How bookmakers set odds, starting from a probability
Every price begins with a view on likelihood. A trader, or more often a pricing model the trader supervises, estimates the chance of each outcome from form, team news, venue and years of past results. For a Premier League match that might mean a 45% chance of a home win, 28% for the draw and 27% for the away side.
Those three numbers add up to exactly 100%, because one of the outcomes has to happen. Turned straight into odds they would give a fair book, where the bookmaker earns nothing over time. No firm trades that way. The real question about how bookmakers set odds is how much they shave off the fair price, and where.
Football models lean heavily on goal expectancy. If the model expects the home team to score about 1.6 goals and the visitors about 1.1, it can produce the chance of every scoreline, and from those the match result, both teams to score and over/under prices. One set of inputs feeds dozens of markets, so one bad assumption spreads everywhere.
The model is only the opening position. Traders adjust it for things a model handles poorly: a manager under pressure, a squad rotating before a cup tie, a pitch frozen at noon. They also watch what rival firms and the exchanges are offering. A price far out of line with the market will attract sharp money within minutes.
From fractional and decimal odds to implied probability
Odds and probability are two ways of writing the same idea. Decimal odds show the total return for a £1 stake, so 2.50 pays £2.50, stake included. To find the probability the price implies, divide 1 by the decimal odds. One divided by 2.50 is 0.40, a 40% chance. That single division is all you need.
Fractional odds, still the default in British racing and on most high-street boards, need one extra step. Divide the first number by the second and add 1. So 6/4 becomes 1.5 plus 1, giving 2.50 in decimal, and 5/4 becomes 2.25. Evens is 1/1, which converts to 2.00 and implies a flat 50%.
Short prices can look odd in fractional form. A favourite at 4/9 converts to roughly 1.44, implying about a 69% chance. A long shot at 33/1 converts to 34.00, which works out at just under 3%. Reading a market as percentages is the first step in seeing how bookmakers set odds.
A worked example: the overround in a three-way market
Say a bookmaker prices a match at 2.10 for the home win, 3.40 for the draw and 3.60 for the away win. These are made-up round odds, chosen to keep the arithmetic easy. Converting each one gives implied chances of 47.6%, 29.4% and 27.8%. On the fair book from our model they would have summed to precisely 100%.
Add the three together and the total comes to 104.8%. That extra 4.8% is the overround, the slice the bookmaker has built into the book. It means the prices on offer are shorter than the true chances warrant. Each outcome has been trimmed a little, and the trims together create the firm’s edge on the match.
The same book in fractional terms reads 11/10, 12/5 and 13/5. Written that way it looks harmless, and that is part of the point. Few punters convert a fraction in their head, so the overround stays hidden in plain sight. Once you see the market as three percentages, the extra 4.8% becomes impossible to miss on any coupon.
The trims are rarely spread evenly. Many firms load more of the overround onto the draw and the outsider, where casual punters care less about the exact price. A favourite backed heavily by the public may also be cut further. Comparing each outcome with the exchange price usually shows you where the bookmaker has taken the biggest bite.
To watch the whole process on a real fixture, open our step-by-step explainer of how bookmakers set odds. It takes one Premier League match priced by every site in our odds record, turns each price into a chance, shows the trim that creates the margin and ends with what shopping around for the best price saves.
Margin and overround: two numbers for the same edge
People use the words margin and overround as if they mean the same thing, but they are calculated differently. The overround is the excess above 100%, the 4.8% in our example. The margin is the share of all money staked that the bookmaker expects to keep, which comes from dividing the overround by the whole book.
In our example the margin is 4.8 divided by 104.8, which gives about 4.6%. Put another way, if punters backed all three outcomes in proportion to the prices, the firm would pay back roughly £95.40 for every £100 taken. The difference between the two figures stays small at low overrounds and grows as the book gets fatter.
We quote the margin in our tables because it answers the question a punter cares about: how much of my stake does the price cost me over time? A bookmaker charging a 4% margin hands back more of every pound than one charging 8%. Over a season of weekly bets that difference adds up far faster than most people expect.
Why the margin differs by market and league
Margins are not fixed across a bookmaker. The same firm will price a Premier League match tightly and a League Two match more generously in its own favour. Top-flight games draw huge volumes, sharp punters and constant comparison, so there is less room to hide a fat margin. Lower-league fixtures attract less scrutiny and less informed money.
The table shows the Premier League match-result margins we measure. For the full picture across all seven competitions we track, and for exchange books with commission included, see our bookmaker margins tables. Two patterns stand out in that data: lower-league football carries wider margins than the Premier League at most firms, and exchanges sit lowest.
The spread between bookmakers matters just as much. On identical matches, the cheapest and the dearest fixed-odds firms in our data are several percentage points apart. That is a large gap. A punter who takes the dearer price on every bet is handing over noticeably more of each stake for exactly the same outcome, with nothing extra in return.
Market type changes the overround too. A two-way market such as over or under 2.5 goals spreads the edge across two outcomes, while an outright with twenty runners spreads it across twenty. Correct-score and first-goalscorer markets usually carry far bigger overrounds than the match result, because each individual price is long and harder for punters to compare.
Trading, liabilities and why prices move
Once a market opens, the trader’s job shifts from estimating probability to managing risk. Each bet adds a liability, the amount the firm would pay out if that outcome wins. If too much money piles onto one side, the bookmaker faces a heavy loss on that result even if its original price was perfectly sound.
The usual response is to shorten the price on the busy outcome and lengthen the others. That makes the popular side less attractive and the neglected ones more tempting, pulling money back towards balance. A perfectly balanced book is rare. Most firms are content to hold a position on some results, provided the overall margin stays in their favour.
Information moves prices more sharply than money does. When a team sheet lands an hour before kick-off and the first-choice striker is missing, every goal-related market shifts at once. Traders also follow the exchanges closely, since prices there react within seconds to informed money. A bookmaker slow to follow a big exchange move will be picked off.
Not all money is treated equally. Bets from accounts with a history of beating the closing price tell the trader something, so they can move the line further than a much larger stake from a casual customer. The same logic explains why some winning accounts end up with reduced stakes, which we cover in our guide to betting account restrictions.
Understanding how bookmakers set odds also explains why prices across firms cluster so closely. Most traders watch the same exchange markets and the same handful of sharp rivals, so opening lines tend to converge quickly. The differences that remain come mostly from the margin each firm chooses to charge, and that is exactly what our tables measure week by week.
How bookmakers set odds on outrights and in-play markets
Outright markets, such as the Premier League winner or a golf tournament, follow the same principle on a larger scale. Each runner gets a price, and the implied chances can add up to 120% or more. With dozens of options, an extra point or two on each long shot is hard to spot, and the overround grows accordingly.
In-play betting shows how bookmakers set odds at speed. A goal, a red card or a penalty changes every probability in an instant, so markets are suspended for a few seconds while the model recalculates. Live prices tend to carry wider margins than pre-match ones, partly to cover the risk of a punter watching a faster feed than the trader.
A side point worth knowing: prices for a weekend fixture are often published days ahead with low maximum stakes, partly to test opinion before the bigger money arrives. The price left at kick-off has absorbed far more information, so many professionals judge their own bets by comparing them with that final closing line instead of the early one.
Working out a market’s margin yourself
You can check any market in about five minutes with a phone calculator. Write down every price in the market, convert each one to decimal if needed, and divide 1 by it. Add up the results. The total minus 100% is the overround. Divide the overround by the total and you have the margin. Our betting margin calculator runs the same sums for any market as you type the prices.
Two-way markets are even quicker. Suppose over 2.5 goals is priced at 1.90 and under at 1.90, again round example odds. Each implies 52.6%, so the book sums to 105.3% and the overround is 5.3%. A fair price for a genuine coin flip would be 2.00 on both sides, and the missing 0.10 is the bookmaker’s charge.
Do it on two or three bookmakers for the same match and the differences jump out. Do it again on a League Two fixture and you will usually see the total climb. Check the exchange as well, remembering that its commission comes off winnings, so its true cost sits a little above what the raw prices suggest.
A lower margin does not make a bet a good one, and no price removes the chance of losing your stake. It simply means you are paying less for the same outcome. Before betting at all, set a deposit limit in your account, and remember that GAMSTOP blocks every UK-licensed site with one registration.
Deposit limits changed on 30 September 2026. Under Gambling Commission rules, only a gross limit, counting all money paid in without netting off withdrawals, may be called a deposit limit. That makes your chosen figure a clean ceiling on spending. Pair it with the cheapest price you can find and both sides of the bet stay under control.
Frequently asked questions
How do bookmakers set odds?
Traders start from a model of how likely each outcome is, using form, team news and historical data. They convert those chances into prices, then shorten every price slightly so the book carries a margin. The odds then move as money arrives and new information comes in.
What is the overround in betting?
The overround is the amount by which the implied probabilities of every outcome in a market add up to more than 100%. A three-way football market summing to 105% has an overround of 5%. It is the bookmaker's built-in edge on that market.
What is a bookmaker margin?
The margin is the share of every pound staked that the bookmaker expects to keep if money is spread across the outcomes in line with the prices. It is calculated from the overround and is always a little smaller than it, because the bookmaker keeps its edge on a larger total.
How do I work out implied probability from odds?
Divide 1 by the decimal odds. Odds of 2.50 imply a 40% chance. For fractional odds, convert first: 6/4 becomes 1.5 plus 1, so 2.50, giving the same 40%.
Why do odds change before a match?
Prices move when the bookmaker learns something new, such as a team sheet or an injury, and when money comes in heavily on one side. Traders shorten the outcome that is building up liability and lengthen the others to attract balancing bets.
Do all bookmakers use the same margin?
No. Our weekly margin tables show a gap of several percentage points between the cheapest and the dearest fixed-odds bookmakers on the same football matches. The gap also widens in lower leagues at most firms.
Why are betting exchange prices usually better?
On an exchange, punters set prices against each other and the exchange takes a commission on net winnings. Once that commission is included, exchange books still sit below every fixed-odds bookmaker in our football tables.
Can a market have an overround below 100%?
Within one bookmaker it is very rare and usually a pricing error that is corrected quickly. Across different bookmakers, combining the best price for each outcome can sometimes bring the total close to or under 100%, though accounts that do this often face stake limits.
Are horse racing margins higher than football margins?
Races with large fields usually carry a much bigger overround than a three-way football market, because the bookmaker spreads its edge across many runners. Our own measurements cover football only, so our tables do not rate racing prices.

















