Bookmaker Margin Explained: How the Overround Works

A bookmaker does not have to predict the result to make money. Price a market so the outcomes' implied probabilities add up to more than 100%, and the book keeps the excess whichever way the game goes. That excess is the margin, the built-in expected profit sitting inside every price you see.

What the margin is

When a bookmaker prices a two-way market, they shorten the true odds on each outcome. An implied probability is a price read as a probability: 1 divided by decimal odds. Odds of 2.00 imply a 50% chance.

A fair, margin-free market would have all implied probabilities summing to exactly 100%. Real markets never do. The excess is the margin, and it guarantees the bookmaker takes more in stakes than it pays out to winners.

The same concept goes by other names. Overround, vigorish, and juice describe the same edge from slightly different angles. Margin, calculated via the implied-probability method, is the practical number to compute and compare across markets, so this article uses that term throughout.

Once you can calculate it for any market, you have the metric that separates sharp prices from soft ones.

Implied probability

Converting a price to a probability depends on which odds format you're reading. Each format packages the same information differently. The margin calculation in the next sections uses decimal odds, but you can convert from any format first and the result is identical.

For decimal odds, divide 1 by the odds:

Implied probability = 1 / decimal odds

Example: 2.10 → 1 / 2.10 = 47.6%

Fractional odds divide the denominator by the sum of numerator and denominator. The fraction 11/10 means you win 11 units for every 10 staked:

Implied probability = denominator / (numerator + denominator)

Example: 11/10 → 10 / (11 + 10) = 10 / 21 = 47.6%

American odds split on whether the line is negative or positive. Negative lines (favorites) use the absolute value divided by itself plus 100; positive lines (underdogs) use 100 divided by the odds plus 100:

Negative line:  |odds| / (|odds| + 100)
Positive line:  100 / (odds + 100)

Example: −110 → 110 / (110 + 100) = 110 / 210 = 52.4%

All three formats encode the same information. Pick the format your book displays and apply the matching formula before summing. The resulting implied probability is what you carry into the margin calculation. The original format no longer matters once you have it.

The formula

Three steps:

Step 1: Implied probability = (1 / decimal odds) × 100
Step 2: Sum the implied probabilities for all outcomes in the market
Step 3: Margin % = sum − 100

The formula works for any number of outcomes. Add as many terms as the market has, then subtract 100. A tennis match uses two terms; a football 1X2 market uses three; a horse race with 12 runners uses twelve.

Margin % = [ (1/odds_A) + (1/odds_B) + ... + (1/odds_N) ] × 100 − 100

A two-way market

Take a money-line market with two outcomes, priced at 1.926 and 2.02:

Option A:  1/1.926 × 100 = 51.92%
Option B:  1/2.02  × 100 = 49.50%

Sum = 101.43%
Margin = 101.43% − 100% = 1.43%

That 1.43% is a tight line, characteristic of a sharp book on a major market. You can verify this yourself on any book, on any event, in under a minute.

A softer two-way line of 1.90 / 2.00 produces a sum of 102.63%, a 2.63% margin. Lower odds on both sides means a bigger margin, with the bookmaker taking more from each price simultaneously.

A three-way market

Football 1X2 markets carry higher margins than two-way markets because the bookmaker spreads that margin across three prices. A market priced at Home 2.10 / Draw 3.60 / Away 3.40:

Home:  1/2.10 × 100 = 47.6%
Draw:  1/3.60 × 100 = 27.8%
Away:  1/3.40 × 100 = 29.4%

Sum = 104.8%
Margin = 104.8% − 100% = 4.8%

Each outcome looks reasonably priced in isolation. Summed, the book takes 4.8% before you've placed a bet. The margin is built in from the moment the bookmaker sets the odds. It is a structural feature of every three-way market, not a rounding error.

Margin vs vigorish

Margin and vig measure the same bookmaker edge, but from different reference points. Margin is profit relative to payouts; vig is profit relative to turnover (total stakes taken). Because payout is always less than turnover on a winning book, margin is slightly larger than vig.

For practical comparison, use the margin. You can compute it directly from any set of odds without knowing how the book balanced.

Expected profit on turnover from a book of O%:

Expected profit % = (O − 100) / O

Example: 104.8% book → (104.8 − 100) / 104.8 ≈ 4.58% expected profit per unit staked

That 4.58% is the structural headwind on every bet in that market. It compounds over every wager you place.

How high margins run in practice

Sharp books (Pinnacle, Betfair Exchange) run tight margins, commonly cited at 1.5-3% on top markets and up to about 2-4% on football. Recreational and soft books build in much larger margins, commonly cited at 6-8% and sometimes higher, with figures of 8-12% appearing on some books and markets.

By market tier, typical overrounds break down roughly as follows:

Accumulator margins compound across legs because each leg carries its own margin and they multiply together.

Margin is the best way to compare two odds on the same outcome. Shopping the same bet across books and taking the lower-margin price is the most direct way to reduce the structural headwind on every bet. Over hundreds of wagers, the difference between paying a 2% margin and an 8% margin compounds into a significant gap in returns.

Removing the margin

Before you can find value, you need the fair price: the odds stripped of the bookmaker's cut. That process is called devigging. The simplest method is proportional devigging: divide each outcome's implied probability by the total sum of all implied probabilities, then convert back to decimal odds.

Fair probability = (1 / odds) / sum_of_implied_probabilities

Two-way example: 1.926 / 2.02 → sum = 1.0143
  Fair prob A = 0.5192 / 1.0143 = 51.19%  → Fair odds A = 1 / 0.5119 = 1.953
  Fair prob B = 0.4950 / 1.0143 = 48.81%  → Fair odds B = 1 / 0.4881 = 2.049

The two fair odds now sum to exactly 100% implied probability. The margin is gone.

Two other methods are worth knowing. The additive method removes an equal slice from each outcome's implied probability rather than scaling proportionally; it behaves differently when the book is lopsided. The Shin method applies a correction derived from insider-trading models and is considered more theoretically sound on binary markets.

For most practical purposes, proportional devigging is accurate enough and the easiest to compute by hand.

From fair price to edge

Once you have the fair probability, you can compare it to any offered price. If your model assigns a higher probability than the devigged fair probability implied by the book, you have a reason to look harder. If an offered price is better than the fair price, you have edge.

The edge formula expresses this directly:

Edge = (model_prob × offered_odds) − 1

Example: model assigns 55% probability to an outcome priced at 2.10
  Edge = (0.55 × 2.10) − 1 = 1.155 − 1 = 0.155 → +15.5%

A positive edge means the odds are too generous relative to your model. A negative edge means the bookmaker's price is too short. Margin is what the bookmaker takes; edge is what you return above it.

Which is why the margin figure you started with is not just background. Every point of margin baked into a price is a point the edge has to clear before it means anything, so the routine that pays off is dull and repeatable: compute the margin on both sides of a bet, take the price with the lower one, and only then ask whether there is any edge left worth backing.