How to Calculate No-Vig Odds and True Probability

No-vig odds — also called fair odds or true odds — are what a two-sided betting market would show if the sportsbook removed its built-in margin entirely. Every posted price already has a small profit cushion baked in, which means the implied probabilities of both sides of a market always add up to slightly more than 100%. Calculating the no-vig number strips that cushion back out, leaving a pair of probabilities that add up to exactly 100% and a matching pair of “fair” prices.

This matters because a sportsbook’s posted price is not a neutral read of an outcome’s chances — it is a business price with a margin attached. A bettor trying to judge whether their own probability estimate for a game is actually better than what the market is offering needs to compare it against the fair, de-margined number, not the raw posted odds. This article walks through the exact three-step formula for finding no-vig probability and fair odds from any two-sided market, with two complete worked examples — one for an evenly priced market and one for an uneven one — so the method generalizes to any odds a reader is looking at.

What “No-Vig” or “Fair” Odds Actually Mean

Every single price on a bet slip converts into an implied probability using the standard odds formulas — a favorite’s negative American odds divide into a probability above 50%, an underdog’s positive odds into one below 50%. In a truly fair market, the implied probabilities of every possible outcome would add up to exactly 100%, because one of those outcomes has to happen. A two-way market — Team A wins or Team B wins — has only two implied probabilities to check against that 100% line.

In practice, a sportsbook’s actual posted odds always imply slightly more than 100% combined. That extra percentage above 100% is the vig, already covered in detail in how vig is calculated as a sportsbook’s margin. No-vig odds — sometimes written “de-vigged” odds — are what remains once that extra percentage is mathematically distributed back out and removed, so the two sides’ probabilities land on exactly 100% again, in the same proportion to each other as before.

The result is not a prediction of what will happen in the game. A no-vig probability is a mathematical baseline — the market’s price with its business margin subtracted — useful specifically for comparing against a bettor’s own probability estimate or against a different sportsbook’s price on the same game.

The Three-Step Formula for Removing the Vig

The full procedure works the same way for any two-way market — moneyline, point spread, or total — regardless of sport, and it only requires the posted odds on both sides. It builds directly on how to read American odds and convert them into implied probability, so that conversion should already feel familiar before working through the steps below.

Step 1: Convert Each Side to Raw Implied Probability

Convert every side of the market into implied probability separately, using the standard formulas: for a favorite, divide the absolute value of the odds by that value plus 100; for an underdog, divide 100 by the odds plus 100. A two-way market produces exactly two numbers at this step.

Step 2: Add Both Probabilities Together

Sum the two raw implied probabilities from Step 1. Whatever that total exceeds 100% by is the market’s vig, expressed as a percentage. A total of 104.76% carries 4.76 percentage points of vig; a total of 104.14% carries 4.14.

Step 3: Normalize Each Side to Find the No-Vig Probability

Divide each side’s raw implied probability from Step 1 by the total from Step 2. This proportionally shrinks both numbers until they add up to exactly 100%, without changing how much more likely one side is than the other relative to itself. The two resulting numbers are the no-vig, or fair, probabilities.

Converting each fair probability back into American or decimal odds — using the same conversion formulas from Step 1 in reverse — produces the final no-vig fair odds. Because sportsbooks quote odds in whole numbers, converting a fair probability back into American odds usually requires rounding, so the reversed price is a close equivalent rather than a mathematically exact mirror of the probability.

Two Worked Examples: An Even Market and an Uneven Market

Suppose a sportsbook lists a hypothetical point spread at -110 on both sides — the standard price for most spreads and totals. Each side’s raw implied probability is 110 ÷ 210 ≈ 52.38%. Adding both sides gives 104.76%, so this market carries about 4.76% vig — the most common baseline figure in sports betting.

Dividing 52.38% by 104.76% gives exactly 50% for each side — a genuinely even matchup, once the vig is removed. A market that reduces to a perfect 50/50 split converts to even odds, sometimes labeled “pick’em” on a bet slip.

Now suppose a hypothetical moneyline lists Team A at -130 and Team B at +110 — a more uneven, and more typical, matchup. Team A’s raw implied probability is 130 ÷ 230 ≈ 56.52%; Team B’s is 100 ÷ 210 ≈ 47.62%. The two sides sum to 104.14%, a smaller vig than the -110/-110 example even though the market itself is less balanced.

54.27% and 45.73% add up to exactly 100%, confirming the vig has been fully normalized out on this hypothetical example. These figures are illustrative only, not a current market price.

How to Use No-Vig Probability to Evaluate Your Own Estimate

The practical reason to calculate a no-vig probability is to have a fair benchmark to compare a personal probability estimate against — comparing an estimate to the raw posted price would unfairly measure it against a number that already includes the sportsbook’s margin. Comparing an estimate to the no-vig fair probability, not the raw posted price, is the more accurate way to judge whether that estimate actually disagrees with the market.

Using the -130/+110 example above, suppose a bettor’s own research suggests Team A actually has closer to a 60% chance to win, above the 54.27% no-vig fair probability the market implies. That gap between a bettor’s estimate and the market’s fair number is what a positive expected value evaluation is built on — it does not guarantee the bet wins, only that the estimated probability implies more theoretical value than the fair market price suggests. On a hypothetical $100 stake at the original -130 price, a win pays a $76.92 profit ($176.92 total payout); a loss forfeits the full $100 stake, regardless of how the estimate compared to the fair number.

A no-vig probability is also the right number for comparing the same game across two different sportsbooks. Two books can post different raw odds on identical matchups while implying very similar no-vig fair probabilities, or the reverse — genuinely different fair probabilities that reveal one book’s line is meaningfully out of step with the other’s.

Common Mistakes When Calculating No-Vig Probability

The most serious mistake is treating a no-vig probability as a prediction of what will happen. A 54.27% no-vig probability does not mean Team A wins 54.27% of the time in any single game — it means that, according to the market’s own pricing once the margin is stripped out, that is the fair theoretical chance, subject to all the uncertainty of a live sporting event.

Another common error is normalizing only one side and assuming the other side’s fair number automatically follows without checking. Both sides must be divided by the same total from Step 2 — skipping this and eyeballing the second number is a frequent source of a no-vig pair that doesn’t actually add back up to 100%.

Rounding too early in the process compounds error by the final step. Carrying at least four decimal places through Steps 1 and 2 before rounding a final answer keeps a small rounding choice from meaningfully shifting the resulting fair odds, especially on markets with tighter vig.

Finally, bettors sometimes use the fair probability and the fair odds interchangeably in conversation, which causes confusion when comparing numbers. A fair probability is a percentage; fair odds are the converted price — the two describe the same underlying number in different formats, not two different calculations.

Where This Calculation Comes Up in Real Betting Decisions

No-vig calculations show up most often in two everyday situations: comparing a personal probability estimate against a single market, and comparing the same game’s price across multiple sportsbooks. Bettors who regularly compare posted lines across several sportsbooks — a practice generally called line shopping — often convert each book’s price to its no-vig equivalent first, since a book with slightly worse raw odds can still imply a very similar or better fair probability once its own margin is accounted for.

Odds-tracking tools and calculators automate this exact three-step formula. Understanding the underlying calculation is what makes an automated tool’s output verifiable rather than something to simply trust, and it is the same math a bettor would use by hand on any single market without needing a tool at all.

Related Concepts to Learn Next

This calculation depends on being comfortable converting a single price into implied probability and understanding how vig builds a margin into a market in the first place — both covered above and worth reviewing first if either felt unfamiliar. A market that reduces to an exact 50/50 split, like the first example above, converts to even odds, or “pick’em”. Once no-vig probability feels automatic, the natural next step is applying expected value — using the fair, de-vigged number as the baseline for judging whether a personal probability estimate actually beats what a market is offering.