bookmaker margins for bettors: the cut the book takes, and how to shop it

what the margin is

Convert the odds on any market into implied probabilities and they never quite add up to 100%. Home 2.50, draw 3.30, away 3.00 sums to 103.6%, not 100%. The excess is the margin: the overround, the vig, the bookmaker's cut, folded invisibly into every price. The implied probability guide explains the odds-to-probability conversion in full; this guide is about the margin itself, how big it is, how to measure it, and what it costs you.

The margin is not an accident of rounding. It is how the book charges for taking your bet, the way a casino edge is how the casino charges for its tables. It is paid on every bet, win or lose: a winning bet at a fat-margin book returns less than the same bet at a thin-margin book, and a losing bet was always going to lose a little more of your stake in expectation.

worked example: the same match at three books

Take a Saturday 3pm: Manchester City against Arsenal, priced three ways at three books. Book A prices home 2.50, draw 3.30, away 3.00. The implied probabilities are 40.0% + 30.3% + 33.3%: a 103.6% book, a margin of 3.6%. Book B prices the same match at 2.30, 3.20 and 2.75: implied 43.5% + 31.3% + 36.4%, a 111.1% book, a margin of 11.1%. The exchange prices 2.62, 3.45 and 3.05: implied 38.2% + 29.0% + 32.8%, a 99.9% book before commission, and at 2% commission an effective book of about 101.9%.

Same match, same three outcomes, and Book B is taking nearly 7.5 points more than Book A. The odds converter does the whole sum in seconds: enter the three prices and read the total implied probability, which is the size of the margin.

why the headline price lies

Notice the trap in that row of prices. Manchester City at 2.30 looks like the big price of the three books, but it is not a gift: it is Book B dressing an 11.1% margin up as generosity. Books compete on single headline prices because that is what punters read, then widen everything else to pay for it. Judging a book by the price of one favourite is judging a supermarket by the price of milk. The only honest measure is the whole market: add the implied probabilities of every outcome you are actually prepared to bet, and compare that total across books. The book with the smallest sum is taking the smallest cut.

Where a market sums to 103-105%, the book is earning a living. Where the sum creeps past 108% on a mainstream market, you are paying for the brand, the advert and the free-bet offer, all folded into your prices.

what the margin costs across a year

Seven and a half points does not sound like much until you count how often you pay it. On a £10 stake, betting at an 11.1% book instead of a 3.6% book costs about 75p per bet in expectation: not on losing bets, on every bet. Bet twice a day, five days a week, fifty weeks a year, and that is five hundred bets: roughly £375 a year handed to the book for no other reason than picking the wrong shop.

That is the quiet arithmetic of margins. The margin is the one number that applies to every bet you place, and it is deducted from the price before the race is even run. You can pick the right horse and still lose money to the wrong book. The reverse is just as true: pick the right book and you can lose the race and still come out ahead of someone who backed the winner at the wrong one.

Notice that the difference is not a tip or a system. It is pure shopping, available to every bettor on every bet, with no opinion about the sport required. A regular who bets the same selections through the low-margin shop instead of the high-margin one has improved their results by several points without improving their judgement at all.

the exchange discount

The exchanges are where the margin argument bottoms out. A bookmaker's margin is collected invisibly inside the price; an exchange charges commission openly on winnings instead. The exchange book in the worked example summed to 99.9% before commission, because punters lay as well as back and the two sides roughly cancel out, and 2% commission on the winning side brings the effective total to about 101.9%: cheaper than Book A and far cheaper than Book B.

That is why sharp bettors price everything against the exchange first. The exchange book is the nearest thing betting has to a fair market, so its stripped probabilities are the best baseline for judging a bookmaker's price. When the book's best price on a horse is 2.50 and the exchange lays 2.62, the difference is the book's margin on that one selection, and it is usually several points fatter than the headline number suggests.

where the fat margins live

Margins are not uniform; they track liquidity and attention. The more money and the more eyes on a market, the thinner the cut. Main football 1X2s and big-race win markets are the thinnest, often 103-106% at the sharp books. Everything else is fatter: in-play markets, small leagues, second divisions, where 108-115% is ordinary. Horse racing win books run a step above football, and the place market is fatter still — the each-way guide shows the place book running near double the win margin.

Special offers sit inside this picture rather than outside it. A book that prices its football at 104% and its in-play at 112% is the same book with two faces; the bonus and the boosted price are marketing paid for by the fat margins elsewhere. Once you can read the whole book, offers stop looking like gifts and start looking like signposts to where the margin is widest.

Multiples are the worst of all, because margins multiply instead of adding. An accumulator with five legs, each priced on a 105% book, is a 127.6% book overall: one losing leg beats you, and even a clean sweep pays a fifth of its fair price less. The dutching calculator shows the same number from the other side: its book% readout is exactly the margin you are paying across the selections you stake.

stripping the margin to find value

Once you know the margin's size, you can remove it. Divide each implied probability by the total book. City at 2.50 on Book A's 103.6% book carries a raw implied probability of 40.0%, but a true implied probability of 38.6% (40.0 ÷ 103.6). That stripped number is the price the book is really offering, and it is the number to beat. Rate City at 45% to win and you have value at both books, and more at Book A. Rate them 38% and you have value at neither: the margin ate the difference.

The value betting guide is the full framework for that comparison, and the odds converter strips and compares the numbers for you. Then size the edge like any other: write down your probability, and stake a fraction of your bankroll with the staking calculator. The margin matters less the bigger your edge, but it never stops mattering: a 3% margin on a 5% edge is a third of your profit, gone.

the margin checklist

A short routine before any bet, in order: list every outcome in the market you are betting and convert each price to implied probability, using the odds converter when the arithmetic is tedious; add the probabilities and write down the total — that is the margin, and anything over about 106% on a mainstream market is a fat book; compare the same market at two other books and bet at the smallest total; strip the margin from the price you are taking and compare the true implied probability with your own estimate, per the implied probability method; treat headline prices and free-bet offers as marketing, not value; and apply the value betting test: a bet is only worth making when your probability beats the stripped price.

the honest endnote

Margins are not a scandal and not a trap; they are the rent on the building. The skill is knowing how much you are paying and choosing the cheapest door. A bettor who shaves five points off the margin on every bet has done the work of a decent winning streak with none of the variance, because the margin is the only number in betting you can control completely, before any race is run. Choose thin books, strip the margin, beat the stripped price, and size with discipline. And on the rare day when one book's margin is small enough that the sums across two books dip below 100%, that is the shape covered in the arbitrage guide: the one bet that needs no opinion at all. Bet within your means, treat the bankroll as money you can afford to lose, and if betting stops being enjoyable, stop. 18+ | BeGambleAware.