Why Remove the Vig?
Bookmakers don't offer fair odds. They inflate both sides of a market, ensuring that the implied probabilities sum to more than 100%. That excess — the vig — is their margin. But if you want to compare the market's estimate to your own model, you need the true probabilities, not the vig-inflated ones. Devigging is the process of removing that margin to recover the underlying probability estimates.
There are several methods, each with different assumptions about how the book applies its margin. Choosing the right method matters. A poor devigging approach can systematically misprice your edge estimates.
The Proportional (Multiplicative) Method
The simplest and most widely used approach. It assumes the book applies the same proportional margin to all outcomes. You divide each implied probability by the total sum.
Formula: True Prob_i = Implied Prob_i / Sum of all implied probs
Example: A moneyline at -150 / +130.
Implied: 60.0% + 43.5% = 103.5%
True: 60.0/103.5 = 58.0% and 43.5/103.5 = 42.0%
You don't have to do the arithmetic by hand. Drop the prices into the No-Vig Calculator and it returns both the proportional and power-method estimates side-by-side.
Pros: Simple, fast, no parameters to estimate.
Cons: Assumes equal proportional margin on favorites and underdogs, which is often false. Books typically apply more margin to underdogs.
The Power (Odds Ratio) Method
The power method assumes the book applies margin by raising the true probabilities to a common power. This produces different effects on favorites and underdogs that better match observed bookmaker behavior.
Formula: True Prob_i = Implied Prob_i^k, where k is chosen so that the true probs sum to 1.
Finding k requires numerical methods — you can't solve it directly. But with a simple root-finding algorithm (or a spreadsheet's Goal Seek), it's straightforward.
For the same -150 / +130 line, the power method with k ≈ 0.96 gives:
True: 58.5% and 41.5%
Notice how the power method gives slightly more probability to the favorite than the proportional method. This is generally more accurate because books tend to compress probabilities toward 50% by adding relatively more margin to underdogs.
Pros: More realistic than proportional. Better handles extreme favorites.
Cons: Requires numerical solving. Can produce counterintuitive results with three-way markets.
The Shin Method
The Shin method, developed by economist Hyun Song Shin, assumes the bookmaker's margin comes from insider information — some bettors know something the book doesn't. The book adds margin to protect against this asymmetric information.
Formula: True Prob_i = sqrt(z * Implied Prob_i^2 + (1-z) * Implied Prob_i), where z is solved so that true probs sum to 1.
This is more complex than the power method but captures an important reality: books face adverse selection from informed bettors, and their margin structure reflects this.
For the same -150 / +130 line, Shin gives results very close to the power method: approximately 58.4% and 41.6%.
Pros: Theoretically grounded. Best for markets with significant insider information (horse racing, some soccer markets).
Cons: More complex. Overkill for efficient markets like NFL sides.
Wisdom of the Crowds (Additive) Method
The additive method simply subtracts an equal amount of margin from each outcome.
Formula: True Prob_i = Implied Prob_i - (Total Overround - 1) / n
For our example: 60.0% - 1.75% = 58.25% and 43.5% - 1.75% = 41.75%.
This method is generally inferior. It can produce negative probabilities for heavy underdogs and has no theoretical justification. Most bettors should avoid it.
Which Method Should You Use?
For most bettors, the power method strikes the best balance between accuracy and complexity. It produces better estimates than proportional devigging with only modest additional effort.
However, the difference between methods is often small in practice. For moneylines where both sides are between -200 and +200, all methods produce similar results. The choice matters more for heavy favorites, three-way markets, and props with many outcomes.
A pragmatic approach: use the power method as your default, but don't obsess over the difference. The error in your probability estimates from model uncertainty almost always dwarfs the error from devigging method choice.
Devigging in Practice
Here's a workflow for using devigged probabilities:
- Get the best available odds across your books.
- Convert to implied probabilities.
- Apply your chosen devigging method to get true probability estimates — the No-Vig Calculator handles steps 2 and 3.
- Compare to your model's probability estimate.
- If your estimate exceeds the market's by your minimum edge threshold, calculate Kelly stake (use the Kelly Calculator) and place the bet.
"The vig is not your enemy. It is the cost of doing business. The question is not whether the vig exists but whether your edge exceeds it."
