Betting maths guide
What Are No-Vig Fair Odds?
No-vig fair odds are an estimate of what a market's prices might look like after removing the bookmaker's overround. They are useful for understanding market structure, but they are not the same thing as the true probability of an outcome.
Start with implied probability
Decimal odds can be converted into an implied probability by dividing 1 by the odds. Decimal odds of 2.00 imply 50%; odds of 4.00 imply 25%.
Implied probability = 1 ÷ decimal odds
In a market with several outcomes, add the implied probabilities together. If the total exceeds 100%, the market contains an overround.
For example, a three-way football market priced at 2.10, 3.40 and 3.60 produces implied probabilities of 47.62%, 29.41% and 27.78%. Together they total 104.81%.
What “removing the vig” actually means
One common method is proportional normalisation. Each implied probability is divided by the total book percentage so that the adjusted probabilities add to exactly 100%.
Raw implied probability = 1 ÷ odds
Total book = sum of all raw implied probabilities
No-vig probability = raw implied probability ÷ total book
No-vig decimal odds = 1 ÷ no-vig probability
This removes the overround proportionally. It is transparent and reproducible, which is why Betting Maths uses it in its basic no-vig calculator.
Two-outcome example
Suppose both outcomes are priced at 1.91. Each price implies about 52.36%, so the total book is about 104.71%.
| Outcome | Odds | Raw implied probability | No-vig probability | No-vig odds |
|---|---|---|---|---|
| A | 1.91 | 52.36% | 50.00% | 2.00 |
| B | 1.91 | 52.36% | 50.00% | 2.00 |
Because the market is perfectly symmetrical, proportional normalisation returns a 50/50 fair-price estimate.
Three-outcome example
Now take the football prices 2.10, 3.40 and 3.60.
| Outcome | Odds | Raw implied | No-vig probability | No-vig odds |
|---|---|---|---|---|
| Home | 2.10 | 47.62% | 45.43% | 2.20 |
| Draw | 3.40 | 29.41% | 28.06% | 3.56 |
| Away | 3.60 | 27.78% | 26.50% | 3.77 |
The raw probabilities total 104.81%. After normalisation they total 100%. The adjusted odds are therefore longer than the quoted odds because the overround has been removed.
No-vig probability is not automatically “true probability”
This distinction matters. A no-vig calculation starts with bookmaker prices and mathematically rescales them. It does not independently estimate how likely each event really is.
The market may be informative, but it can still contain bias, uneven margin allocation, stale prices, differing limits, liquidity effects or bookmaker-specific opinions. Calling the result “fair odds” is shorthand for fair under the chosen margin-removal method.
Does every bookmaker spread margin proportionally?
No-vig calculations often assume proportional margin allocation because it is simple and transparent. Real bookmakers do not have to price markets that way.
A bookmaker can shade one outcome more heavily than another. Favourite-longshot effects, liability management, customer behaviour and market-making choices can all make the true margin distribution uneven.
That means two different vig-removal methods can produce slightly different fair-probability estimates from the same quoted prices. The correct interpretation is therefore an estimate, not a revealed fact.
Book percentage, overround and no-vig odds are different things
| Term | What it describes |
|---|---|
| Implied probability | The percentage implied by one quoted price. |
| Book percentage | The total of all implied probabilities in the market. |
| Overround | The amount by which the book exceeds 100%. |
| No-vig probability | An adjusted probability after removing overround using a stated method. |
| No-vig odds | The decimal odds corresponding to that adjusted probability. |
What real Premier League data looked like
Betting Maths has also measured overround in observed markets rather than relying only on examples. In our 10-match Premier League opening-weekend sample for 2025/26, cross-bookmaker average opening prices produced an average book of 105.56%, while combining the best available listed price for each outcome reduced the average book to 101.74%.
That research is useful context for this guide because it shows that overround is not just a textbook idea: it is measurable in recorded markets, and the size of the book changes depending on which prices are used.
No-vig odds and expected value are not the same calculation
No-vig odds remove margin from the market prices you supply. Expected value asks what happens when a price is compared with a probability estimate.
You can use a no-vig probability as one benchmark, but doing so does not prove that benchmark is the true probability. A genuinely independent EV calculation requires a probability estimate you are prepared to defend.
When to use the calculator
Use the No-Vig Fair Odds Calculator when you want to understand how a simple two-outcome market looks after proportional margin removal. For larger markets, the same normalisation idea can be applied across all outcomes.
Use the result as an educational benchmark, not as proof that a selection is mispriced.
