Bookmaker odds include a built-in profit margin known as the vig or overround. This means the implied probabilities from the published odds always add up to more than 100%.
De-vigging (also called de-juicing or margin removal) is the process of removing this margin to estimate the bookmaker's underlying probability for each outcome.
It helps answer one important question:
What does the bookmaker believe the true probabilities are before adding their profit margin?
De-vigging is widely used by professional bettors, quantitative analysts, and betting models because it provides a fairer reference point for evaluating value.
Suppose a bookmaker offers odds that imply probabilities adding up to 105%. That extra 5% represents the bookmaker's expected profit.
If you compare your own probability estimates against these inflated probabilities, you may incorrectly conclude that a bet has no value.
Removing the margin allows you to compare your analysis against a much fairer estimate of the market's opinion.
The simplest and most commonly used method is proportional de-vigging.
The idea is straightforward: divide each implied probability by the total implied probability of the market.
Adjusted Probability = Raw Probability ÷ Total Raw Probability
This scales every probability proportionally until the total equals exactly 100%.
Consider the following football Match Result market:
First, convert each price into its implied probability.
Total implied probability:
47.62% + 29.41% + 27.78% = 104.81%
Now divide each probability by 104.81%.
The de-vigged probabilities become:
The adjusted probabilities now total exactly 100%, representing the bookmaker's estimated fair probabilities before adding its margin.
Once the probabilities have been adjusted, they can be converted back into decimal odds.
Fair Odds = 1 ÷ Adjusted Probability
Using the previous example:
These are the bookmaker's estimated fair odds after removing the built-in margin.
Professional analysts often use more advanced techniques than simple proportional scaling.
One of the best-known methods is the Shin model.
The Shin method assumes bookmakers do not spread their margin equally across every outcome.
Instead, they often apply a larger margin to outsiders and a smaller margin to favourites because recreational bettors tend to overvalue long shots.
The model repeatedly adjusts probabilities using an estimated parameter until the probabilities sum to exactly 100%.
Although more mathematically accurate in many situations, the Shin method requires iterative calculations and is usually performed using specialist software or statistical packages.
Removing the margin is useful in several situations.
De-vigging provides a fair benchmark before deciding whether your own estimated probability indicates value.
By removing the margin from multiple bookmakers or related betting markets, you can check whether the implied probabilities are internally consistent.
Many professional bettors compare the prices they obtained against the de-vigged closing market, using it as a measure of betting performance over time.
De-vigging removes the bookmaker's margin, but it does not guarantee that the resulting probabilities are correct.
If the bookmaker initially mispriced an event, removing the margin simply produces a margin-free version of the same incorrect opinion.
In other words, de-vigging estimates the bookmaker's view, not necessarily the event's true probability.
This is why successful bettors combine de-vigged probabilities with their own research, statistical models, and market knowledge rather than relying on bookmaker prices alone.
De-vigging removes the bookmaker's built-in margin to estimate the fair probabilities implied by the market. The simplest method scales each implied probability proportionally until they total 100%, while more advanced approaches such as the Shin method account for uneven margin allocation. Although de-vigging provides an excellent reference point for analysing value, it should always be viewed as the bookmaker's best estimate rather than an absolute measure of true probability.
