Implied Probability: How to Calculate It From Football Odds
How to calculate implied probability from decimal and fractional football odds, remove the bookmaker margin and find your break-even rate.
Every football price is a probability in disguise. A home win at 2.50 is the market saying "about 40%". Once you can read odds as percentages, you can compare prices across formats, spot how much margin is built into a market, and check whether a model's view of a match differs from the price you are being offered.
This guide covers the implied probability formula for decimal and fractional odds, how to strip out the bookmaker margin (overround), how it works for accumulators, and how to use it alongside your own betting record.
What implied probability means
Implied probability is the chance of an outcome that a price represents. It is a translation of the odds, not a prediction. If the price says 40%, that is what the market charges for the outcome, not a promise of how often it will happen.
It is also your break-even rate. At decimal odds of 2.50, you need the selection to win more than 40% of the time over the long run just to avoid losing. Below that, the bet loses units in the long run, however good it felt at the time.
The implied probability formula for decimal odds
Decimal odds show the total return per unit staked, including the stake. The formula is:
Implied probability = 1 ÷ decimal odds × 100
- 2.00 → 1 ÷ 2.00 = 50%
- 2.50 → 1 ÷ 2.50 = 40%
- 4.00 → 1 ÷ 4.00 = 25%
- 1.25 → 1 ÷ 1.25 = 80%
Lower odds mean a higher implied probability; higher odds mean a lower one. In a spreadsheet, =1/B2 turns a whole column of decimal prices into probabilities.
How to calculate implied probability from fractional odds
Fractional odds, the traditional UK format, show profit relative to stake. For odds written as a/b:
Implied probability = b ÷ (a + b) × 100
- 6/4 → 4 ÷ (6 + 4) = 40%
- Evens (1/1) → 1 ÷ 2 = 50%
- 5/2 → 2 ÷ 7 = 28.6%
- 1/2 → 2 ÷ 3 = 66.7%
Notice that 6/4 and 2.50 give the same answer: they are the same price written two ways. To convert fractional to decimal, divide a by b and add 1 (6/4 → 1.5 + 1 = 2.50). If you would rather not do it by hand, the Beathem odds converter switches between decimal, fractional and American formats.
Why the percentages add up to more than 100%
In a 1X2 (home, draw, away) market, exactly one outcome must happen, so true probabilities should total 100%. Bookmaker prices don't. Take the worked example from the Beathem implied probability calculator, with prices of 1.80, 3.60 and 4.50:
- Home: 1 ÷ 1.80 = 55.56%
- Draw: 1 ÷ 3.60 = 27.78%
- Away: 1 ÷ 4.50 = 22.22%
Added together, the "book" is 105.56%. The extra 5.56% is the overround, or bookmaker margin (also called the vig). It is how the price-setter is paid, and it means every raw implied probability in that market is slightly inflated.
Implied probability vs true probability: removing the margin
To estimate what the market really thinks, divide each implied probability by the book total. This is called proportional margin removal, and it is the method the Beathem calculator uses:
- Home: 55.56 ÷ 1.0556 = 52.63%
- Draw: 27.78 ÷ 1.0556 = 26.32%
- Away: 22.22 ÷ 1.0556 = 21.05%
These "fair" probabilities now add up to 100%. The matching fair odds are the advertised price multiplied by the book sum: 1.90, 3.80 and 4.75.
Two caveats. First, proportional removal is an approximation; price-setters don't always spread the margin evenly, and other methods (such as power or odds-ratio methods) exist. Second, a margin-free market probability is still only the market's opinion. "True probability" is something nobody can observe directly; it can only be estimated by a model and then tested against results.
Using implied probability to compare a model with the market
This is where implied probability gets useful. If a model gives an outcome 50% and the margin-free market gives it 42%, the gap between them is 8 percentage points. Beathem's value predictions page works this way: it lists fixtures where the model probability sits clearly above the de-vigged price on 1X2 or Over 2.5, ranked by the largest edge.
A gap is a question, not an answer. The market may know something the model doesn't, such as team news or a late line-up change. That is why it matters whether a model's probabilities hold up over time. Beathem publishes its prediction track record for that reason. In the window 18 October to 16 November 2026 (model v2.3.0, checked on 17 November 2026), when the top 1X2 probability sat in the 50–60% band, the pick matched the result in 66.3% of 104 graded fixtures. The 60–70% band shows 88.9%, but on only 9 fixtures, which is far too small a sample to lean on. A well-calibrated 55% call should still miss roughly 45 times in 100. The methodology page explains how the probabilities are built.
Implied probability for accumulators
An acca's combined decimal odds are the legs multiplied together, so its implied probability is the legs' probabilities multiplied together. Two legs at 1.80 and 1.90:
- Combined odds: 1.80 × 1.90 = 3.42
- Implied probability: 1 ÷ 3.42 = 29.2% (the same as 55.6% × 52.6%)
Each leg also carries its own margin, and those margins compound. That is one reason long accas look generous on the slip and are hard to beat over time. This simple calculation also assumes the legs are independent; same-match legs usually aren't. The acca calculator shows combined odds and returns, and the Smart Acca builder shows model probabilities for each leg (free account required).
Check your break-even rate against your real record
Implied probability tells you the hit rate you need. Your own record tells you the hit rate you are actually getting. If you mostly back prices around 2.00, you need to win more than half of those bets to be in profit. Most people overestimate that figure from memory.
Logging every bet in the Beathem bet journal settles them automatically and tracks your hit rate and ROI. ROI is profit divided by total staked, and void or cancelled bets are left out of total staked, because no units were actually risked on them. You can also run the numbers by hand with the ROI calculator, or read our guide to tracking football bets and calculating ROI.
Quick reference
- Decimal: probability = 1 ÷ odds
- Fractional a/b: probability = b ÷ (a + b)
- Overround = sum of a market's implied probabilities − 100%
- Fair probability = implied probability ÷ book total
- Acca probability = product of the legs' probabilities (assuming independent legs)
Research only, not betting advice. 18+. Please gamble responsibly.