A sportsbook's prices add up to more than 100% — the excess is its margin, or vig. Removing it gives the fair probability of each outcome, and there is more than one way to remove it. Here are the four methods TrueEdge supports, the formula for each, and what they give on the same market.
Every price implies a probability: p = 1 ÷ decimal odds. American −150 is decimal 1.667, so p = 60.00%; +130 is decimal 2.30, so p = 43.48%. Together they make 103.48%, and the 3.48 points over 100 is the book's margin. Each method below decides how to take those points back out.
Scales every implied probability down by the same factor. The standard method, and the one most calculators use.
Takes the same number of percentage points off every outcome. Breaks down for extreme longshots, where it would go below zero; we fall back to multiplicative there.
Raising each probability to a power above 1 shrinks small probabilities proportionally more, so the longshot gives up more of the margin.
Shin (1992) models the margin as the book's protection against better-informed bettors. Like power, it pushes more of the margin onto the longshot.
pᵢ is the implied probability of outcome i, Σp their total and n the number of outcomes. Power and Shin have no closed form for k and z; both are found numerically.
On a market with a modest favourite the methods land within a couple of cents of each other. On a heavy favourite they spread out, because they disagree about how much of the margin sits on the longshot.
| Method | Favourite fair | Underdog fair |
|---|---|---|
| Multiplicative | 57.98% (-138) | 42.02% (+138) |
| Additive | 58.26% (-140) | 41.74% (+140) |
| Power | 58.40% (-140) | 41.60% (+140) |
| Shin | 58.26% (-140) | 41.74% (+140) |
| Method | Favourite fair | Underdog fair |
|---|---|---|
| Multiplicative | 76.19% (-320) | 23.81% (+320) |
| Additive | 77.50% (-344) | 22.50% (+344) |
| Power | 78.24% (-359) | 21.76% (+359) |
| Shin | 77.50% (-344) | 22.50% (+344) |
The no-vig calculator does this for one market you paste in. TrueEdge does it for every market across US sportsbooks and exchanges, and lists the bets priced better than fair.
None is right for every market. Multiplicative is the standard and is fine when both sides are close in price. When one side is a heavy favourite, books tend to load more of their margin onto the longshot, and the power and Shin methods take more off the longshot to match. On -400 / +300 the fair underdog price is +320 multiplicative, +344 Shin and +359 power.
On a two-way market they always do: with only two outcomes Shin's formula reduces to splitting the margin equally between the sides. They differ once a market has three or more outcomes.
The +EV feed's fair price is multiplicative by default. Each +EV row also carries a second, longshot-adjusted fair value computed with Shin, so a bet that only looks +EV under the gentler method can be told apart from one that is +EV under both.