 81.9% at 6.0, 86.0% at 8.0 and 93.4% at 21.0. What caps it in practice is the liability printed beside the lay stake: converting a $10 token at 21.0 needs $190.66 sitting in the exchange, against $22.61 at 4.0. The common advice to work at 4.0 to 6.0 is about how much cash gets tied up, and about how far back and lay prices drift apart on long shots, not about where this formula peaks.
  • A stake-returned token moves the other way, and carrying the wrong rule across is the likely mistake. Its retention edges down as the odds lengthen: 99.0% at 2.0, 98.5% at 4.0 and 98.2% at 8.0, on the same 2% commission with matching prices. The two token types want opposite ends of the board, and the mode selector is the only thing on the page that tells them apart.
  • Matching the lay to the back does not take the qualifying loss to zero. Commission is still charged on the lay when it wins, which leaves $0.10 on a $10 qualifier at 2.0 and $0.17 at 6.0, at 2% commission. Closing the gap shrinks the loss toward that residue rather than to nothing.
  • Nothing here prices the bet not standing. The arithmetic covers two bets that both settle as placed, and it cannot see a bookmaker voiding the bet, limiting the account, or settling an offer on terms other than the ones you read. Those leave you holding one side of a position that was only balanced while both sides existed.
  • An exchange with a two-sided market and a stated commission is what this maths assumes you have access to. That is an ordinary assumption in the UK and Australia and an uneven one in the US, where much of what offers a lay side runs as a federally regulated prediction market rather than a state-licensed sportsbook exchange. Check what actually operates where you are before treating a lay stake as an order you can place.
Evidence

The figures are computed by layBet in math.ts, and math.test.ts pins them: a $10 qualifier at 3.0 against a lay of 3.2 at 2% commission gives a lay stake of $9.43, a liability of $20.75 and a locked-in loss of $0.75; a $10 stake-not-returned token at 5.0 against 5.2 gives $7.72 laid and $7.57 kept, which is the 75.7% quoted above; the stake-returned version of the same inputs keeps $9.46.

Two further tests pin the identities the page rests on, that the winning and losing branches agree to eight decimal places in each of the three modes, and that a lay matching the back at zero commission produces a lay stake equal to the back stake and no qualifying loss at all.

The retention and liability figures quoted in the limits are that same function evaluated at other prices.

The commission model is the simple per-bet case: Betfair's own Automation Hub documentation, betfair-datascientists.github.io/gettingStarted/commission (first retrieved 2026-09-09, re-verified live 2026-10-01, which corrected this page's own citation: the worked example reads 8% commission and $12.00 owed, not the 7% and $10.50 previously quoted here), is the source for commission being calculated on total market profit and cumulative across a market; that page's own worked example (a $50 back bet at 4.00, 8% commission, $12.00 owed on $150 winnings) is a single-bet racing scenario at a different rate from this page's shipped 2%, not a matched-betting worked example this tool's figures are checked against.

A passing test means the code computes this model correctly, not that the prices you type are prices you will be filled at.

Gambling can cause serious harm. The National Problem Gambling Helpline is free, confidential and open 24/7, call or text 1-800-MY-RESET (1-800-697-3738), or chat at ncpgambling.org/chat.

Read up

Go deeper

Related tools

Try next

We’d like to use privacy-friendly analytics (Google Analytics, Vercel Analytics and Speed Insights) to see which tools and pages are actually useful, so we can improve them. Nothing is sold, and nothing loads unless you say yes. See our privacy policy for details.