Add up the implied probabilities from a football market and you will almost never get 100%. You get something like 104% or 106%, and that gap is not a typo. It is the margin, also called the overround, and it is built into every set of prices.
Understanding it takes about five minutes and changes how you read odds. This guide explains what the margin is, how to calculate it, where it hides inside a market and how to remove it when you want a cleaner probability.
Key Takeaways
- The margin (overround) is the amount by which implied probabilities exceed 100%.
- You calculate it by adding up the implied probabilities for every outcome in a market.
- Overround and margin are related but not identical, so check which one a source is quoting.
- The margin is not always shared evenly across outcomes, and longshots often carry more of it.
- A study covering 2005/06 to 2017/18 found the average Premier League margin fell from about 9% to about 4%.
What the Margin Actually Is
A fair set of prices would have implied probabilities that add up to exactly 100%. Real prices include a cushion on top, so the total comes out higher. That cushion is the overround, and it is the reason a market never looks perfectly balanced.
You can think of it as the price of running the market. It covers costs and gives the operator some protection against pricing errors. Competition squeezes it, which is why it varies by sport, league and market type.
How to Calculate the Margin
The maths takes three short steps. Let’s use the same illustrative match prices as in the odds explainer: Team A at 2.10, a draw at 3.40 and Team B at 3.60.
1. Convert Each Price to an Implied Probability
Divide 1 by each decimal price. That gives 47.6% for Team A, 29.4% for the draw and 27.8% for Team B.
2. Add Them Up
47.6% + 29.4% + 27.8% = 104.8%. This total is often called the overround, or the book percentage.
3. Work Out the Margin and Payout
The 4.8 extra points is the overround. Many people also express the margin as a share of the total, which is 4.8 ÷ 104.8 = about 4.6%.
The flip side is the payout percentage. Divide 100 by 104.8 and you get about 95.4%, meaning that across all outcomes the market returns roughly 95.4 for every 100 staked in total.
Sources use “margin”, “overround” and “vig” loosely, so check whether a figure is the 4.8-point version or the 4.6% version before comparing them.
Two-Way and Three-Way Markets
The same maths works for any market. A two-way over/under line priced at 1.90 on each side gives 52.6% + 52.6% = 105.3%, or a payout of 95%.
A three-way result market with prices of 1.50, 4.20 and 6.50 gives 66.7% + 23.8% + 15.4% = 105.9%. Comparing totals is the easiest way to see which of two markets carries the lower margin, since the smaller the total, the thinner the cushion.
Where the Margin Sits
The total only tells you how big the margin is, not where it lives. There is no rule that it must be split in proportion to each outcome’s probability, and in practice it often is not.
A well-known pattern called the favourite-longshot bias shows up in many betting markets. Longshots tend to be priced as more likely than they really are, while favourites are priced a little less likely than they really are. That suggests a bigger share of the margin can sit on the outsiders.
1. Basic Normalisation
This is the method used in the odds-to-probability guide: divide each implied probability by the total. It assumes the margin is spread in proportion to each outcome’s size, and it is quick to do by hand.
2. The Power Method
This method raises each implied probability to a power so that the set adds up to 1. In practice, it takes a little away from every outcome but takes proportionally more from the longshots.
3. Shin’s Method
Shin’s method comes from an economic model that assumes some participants have inside information. It was developed with horse racing in mind, and it also shifts probability away from longshots and towards favourites compared with basic normalisation.
4. Comparing the Methods
Here is the 1.50, 4.20 and 6.50 market run through all three, calculated from the same prices.
| Outcome | Raw Implied | Basic Normalisation | Power Method | Shin’s Method |
|---|---|---|---|---|
| Favourite (1.50) | 66.7% | 63.0% | 64.9% | 64.3% |
| Draw (4.20) | 23.8% | 22.5% | 21.6% | 22.0% |
| Longshot (6.50) | 15.4% | 14.5% | 13.6% | 13.7% |
| Total | 105.9% | 100% | 100% | 100% |
The differences are small, around 1 to 2 percentage points for the favourite, and no method is definitively the most realistic. Which one fits best depends on the market and the setting, so basic normalisation is a perfectly sensible default for everyday reading.
Comparing Two Sets of Prices
The total is the quickest way to compare markets. Imagine the same over/under line shown with two sets of prices, one at 1.90 on each side and one at 1.95 on each side.
| Prices | Implied Probability Each | Total | Payout |
|---|---|---|---|
| 1.90 / 1.90 | 52.6% | 105.3% | 95.0% |
| 1.95 / 1.95 | 51.3% | 102.6% | 97.5% |
The gap between 1.90 and 1.95 looks tiny, yet the margin roughly halves. That is why it pays to calculate the total rather than judge a price by eye.
Why Margins Exist
A market operator has to quote prices before the outcome is known, and those prices can be wrong. The margin gives breathing room for that uncertainty and covers the ordinary cost of operating.
It also reflects competition. When many sources offer similar markets, the pressure to tighten prices grows, and that shows in the numbers over time.
How Margins Have Changed
A study of six large bookmakers covering the 2005/06 to 2017/18 seasons found the average margin on Premier League matches fell from about 9% to about 4%. The same research found lower English divisions, including the Championship, League One and League Two, dropped from about 11% to about 6%.
Two takeaways follow. Margins are not fixed, and in that study the bigger top-flight league carried a smaller margin than the lower divisions. If you are reading prices for a top-tier fixture, the gap between raw and adjusted probabilities is typically smaller than for a niche competition.
Why Totals Differ Between Sources
Different sources can quote different prices for the same match, so their totals can differ too. A market with slightly longer prices on every outcome will have a lower total than one with slightly shorter prices.
Totals also move as prices move. If you want a fair comparison, take the figures at the same moment and from the same type of market, then compare the totals on that basis.
Why the Margin Matters When Reading Odds
If you skip the margin, every probability you calculate is a little too high. For a quick glance, that is fine. For a precise comparison with another forecast, it can create a gap that is not really there.
Suppose a model gives a team 45% and the raw implied probability from the market is 47.6%. That looks like a 2.6-point disagreement, yet after removing the margin the market figure is 45.4%, and the forecasts are almost identical.
Margins and AI Predictions
Model forecasts have no margin, because they are simply probabilities that add up to 100%. That makes them easy to line up with margin-adjusted market figures. The AI football predictions hub lists these percentages for upcoming fixtures.
For top-flight English matches, the Premier League predictions page is the natural place to start, and the Premier League profile explains the competition. For Spanish fixtures, the La Liga predictions page works the same way.
Read More: How to Use AI Football Predictions shows how to read the percentages on a prediction card.
A Quick Margin Checklist
Use this routine on any market:
- Convert every outcome to an implied probability, including the draw in a three-way market.
- Add them all up. Missing an outcome will understate the total.
- Subtract 100 for the overround, or divide 100 by the total for the payout percentage.
- Compare like with like. Two-way and three-way markets, and different leagues, can carry different margins.
Common Misunderstandings
A few ideas trip people up regularly:
- Assuming 100% is the normal total. In real markets it is always above 100%.
- Mixing up overround and margin. The 4.8-point figure and the 4.6% figure describe the same market.
- Assuming the margin is spread evenly. Longshots often carry more of it.
- Treating all markets alike. Margins vary by league, market and over time.
- Over-engineering the fix. For everyday reading, dividing by the total is usually enough.
Conclusion
The margin is the gap between the implied probabilities in a market and 100%. Add up the probabilities, subtract 100 and you have the overround. Divide by the total to turn it into a margin-adjusted forecast.
Knowing where the margin comes from, and that it may sit unevenly across outcomes, makes you a more careful reader of odds. Strip it out before comparing a market with any model and the picture gets a lot clearer.
This article is for educational purposes only and does not encourage gambling. Readers must be 18 or over, and anyone who feels gambling is affecting their life should contact a local support service.