Methodology
- Convert each side of the market from odds to implied probability.
- Add both implied probabilities to measure the overround.
- Divide each side by the total implied probability to estimate no-vig share.
- Compare the no-vig probability to your model projection and minimum edge threshold.
Example output
Two-way no-vig example
A fair probability estimate from a common spread price.
| Side | Book price | No-vig estimate |
|---|---|---|
| Team A spread | -110 | 50.00% |
| Team B spread | -110 | 50.00% |
| Team A moneyline | -125 | 53.19% |
| Team B moneyline | +105 | 46.81% |
No-vig creates a benchmark; it does not replace matchup, injury, and limit analysis.
Why no-vig matters
Sportsbooks build margin into prices. Without removing that margin, a bettor may mistake hold for market confidence or compare a model against a distorted number.
- Turns -110 / -110 into approximately 50% / 50%
- Helps identify whether a listed price is actually better than fair
- Creates a common baseline for model marketplaces and reports
- Works best on liquid markets with competitive pricing
Where no-vig can mislead
No-vig prices are estimates, not truth. Same-game props, long-shot markets, and stale lines can include asymmetric margin or limits that make simple normalization less reliable.
- Use multiple books when estimating market consensus
- Be careful with futures and long-shot props
- Do not treat no-vig as a standalone betting signal
- Require a cushion between model price and fair market price
Responsible-use note
Analytics should support disciplined decision-making, not guaranteed outcomes. Bet only where legal, never risk money you cannot afford to lose, and use limits before volume increases.
FAQ
Is no-vig probability the same as true probability?
No. It is an estimate derived from market prices after removing margin. True probability is unknown before the event.
Can no-vig math be used for props?
Yes, but props are often less liquid and more sensitive to stale data. Use extra caution and compare multiple books when possible.
What edge threshold should I require?
That depends on market quality, model confidence, and bankroll rules. Many users require a larger cushion for lower-liquidity markets.