ALT: Woman using a betting app on her phone from a VIP stadium lounge during a football match.
Three decimal prices can reveal more than the potential payout attached to each football outcome. In the worked market, odds of 2.10 for the home win combine with 3.40 for the draw and 3.60 for the away win to produce raw implied probabilities totalling 104.81%. The excess above 100% is a simple overround of 4.81 percentage points. The same distinction between quoted price and implied probability also appears in online casino information, although the calculation here concerns a three-way football market. Four stages turn the displayed odds into a comparable margin figure without treating that figure as a prediction of which team will win.
Step 1: One price snapshot fixes the market for calculation
The three prices belong to the same market and work best as one simultaneous snapshot. If the home price is recorded at one moment and the draw price several minutes later, the market may already have moved before the away price is captured. The resulting calculation would then combine figures that never existed together.
That matters because the overround describes the structure of a particular set of quoted prices. Even a small change in one outcome alters the implied-probability total.
Decimal format makes the next stages straightforward. A price of 2.10 already expresses the total potential return per unit staked. For margin analysis, however, the important information is the probability implied by that number rather than the payout itself.
A spreadsheet can begin with the three decimal prices in one column. Keeping the timestamp alongside the market snapshot also makes later comparisons cleaner when several matches or different moments are being analysed.
Step 2: Inverse odds produce the raw probabilities
The basic conversion is:
1 ÷ decimal odds × 100
For the home win at 2.10, the calculation is:
1 ÷ 2.10 × 100 = 47.62%
The draw at 3.40 gives approximately 29.41%. The away price of 3.60 produces about 27.78%.
These are raw implied probabilities. They are useful because they put all three prices onto the same percentage scale, but they do not yet form a complete probability distribution. Adding them gives more than 100%.
The full example looks like this:
| Outcome | Decimal odds | Raw implied probability | Normalised probability |
| Home win | 2.10 | 47.62% | 45.43% |
| Draw | 3.40 | 29.41% | 28.06% |
| Away win | 3.60 | 27.78% | 26.50% |
| Total | — | 104.81% | 100.00% |
The displayed normalised rows add to 99.99% because each figure has been rounded to two decimal places. Using the unrounded values produces a total of 100%.
In spreadsheet form, a raw-probability cell can contain =1/B2 when the decimal price sits in B2. Formatting that result as a percentage converts the decimal fraction into the familiar percentage display.
Step 3: The combined total reveals the 4.81% overround
Adding the three unadjusted probabilities gives approximately:
47.62% + 29.41% + 27.78% = 104.81%
The amount above 100% is the simple overround:
104.81% − 100% = 4.81 percentage points
In a spreadsheet, the market total can be calculated with =SUM(C2:C4) when the raw probabilities occupy those cells. The overround then follows from =C5-1 if C5 contains the total in decimal form rather than percentage points.
The distinction between the 104.81% booksum and the 4.81% overround matters. One describes the complete raw total. The other describes the excess above a probability distribution that would sum to exactly 100%.
That 4.81% figure is also not automatically the expected loss attached to every possible selection. The built-in margin may not be distributed proportionally across all three prices. Research on football markets has found that simple inverse-odds methods can retain biases between shorter and longer prices.
A smaller calculated overround can therefore help compare market structures, but it does not identify the winning outcome. It also does not establish that a particular selection offers a favourable return.
The same separation between displayed probability information and payout structure can apply to casino games, although those products use different mechanics from a three-way football market.
Step 4: Normalisation rescales the market to 100%
Normalisation removes the excess proportionally so that the three outcomes form a 100% distribution.
The formula for each outcome is:
raw implied probability ÷ combined raw total
For the home win:
47.62 ÷ 104.81 ≈ 45.43%
The draw falls from 29.41% to about 28.06%. The away outcome moves from 27.78% to about 26.50%.
In a spreadsheet, a formula such as =C2/$C$5 divides the home probability by the fixed market total. Copying the same structure to the remaining outcome rows produces the other normalised values.
This rescaling is useful when two sets of three-way prices need comparison on the same 100% basis. Raw inverse odds cannot provide that directly because their total includes the margin.
Proportional normalisation still has limits. It assumes that the excess probability can be removed from every outcome in proportion to its raw implied probability. Real pricing may distribute margin differently, so the adjusted percentages are analytical estimates rather than objectively correct forecasts.
Two-way markets deserve separate treatment for the same reason. A market containing only two mutually exclusive outcomes has a different structure from a home-draw-away market. Comparing headline overrounds without acknowledging that difference can make unlike markets appear directly equivalent.
Rounding adds another small complication. Calculations performed with full precision can give a slightly different displayed total from calculations based only on percentages rounded to two decimal places. Spreadsheets reduce that problem when the underlying formulas retain the original decimal values.
The completed four-step calculation therefore answers a narrow but useful question: how much implied probability sits above 100% in one quoted three-way market, and what do the prices look like after proportional normalisation? In this example, the answer is a 4.81-point overround with normalised probabilities centred around 45.43% for the home outcome.
That result improves transparency around the market calculation. It does not turn the normalised percentages into match predictions, and it cannot create a guaranteed positive return. The figures describe how the quoted prices fit together at one moment; the football result remains a separate question.
