Divide 1 by decimal odds and you get the implied probability. Add those up across a market and the total passes 100%. The excess is the margin. Strip it out and you get margin-free probabilities and margin-free odds, which are still estimates. Set a price against a probability estimated elsewhere and you get a theoretical value. Set the price you took against the final price and you get the closing line value.
The journey in short
Odds are prices, and this whole guide is about reading them. You convert a price into a probability, take out the margin it carries, then compare it with an independent estimate or with the final price. Each step answers a different question. None of them tells you who wins.
Seven ideas come up, in the order you use them: the price itself, the implied probability, the margin, the margin-free probability, the fair price, the theoretical value and the closing line value. The glossary below gives each one a line.
These seven ideas describe a price and how it moves. None of them turns odds into a tip, and none of them makes a bet "safe".
What are odds?
Odds are the price at which an operator agrees to take on a risk. In decimal format they show what a winning stake returns, stake included. Odds of 2.20 return £22 on a £10 stake, so £12 of net profit.
That price isn't a neutral reading of the sport. It mixes three things: the bookmaker's own estimate, the margin it wants to keep and adjustments for its exposure, meaning how the stakes already taken are spread. Two operators can show different prices for the same match without either being wrong. They don't share the same estimate, the same exposure or the same commercial policy. Odds movements are the story of those adjustments.
| Idea | What it measures |
|---|---|
| Odds | The price displayed for an outcome |
| Implied probability | What that price assumes, margin included |
| Book percentage | The sum of the implied probabilities of a market |
| Overround | The part of the book percentage above 100% |
| Margin-free probability | The implied probability once the overround is spread out |
| Fair odds | The odds that match an estimated probability |
| Closing line value | The gap between the price taken and the closing line |
How do you turn odds into a probability?
Divide 1 by the decimal odds, then multiply by 100: (1 ÷ odds) × 100. Odds of 2.00 give 50%, odds of 1.80 give 55.56% and odds of 2.20 give 45.45%.
The conversion is mechanical. It needs no sports data, only the price. That's its strength and its limit. It translates faithfully what the price assumes, but it corrects neither the margin nor any estimation error by the operator.
The notation changes nothing. Odds of 2.50 are written 3/2 in fractional and +150 in American. Three notations, one price, the same 40%. The odds formats page has the detail.
| Decimal | Fractional | American | Implied probability |
|---|---|---|---|
| 1.50 | 1/2 | −200 | 66.67% |
| 1.80 | 4/5 | −125 | 55.56% |
| 2.00 | 1/1 | +100 | 50.00% |
| 2.20 | 6/5 | +120 | 45.45% |
| 2.50 | 3/2 | +150 | 40.00% |
| 3.40 | 12/5 | +240 | 29.41% |
| 5.00 | 4/1 | +400 | 20.00% |
For the full formula, a table of examples and a calculator, head to implied probability.
Why does the total go over 100%?
Because odds carry a margin. Add up the implied probabilities of every outcome in a market. The total, called the book percentage, is almost always above 100%. The excess is the overround.
Take a two-outcome market priced at 1.80 and 2.00:
1 ÷ 1.80 = 55.56%;1 ÷ 2.00 = 50.00%;- total = 105.56%, an overround of 5.56 points.
That 5.56 isn't a profit. It's the excess sitting in the prices at one moment. What the operator really keeps depends on the stakes it takes, how they're spread across the outcomes, the results and its costs. The full calculation, the gap between overround and theoretical margin and the reasons margins vary from one market to another are on the bookmaker margin page.
How do you take the margin out of a market?
Divide each implied probability by the market's total. It's the most common method. The total comes back to exactly 100% and each outcome gets what's called a margin-free probability.
Take a 1X2 market priced at 2.10 / 3.40 / 3.60. The implied probabilities are 47.62%, 29.41% and 27.78%, a total of 104.81%. After proportional normalisation they become 45.43%, 28.06% and 26.50%. The matching margin-free odds are 2.20, 3.56 and 3.77.
This method spreads the excess in proportion to each probability. Other approaches exist (power, logarithmic, Shin, favourite-longshot bias corrections) and give different results on the same prices, especially for big longshots. So a margin-free probability is never the probability of the event. It's an estimate that depends on the observed prices and on the method. The full method and its limits are on margin-free odds.
What are fair odds?
Fair odds are the odds that would match an estimated probability exactly: fair odds = 1 ÷ estimated probability. Estimate 50% and the fair odds are 2.00.
People often mix them up with margin-free odds. Margin-free odds come from one operator's prices once its overround is removed. Fair odds come from a probability that arrives from elsewhere: a model, a consensus of several sources, your own analysis. The two can coincide, but they answer different questions. I keep the two strictly apart: margin-free odds only clean up a price, while fair odds bring in an outside estimate. The fair odds page covers the conversion in all three formats, and what happens when fair odds are simply wrong.
What is theoretical value?
Theoretical value appears when the estimated probability of an outcome is higher than the probability needed to justify the odds on offer. With odds of 2.20 and an estimate of 50%, the theoretical gap is 2.20 × 0.50 − 1 = +10%.
You need two ingredients: the odds and an estimate that's independent of them. The bookmaker's price alone isn't enough, since you'd be comparing a price with itself. And because everything rests on the quality of the estimate, a positive gap can be nothing more than an over-optimistic estimate. The value bet page explains where that estimate comes from, why it can be wrong and what the calculation doesn't prove.
What is closing line value?
Closing line value compares the price you got with the last price shown before the match starts. Odds taken at 2.20 against a close of 2.00 give 2.20 ÷ 2.00 − 1 = +10%.
The closing price is the last one available, so it holds the most information. That doesn't make it exact. A closing market is still a market, with its margin, its liquidity and its errors. A positive CLV only says you took a price above the final one. It guarantees neither a profit on that bet nor profitability over time. The closing line value page covers the two calculation methods and how to pick the closing line.
What OddScore does with all this
OddScore runs this journey continuously, across several bookmakers and successive snapshots. It converts odds into probabilities, removes each operator's margin before any comparison, then tracks how those prices move until kickoff.
The formulas fit in three lines. What takes time is repeating them across dozens of operators and thousands of markets at regular intervals, and no manual reading keeps up. The AI reading summarises what the market says before the match. It isn't a bet to follow, and OddScore gives no staking advice.
Where to start
New to betting altogether? Sports betting for beginners covers the market, the stake and the settlement that everything above takes for granted.
Here are five steps, in order, so you don't get lost.
- Convert a price. Start with implied probability. It's the one essential calculation, and everything else hangs off it.
- Measure the margin. Go to bookmaker margin to see why the total passes 100% and what the overround measures.
- Remove the margin. Continue with margin-free odds to compare two operators on common ground.
- Estimate a fair price. Carry on with fair odds, then value bet if you have an independent estimate.
- Check the price you got. Finish with closing line value, which sets your price against the market's last one.
Two tools help along the way, no account or sign-up needed: the odds converter to switch between formats and the bookmaker margin calculator to analyse a full market.
To use these probabilities when analysing a match, read odds and predictions.
Keep this in mind on every page. These calculations show what a price assumes, they help you estimate a gap between two prices or between a price and an estimate, and they never say an outcome will happen.
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.
Implied probability: converting it from odds
The formula, worked examples and a calculator to go from odds to their implied probability.
Convert oddsMargin-free odds: the proportional method
How to spread the overround across the outcomes, and what the result can't tell you.
Remove the marginFair odds: from an estimated probability to odds
Going from an estimated probability to fair odds, in all three formats.
Calculate fair oddsBookmaker margin: definition and calculation
Book percentage, overround, theoretical margin: the measures of the excess built into prices.
Understand the marginArbitrage between bookmakers
Where a price gap between operators comes from, and why it closes fast.
Understand the gapFrequently asked questions
Are odds a prediction?
No. Odds are a price set by an operator. They reflect its commercial estimate of the market, adjusted for the stakes it receives and the risk it's willing to carry. They don't announce a result.
How do you turn odds into a probability?
With decimal odds, divide 1 by the odds, then multiply by 100. Odds of 2.00 correspond to 50%, odds of 2.20 to 45.45%.
Why does the sum of the probabilities go over 100%?
Because odds carry a margin. The total of the implied probabilities, the book percentage, sits above 100%. The part above is the overround.
What are margin-free odds?
They're the odds you get once the margin is spread across the outcomes, most often by proportional normalisation. The result is still an estimate, and it depends on the method you pick.
What's the difference between margin-free odds and fair odds?
Margin-free odds come from the prices one operator displays. Fair odds come from an estimated probability, which can come from a model, a consensus or your own analysis.
Does a positive theoretical value guarantee a profit?
No. It only shows a gap between a price and an estimated probability. If the estimate is wrong, so is the gap, and a bet with a positive theoretical value can easily lose.
What's the closing line for?
It's a benchmark. It shows where the market ended up putting the price. Comparing it with the price you took tells you whether your price was good. It says nothing about whether a strategy is profitable.
Sources and method
This page draws on the standard formulas for converting decimal odds, on economic research into the overround and price formation in betting markets, on the usual ways of normalising probabilities and on OddScore's own odds-analysis method. The Gambling Commission and BeGambleAware links are official reference points.
- Convert each decimal price into an implied probability (1 ÷ odds).
- Add those probabilities to get the book percentage, then subtract 100% to get the overround.
- Normalise the probabilities proportionally to estimate a margin-free market, remembering that the result depends on the spreading method.
- Compare the odds with an independently estimated probability to get a theoretical value, and with the closing line to judge the price you got.