Skip to content
Independent research on sports-betting software · No sportsbook partnerships

Guide · Positive-EV betting

Implied probability and no-vig odds, explained

How to turn betting odds into implied probabilities, measure the sportsbook's margin, and remove it to estimate fair (no-vig) odds.

Reviewed by Owen DavidsUpdated Oct 28, 2026Facts checked Oct 28, 2026

On this page

Every betting price implies a probability. Add up a sportsbook's implied probabilities for every outcome of a market and you'll get more than 100%. The excess is the sportsbook's margin, often called the vig or juice. Strip it out and you get no-vig (fair) probabilities: the sportsbook's best estimate of what will happen, before its cut.

From odds to implied probability

Odds formatImplied probabilityExample
Decimal1 ÷ odds2.50 → 40.00%
American, positive100 ÷ (odds + 100)+150 → 40.00%
American, negative−odds ÷ (−odds + 100)-150 → 60.00%

The odds converter does these conversions for you.

Measuring the margin

Take a common market where both sides are priced -110:

  • Each side implies 110 ÷ 210 = 52.38%.
  • Together: 104.76%. The overround is the 4.76% above 100%.
  • The sportsbook's theoretical hold (its expected share of money wagered, if action is balanced) is 1 − 1 ÷ 1.0476 = 4.55%.

Overround and hold describe the same margin in different ways, so check which one a source means.

Removing the vig

The simplest method, and the one most calculators use, is proportional normalization: divide each implied probability by their total.

Example: Team A -150, Team B +130.

Team ATeam BTotal
Implied probability60.00%43.48%103.48%
No-vig probability57.98%42.02%100.00%
Fair American odds-138+138

The sportsbook's own estimate, without its margin, is roughly 58/42. Try it with the no-vig calculator.

Why the method matters

Proportional normalization assumes the margin is spread in proportion to each side's probability. That's an assumption, not a fact about how sportsbooks price. Many researchers use it because it's simple, but it isn't the only option: Shin's procedure, for example, models a bookmaker setting prices against a mix of ordinary bettors and better-informed "insiders," and there's an open debate about which method to use.[1]

For everyday two-way markets priced near even, the methods give similar answers. They diverge more for heavy favourites, longshots and markets with many outcomes.

Which market to de-vig

No-vig odds are only as good as the market they come from. A sportsbook with low limits that adjusts slowly can be consistently off; de-vigging its prices just gives you a precise version of a poor estimate. Positive-EV bettors usually de-vig a market they consider sharper (higher limits, quick to move) and compare other sportsbooks' prices against it. See What is positive expected value (+EV) betting?.

Sources

  1. 1.Combining historical data and bookmakers' odds in modelling football scores, Egidi, Pauli & Torelli, arXiv:1802.08848 (2018)Independent research · accessed Oct 28, 2026

About this page

Editorial review
Approved by Owen Davids on Oct 28, 2026
AI assistance
Drafted with AI assistance. Facts, sources and conclusions were checked by a human editor before publication.
Facts last checked
Oct 28, 2026

How we research and label evidence: our methodology.