Every price a bookmaker offers has a fee built into it. You never see it on a bet slip, but you pay it on every bet, win or lose. That fee is the bookmaker margin, also called the overround or the vig.
This guide shows how to calculate it, how big it really is in different football markets, and how to remove it to get the fair probabilities the odds imply. If you need a refresher on reading odds and converting them to implied probability first, start with football betting odds explained.
Data: pre-match and closing odds from football-data.co.uk · Premier League, La Liga, Bundesliga, Serie A, Ligue 1 · 2024/25 season · 1,752 matches · pre-match odds collected on Friday afternoons for weekend games and Tuesday afternoons for midweek games (source notes) · Updated: 25 October 2026
TL;DR
- The formula: add up 1 / odds for every outcome. The total is the book; anything above 100% is the overround. 2.10, 3.40 and 3.60 make a book of 104.81%, an overround of 4.81%.
- 1X2 margins (pre-match): Bet365 averaged 5.56%, William Hill 5.79% and Pinnacle 3.60%.
- Over/under 2.5 goals: Bet365 4.86%, Pinnacle 3.56%.
- Asian handicap was the cheapest market: Bet365 2.98%, Pinnacle 2.57%.
- Closing margins: Pinnacle's 1X2 margin fell to 3.01% by kick-off. Bet365's stayed at 5.62%.
- The margin is not spread evenly. Bet365 prices of 7.00 or longer won 6.6% of the time, against a proportional fair probability of 10.6%. Longshots carry more of the margin than favourites.
- Free tool: the bookmaker margin calculator does all of this for 2-way, 3-way and bigger markets.
How to calculate bookmaker margin
Each decimal price converts to an implied probability: 1 / odds. In a fair market the implied probabilities of all outcomes would add up to exactly 100%. Bookmakers shorten every price a little, so the total comes to more than 100%. That total is the book.
Here is a typical 1X2 market:
| Outcome | Odds | Implied probability |
|---|
| Home | 2.10 | 1 / 2.10 = 47.62% |
| Draw | 3.40 | 1 / 3.40 = 29.41% |
| Away | 3.60 | 1 / 3.60 = 27.78% |
| Book | | 104.81% |
From the book you get three related numbers. Sites mix them up, so it is worth knowing which is which:
| Measure | Formula | Example | What it tells you |
|---|
| Overround | book − 100% | 4.81% | How far the prices are inflated. Most sites call this "the margin". |
| Hold | 1 − 1 / book | 4.59% | What the bookmaker keeps from a book balanced to pay the same on every result. |
| Payout | 1 / book | 95.41% | What bettors get back, on average, from those stakes. |
Hold and payout always add up to 100%. In the rest of this article, "margin" means the overround unless we say otherwise.
Two rules make the calculation reliable. First, use every outcome of the same market from the same bookmaker. Leave out the draw and you will get a book below 100%, which means nothing. Second, use prices taken at the same moment, because odds move.
We took the odds for every match in the 2024/25 top five European leagues from football-data.co.uk and calculated the overround for each match, then averaged it. Pre-match odds were collected on the Friday or Tuesday afternoon before each round; closing odds are the last prices before the start.
| Market | Bookmaker | Pre-match overround | Closing overround | Matches |
|---|
| 1X2 | William Hill | 5.79% | 5.82% | 1,326 |
| 1X2 | Bet365 | 5.56% | 5.62% | 1,752 |
| 1X2 | Pinnacle | 3.60% | 3.01% | 1,752 |
| Over/under 2.5 goals | Bet365 | 4.86% | 4.81% | 1,752 |
| Over/under 2.5 goals | Pinnacle | 3.56% | 3.16% | 1,733 |
| Asian handicap (main line) | Bet365 | 2.98% | 2.61% | 1,751 |
| Asian handicap (main line) | Pinnacle | 2.57% | 2.10% | 1,748 |
Match counts are for pre-match odds. William Hill prices were missing for some matches.
Three things stand out.
The Asian handicap is the cheapest market. It carried the lowest margin at both bookmakers. At Bet365, over/under 2.5 was clearly cheaper than 1X2; at Pinnacle the two were almost level (3.56% and 3.60%). That fits the pattern described in our guide to football betting markets: two-way, heavily traded markets carry less margin. The Asian handicap is covered in more depth in our Asian handicap guide.
League barely matters in the top five. Bet365's average 1X2 overround ranged only from 5.51% (Ligue 1) to 5.65% (La Liga). Pinnacle's ranged from 3.54% (La Liga) to 3.69% (Ligue 1).
The sharp bookmaker tightens up before kick-off; the other one does not. Pinnacle's 1X2 margin fell from 3.60% to 3.01% by the close, as more information and money came in. Bet365's was 5.56% early and 5.62% at the close.
BTTS, corners, cards and player props
The public dataset only covers 1X2, over/under 2.5 and the Asian handicap, so we cannot give you market-wide averages for anything else. What we can do is show the calculation. The prices below are illustrative examples we chose to show the arithmetic, not market averages:
| Market (illustrative prices) | Odds | Book | Overround |
|---|
| BTTS yes / no | 1.80 / 1.95 | 55.56% + 51.28% = 106.84% | 6.84% |
| Total corners over / under 9.5 | 1.83 / 1.83 | 54.64% + 54.64% = 109.29% | 9.29% |
| Total cards over / under 3.5 | 1.80 / 1.90 | 55.56% + 52.63% = 108.19% | 8.19% |
| Player shots on target over / under 0.5 | 1.67 / 2.05 | 59.88% + 48.78% = 108.66% | 8.66% |
Put any real prices you see into the margin calculator and you will get the same breakdown in seconds. Expect less-traded markets to cost more than the 1X2 figures above, but check rather than assume: margins vary by bookmaker, league and even by match.
Many player props cannot be measured at all. Anytime goalscorer, "2+ shots on target" and most bet builder legs are offered on one side only. Without the "no" price there is nothing to add up, so the margin is hidden. Our article on how bookmakers set player prop lines explains why that matters.
How to remove the vig
Removing the margin turns the bookmaker's prices into fair (no-vig) probabilities that add up to 100%. There are several ways to do it.
The proportional method
Divide each implied probability by the book. For the 1X2 example above:
| Outcome | Implied | Fair probability | Fair odds | Offered odds |
|---|
| Home | 47.62% | 47.62 / 104.81 = 45.43% | 2.20 | 2.10 |
| Draw | 29.41% | 29.41 / 104.81 = 28.06% | 3.56 | 3.40 |
| Away | 27.78% | 27.78 / 104.81 = 26.50% | 3.77 | 3.60 |
The same works for two-way markets. The illustrative BTTS prices of 1.80 and 1.95 give fair probabilities of 52.00% and 48.00%, or fair odds of 1.92 and 2.08.
The power and Shin methods
The proportional method assumes every outcome carries the same share of the margin. When one side is a big favourite, that assumption matters. The power method raises each implied probability to a power k, chosen so the results add up to 100%; small probabilities shrink more. The Shin method models the bookmaker protecting itself against bettors with inside information, and also takes more margin off longshots.
Take a strong home favourite priced at 1.25, 6.00 and 12.00 (a 105.00% book):
| Method | Home | Draw | Away |
|---|
| Proportional | 76.19% (1.31) | 15.87% (6.30) | 7.94% (12.60) |
| Power | 78.63% (1.27) | 14.50% (6.90) | 6.87% (14.55) |
| Shin | 77.78% (1.29) | 15.20% (6.58) | 7.01% (14.26) |
Fair probability with fair decimal odds in brackets.
For the away side, the fair price is 12.60 or 14.55 depending on the method. If you are judging a 13.00 price, that difference decides whether it looks like value or not.
Which method fits real results?
We split every Bet365 pre-match 1X2 price from the 1,752 matches into odds bands and compared each method's fair probability with how often those outcomes actually won.
| Bet365 odds band | Prices | Implied | Proportional | Power | Shin | Actual win rate | Return at level stakes |
|---|
| Under 1.50 | 367 | 76.0% | 71.9% | 74.4% | 73.5% | 75.7% | −0.6% |
| 1.50-2.49 | 1,222 | 52.2% | 49.4% | 50.3% | 50.1% | 49.4% | −5.3% |
| 2.50-3.99 | 2,078 | 30.8% | 29.2% | 29.0% | 29.0% | 29.1% | −5.9% |
| 4.00-6.99 | 1,193 | 20.8% | 19.7% | 18.9% | 19.1% | 20.0% | −4.0% |
| 7.00 and longer | 396 | 11.2% | 10.6% | 9.3% | 9.7% | 6.6% | −38.5% |
At the extremes, power and Shin were closer to reality: short favourites won more often than the proportional method suggests, and longshots won far less often. In the middle bands, proportional was as good or better. Across all 5,256 prices the overall accuracy (Brier score) was almost identical: 0.1911 proportional, 0.1910 power and 0.1910 Shin.
Treat the longshot row with care. It is one season and 396 prices, so the 6.6% win rate has a margin of error of roughly 2.5 percentage points either way. Pinnacle's prices showed the same pattern: 5.8% actual against 10.2% proportional at 7.00 and longer.
What this means for betting
The margin is the hurdle your edge must clear. A typical Bet365 1X2 market has an overround of about 5.6%, which works out at an expected loss of about 5.3% of every stake (the hold) if the fair probabilities are right. If your own probability is only a point or two above the bookmaker's implied probability, you are not ahead; you have just priced in the margin. Use the fair probability as your benchmark, then check your edge with the expected value calculator.
Shop for prices. Taking the best price across all bookmakers for each outcome cut the average 1X2 book to 100.78% in our sample, and over/under 2.5 to 101.68%. Some of those best prices will be stale or limited, but the direction is clear: one account means paying the full margin every time.
Pick your markets by cost. A bet on the Asian handicap at Pinnacle starts about 2.5% behind (a 2.57% overround is a 2.51% hold); a Bet365 1X2 bet starts about 5.3% behind; a one-sided player prop starts an unknown distance behind. That does not make props bad bets, but the edge has to be bigger.
Be wary of longshots. Prices of 7.00 and longer lost 38.5% of stakes at Bet365 in our sample. Backing outsiders is fine when your numbers support it, but the favourite-longshot bias means the market's fair price for them is usually too generous.
Measure CLV against a no-vig close. Beating the closing odds only proves skill if you beat the fair closing price, not just the price with the margin in it. Our guide to closing line value explains how, and the CLV calculator does the arithmetic.
How to check this with StatPair
StatPair does not show odds or calculate margins inside the app; the free bookmaker margin calculator is a separate tool. StatPair gives you the other half of the comparison: your own probability.
- Open a player or team and check the last 10+ matches for shots, shots on target, fouls, cards or corners, with home/away and opponent filters.
- Count how often the stat cleared the bookmaker's line to get a hit rate.
- Compare that hit rate with the no-vig probability from the margin calculator, not the raw implied probability.
- Check the AI prediction for shots, fouls, cards or corners as a second opinion.
- Only bet when your probability beats the fair probability by more than your margin of error.
Try StatPair free for 7 days → · Free bookmaker margin calculator → · Football betting odds explained →
FAQ
Q: How do you calculate bookmaker margin?
A: Convert each outcome's decimal odds to an implied probability with 1 / odds, then add them up. The total is the book, and the amount above 100% is the overround. For 2.10, 3.40 and 3.60 the book is 47.62% + 29.41% + 27.78% = 104.81%, so the overround is 4.81%.
Q: What is a normal bookmaker margin on football?
A: In the 2024/25 top five European leagues, Bet365's pre-match 1X2 odds averaged a 5.56% overround and Pinnacle's 3.60%, across 1,752 matches. Over/under 2.5 goals averaged 4.86% and 3.56%, and the main Asian handicap line 2.98% and 2.57%.
Q: What is the difference between overround and margin?
A: Overround is the book minus 100%. Hold, often called the true margin, is 1 − 1 / book: the share of a balanced book the bookmaker keeps. For a 104.81% book the overround is 4.81% and the hold is 4.59%. Many sites use "margin" to mean the overround.
Q: How do you remove the vig from odds?
A: The simplest way is the proportional method: divide each implied probability by the book. For a 104.81% book, a 47.62% implied probability becomes 45.43%, or fair odds of 2.20. The power and Shin methods take more of the margin off longshots and fit results better when one side is a big favourite.
Q: Why can't I calculate the margin on anytime goalscorer odds?
A: Because the market is usually offered on one side only. The margin calculation needs a price for every outcome, and without a "no" price there is nothing to add up. One-sided markets hide their margin, so treat their prices with extra caution.
Bottom line
Bookmaker margin is simple to calculate and easy to forget. Add up 1 / odds across a market and the excess over 100% is what you pay. In the 2024/25 top five leagues that was about 5.6% on a Bet365 1X2 price, 3.6% at Pinnacle and under 3% on the Asian handicap at both.
Remove it before you compare a price with your own probability, prefer the power or Shin method when one side is a big favourite, and shop for the best odds. The bookmaker margin calculator does the arithmetic for any market.