Turn each price into an implied probability (1 ÷ odds), add them up to get the book percentage, then take away 100%. What's left is the overround. With odds of 1.80 and 2.00 you get 55.56% + 50.00% = 105.56%, an overround of 5.56 points. That figure measures the excess built into the prices, not the bookmaker's profit.
The bookmaker's margin in three lines of maths
Turn each decimal price into an implied probability (1 ÷ odds), add them all up, then take away 100%. What's left is the market's overround: the excess probability tucked into the odds.
Three steps are enough. The implied probability is 1 ÷ odds. The book percentage is the sum of those probabilities. The overround is the book percentage minus 100%.
Take odds of 1.80 and 2.00:
- 1 ÷ 1.80 = 55.56%
- 1 ÷ 2.00 = 50.00%
- Total = 105.56%
- Overround = 5.56 points
Key point. 105.56% is the book percentage and 5.56 points is the overround. Neither of them is the bookmaker's profit.
What is a bookmaker's margin?
The margin is the edge a bookmaker builds into its odds. It usually shows up as implied probabilities that add up to more than 100%. The excess goes by overround, book margin or vigorish, depending on the market and who's talking.
In a perfectly fair market, the probabilities of all the outcomes would total exactly 100%. Two outcomes estimated at 50% each would both be priced at 2.00. A bookmaker may still offer 1.91 and 1.91. Each price then works out at an implied probability of about 52.36%, and the sum reaches 104.71%.
| Term | Definition |
|---|---|
| Implied probability | The probability you get straight from the odds |
| Book percentage | Sum of the implied probabilities of all outcomes |
| Overround | The part of the book percentage above 100% |
| Underround | A sum below 100%, possible when you combine the best odds from several sources |
| Realised hold | Share of the stakes the bookmaker keeps after settlement |
| Margin-free probability | Estimate obtained once the overround is redistributed |
In everyday talk, "margin" and "overround" pass for synonyms. If you want to be precise they're two slightly different measures, as the section on the overround and the theoretical margin shows.
How do you convert odds into an implied probability?
With decimal odds, divide 1 by the odds, then multiply by 100. Odds of 2.00 give 50%, and odds of 4.00 give 25%.
Implied probability (%) = (1 ÷ decimal odds) × 100
| Odds | Calculation | Implied probability |
|---|---|---|
| 1.25 | 1 ÷ 1.25 | 80.00% |
| 1.50 | 1 ÷ 1.50 | 66.67% |
| 1.80 | 1 ÷ 1.80 | 55.56% |
| 2.00 | 1 ÷ 2.00 | 50.00% |
| 3.00 | 1 ÷ 3.00 | 33.33% |
| 4.00 | 1 ÷ 4.00 | 25.00% |
It's called implied because it's hidden inside the displayed price. It isn't a margin-free estimate yet, and it's certainly not a guaranteed probability.
How do you calculate the book percentage and the overround?
Add up the implied probabilities of every outcome on the same market: the total is the book percentage, and whatever sits above 100% is the overround.
For a market with n outcomes and odds O₁, O₂, … Oₙ:
- Book percentage = (1 ÷ O₁) + (1 ÷ O₂) + … + (1 ÷ Oₙ)
- Book percentage (%) = sum of the implied probabilities × 100
- Overround (%) = book percentage (%) − 100%
Here's a three-outcome example, a 1X2 market priced at 2.10 / 3.40 / 3.60:
- 1 ÷ 2.10 = 47.62%
- 1 ÷ 3.40 = 29.41%
- 1 ÷ 3.60 = 27.78%
- Book percentage = 104.81%
- Overround = 4.81 points
Key point. The overround is calculated market by market. A bookmaker doesn't have one permanent margin.
The same operator can apply very different levels depending on the sport, the competition, the type of bet and the moment you check the price.
A real case: Borussia Dortmund v Eintracht Frankfurt (Bundesliga). The second table applies this calculation to the raw odds of an unnamed reference operator before kick-off. The implied probabilities add up to more than 100%, and the excess is the overround. The chart shows the OddScore margin-free odds for the same match.
From 1 May to the start, the margin-free odds on “Borussia Dortmund win” moved from 1.60 to 1.78 (+11.40%), a margin-free probability of 62.62% then 56.21%. Lowest value seen: 1.57; highest: 1.79. This chart describes a past market movement: it says nothing about the match result and is not betting advice.
| Outcome | First tracked odds | Last odds before the start | Change | Margin-free probability (first, then last) |
|---|---|---|---|---|
| Borussia Dortmund win | 1.60 | 1.78 | +11.40% | 62.62% then 56.21% |
| Draw | 4.90 | 4.55 | −7.12% | 20.41% then 21.97% |
| Eintracht Frankfurt win | 5.89 | 4.58 | −22.17% | 16.99% then 21.82% |
Raw odds from a reference operator, not named, before the start: the implied probabilities add up to 103.42%, a margin of about 3.42%. Dividing each probability by that total gives the margin-free probability (proportional method, slightly different from the OddScore consensus in the chart).
| Outcome | Raw odds | Implied probability | Margin-free probability |
|---|---|---|---|
| Borussia Dortmund win | 1.70 | 58.82% | 56.88% |
| Draw | 4.44 | 22.52% | 21.78% |
| Eintracht Frankfurt win | 4.53 | 22.08% | 21.34% |
| Total | 103.42% | 100.00% |
Source: OddScore. Margin-free odds computed from the odds of the international reference bookmakers, snapshots from 1 May to 8 May 2026, match already played. Data frozen on 11 October 2026.
Full example on a tennis match
We start with Player 1 at 1.80 and Player 2 at 2.00.
- Implied probabilities. Player 1: 1 ÷ 1.80 = 55.56%. Player 2: 1 ÷ 2.00 = 50.00%.
- Book percentage. 55.56% + 50.00% = 105.56%.
- Overround. 105.56% − 100% = 5.56 points.
- Normalised probabilities (implied probability ÷ book percentage). Player 1: 55.56 ÷ 105.56 = 52.63%. Player 2: 50.00 ÷ 105.56 = 47.37%.
- Theoretical margin-free odds. Player 1: 1 ÷ 52.63% ≈ 1.90. Player 2: 1 ÷ 47.37% ≈ 2.11.
| Outcome | Displayed odds | Implied probability | Normalised probability | Estimated margin-free odds |
|---|---|---|---|---|
| Player 1 | 1.80 | 55.56% | 52.63% | 1.90 |
| Player 2 | 2.00 | 50.00% | 47.37% | 2.11 |
The normalised probabilities do add up to 100%.
What's the difference between the overround and the theoretical margin?
The overround measures what sits above 100% in the book percentage. To turn that into a theoretical rate on the total amount, you need the full size of the book.
If B is the book percentage as a decimal: overround = B − 1, theoretical return rate = 1 ÷ B, theoretical margin = 1 − (1 ÷ B).
Using the previous example, B = 1.0556:
- Overround = 1.0556 − 1 = 5.56 points
- Theoretical return rate = 1 ÷ 1.0556 ≈ 94.74%
- Theoretical margin = 100% − 94.74% = 5.26%
| Indicator | Result |
|---|---|
| Book percentage | 105.56% |
| Overround | 5.56 points |
| Theoretical return rate | 94.74% |
| Corresponding theoretical margin | 5.26% |
The figures 5.56 and 5.26 look alike even though they're two different ideas. That's why this table is here.
Saying the market has an overround of 5.56 points is correct. Saying the bookmaker will necessarily keep 5.56% of the stakes is wrong. What it keeps depends on the stakes received, how they're spread, the results and how the market is managed. Economists make the same point: the observable overround doesn't always match the profit rate the odds actually produce. I'd call this the easiest slip in the whole calculation, since 5.56 and 5.26 look so alike.
How do you remove the margin from the odds?
The simplest method is to divide each implied probability by the sum of all the probabilities in the market. This normalisation brings the total back to 100% and gives you an estimate of the margin-free market probabilities.
Normalised probability of outcome i = implied probability of outcome i ÷ sum of the implied probabilities. Estimated margin-free odds = 1 ÷ normalised probability.
Proportional normalisation assumes the margin is spread across all the outcomes in proportion. It's simple and good for learning, but it isn't always realistic. Other methods spread the margin differently, especially when a market has a very strong favourite or several longshots: logarithmic or power methods, the Shin method, or models that correct for the favourite-longshot bias.
Removing the margin always means choosing a spreading method. The result is an estimate, not an objective probability coming to light.
So it's better to say normalised probability, margin-free market probability or theoretical margin-free odds than "true probability".
Why does the margin vary?
The margin depends mostly on market liquidity, competition between bookmakers, how uncertain the outcome is, staking limits and the cost of managing the bet. Popular, highly competitive markets usually show tighter prices than secondary markets or markets that are hard to assess.
- A heavily played market can take more volume with a smaller margin per bet.
- On the most closely followed matches, punters compare prices easily, which can push margins down.
- A secondary market or a very specific bet carries more uncertainty, so it often gets a higher margin.
- Markets that accept large stakes need sharper, more competitive prices.
- The margin can move between the market opening and kick-off.
- An operator may cut its margin on purpose for certain popular events. That's one of the differences between sharp and soft profiles.
A three-outcome market doesn't automatically carry a higher margin than a two-outcome one. The number of outcomes changes the calculation without explaining the final overround: a very liquid 1X2 can sit below a much less competitive two-outcome market.
A betting exchange reads differently. It matches bettors with each other and earns a commission on winnings rather than a margin built into the odds.
Understand how bookmakers make money →
How do you compare margins fairly?
To compare bookmakers, use the same event, the same market, the same settlement rules and an observation time as close together as possible. Comparing different markets leads to a wrong conclusion.
A fair comparison covers the same match, the same type of bet, the same line (over or under 2.5 goals, say), the same settlement rules, odds taken at the same moment, every outcome in the market and the same odds format.
Comparing the margin on the 1X2 of a football match with the winner of a tennis match makes no sense. Comparing the 1X2 of the same match, at the same moment, across three bookmakers does.
If you pick the best price for each outcome across several bookmakers, the calculation no longer describes one operator's margin. It describes a composite market built from several sources. The total can even drop below 100%. That doesn't mean a bookmaker is offering a negative margin: the best prices just come from different operators.
What the overround tells you, and where it stops
The overround shows how much excess is built into the odds of a market. It gives you a sense of how competitive the prices look, the gap between the raw probabilities and a 100% market, and the theoretical return attached to the odds. It also gives you a base for comparing similar markets.
It says nothing about the bookmaker's guaranteed profit, the result of the match, the exact probability of each outcome, the quality of the operator's sporting model, how the stakes are spread, the future profitability of a bet or the exact reason a price moved.
A low margin means the prices sit closer to a 100% market. That doesn't make any bet good value in itself, and it doesn't prove the estimated probabilities are right.
How does OddScore remove the margin?
OddScore removes the margin so it can compare probabilities rather than raw commercial prices. Without that step, a movement can come from a change in probability, a change in margin or both at once.
Two bookmakers can see a match in a similar way and still show different odds, because they don't apply the same margin. OddScore converts the odds into comparable probabilities and strips out the margin built into the prices. Only then does it look at the direction of the movement, its size, its speed, how many operators are involved and how consistent the market is overall.
How OddScore analyses odds movements →
Put it into practice
Apply this guide to real odds.
See a market reading
A real match with its odds, movements and 0-100 signal.
Open the demoMargin calculator
Measure a bookmaker's margin on any market.
Calculate the marginDig into the market
Odds movements are only part of the story. Here are the next topics to read.
How do bookmakers make money?
Stakes, payouts, margin and risk management.
Understand the modelHow does a bookmaker work?
How odds are built, margin, risk and types of operator.
Read the guideHow do bookmakers adjust their odds?
Models, information, exposure and market movement.
See the adjustmentsHow do you read an odds movement?
Direction, size, timing and agreement between bookmakers.
See the movementsFrequently asked questions
What is a bookmaker's margin?
It's the edge a bookmaker builds into its odds. You can spot it when the implied probabilities of all the outcomes add up to more than 100%.
What is the overround?
It's the part of the book percentage that sits above 100%. A market that totals 105% has an overround of 5 points.
How do you convert odds into a probability?
With decimal odds, divide 1 by the odds, then multiply by 100. Odds of 2.50 give an implied probability of 40%.
How do you calculate the overround?
Add up the implied probabilities of every outcome in the market, then subtract 100%. If the total is 106%, the overround is 6 points.
Is the overround the bookmaker's profit?
No. It measures the excess contained in the prices. The profit actually made depends on the stakes received, how they're spread, the results and the operator's costs.
How do you remove the margin from the odds?
The simplest way is to divide each implied probability by the sum of all the probabilities in the market. Once normalised, the total is 100%.
Do margin-free odds give the true probability?
No, they give an estimate. It depends on the market prices and on how you spread the margin across the outcomes.
Why does the margin change from one market to another?
It moves with liquidity, competition, staking limits, how hard the outcome is to assess and the bookmaker's commercial policy.
Does a low margin mean a bet is good value?
No. It means tighter prices, nothing more. It can't tell you whether the market's view of the sport is right.
Does OddScore give tips?
No. OddScore converts and compares odds so market movements are easier to read. The platform gives no staking advice.
Sources & methodology
This page draws on the standard formulas for converting decimal odds, on economic research into the overround and betting-market prices, on the normalisation methods used to estimate margin-free probabilities, and on OddScore's own odds-analysis method. The Gambling Commission and BeGambleAware links point you to the UK regulator and to support for players.
- Convert each decimal price into an implied probability (1 ÷ odds).
- Add the probabilities to get the book percentage, then subtract 100% for the overround.
- Normalise the probabilities to estimate a margin-free market, bearing in mind that the result depends on the spreading method you choose.