No-Vig Odds Explained

Two prices, more than one whole

Take any two-way market and convert both sides to implied probability. A pair like −110 / −110 converts to 52.4% and 52.4%. Add them: 104.8%.

There is no world in which two mutually exclusive outcomes are 104.8% likely. The extra 4.8 points aren't a mistake and they aren't a forecast — they're the vig, the book's margin, baked directly into the price you're quoted.

That leaves you with a problem before you've done any handicapping at all: the posted number is not the market's estimate. It's the market's estimate plus a fee. If you want to know what the market actually believes — the thing you'd compare your own number against — you have to take the fee back out.

That operation is called de-vigging.

Step 1 — convert the price to implied probability

American odds convert like this:

• Negative odds (favorite): p = |odds| / (|odds| + 100). So −150 → 150 / 250 = 60.0%.
• Positive odds (underdog): p = 100 / (odds + 100). So +130 → 100 / 230 = 43.5%.

Those two are the raw implied probabilities. Note what they sum to: 60.0 + 43.5 = 103.5%.

Step 2 — measure the overround

That 103.5% is the overround: how much more than a whole the market is charging. Here it's 3.5 percentage points.

The related number — and the one books actually quote internally — is the hold: the overround expressed as a share of the total, (103.5 − 100) / 103.5 = 3.4%. Same market, two denominators, two slightly different numbers. Both are worth knowing, and mixing them up is a common way to mis-state what a market costs.

Step 3 — proportional de-vig

The standard method — and the one Blitzen uses everywhere it needs a fair price — is two-sided proportional de-vig: divide each side's raw probability by the sum of both.

fair_a = raw_a / (raw_a + raw_b)
fair_b = raw_b / (raw_a + raw_b)

Running our example:

• Favorite: 0.600 / 1.035 = 58.0%
• Underdog: 0.435 / 1.035 = 42.0%
• Sum: 100.0% — a real probability distribution.

Convert 58.0% back to American odds and you get roughly −138; the 42.0% side is about +138. That symmetric pair is the fair line: what the market believes, with the fee removed.

Honest caveat on the method. Proportional de-vig assumes the margin is spread evenly across both sides. It usually isn't — books tend to load more of it onto longshots (the favorite-longshot bias), so proportional de-vig slightly under-prices favorites and over-prices dogs at extreme numbers. Alternative methods exist (multiplicative, additive, Shin, power) and they disagree most exactly where the price is most lopsided. Proportional is the honest default: it's transparent, reproducible, and the same on every screen. It is not the last word.

What the fair number is actually for

Three uses, in ascending order of how much work they require of you:

1. Comparing books. Two books quoting the same game will produce different fair lines. The spread between them is a measure of how much the market disagrees with itself — which is information, and it's free.
2. Comparing to your own estimate. If you have an independent probability for a side, the fair line is the honest thing to compare it against. Comparing your estimate to the posted price flatters you by exactly the size of the vig — you'd be counting the book's fee as if it were your own insight.
3. Comparing to the close. The closing price, de-vigged, is the market's final and sharpest estimate. Grading your entries against that is the whole basis of closing line value.

What it is not for

De-vigging tells you what a market believes. It does not tell you what will happen, and a gap between your number and the fair number is not money — it's a disagreement, and one of you is wrong. Two things that gap has to survive before it means anything:

• Your probability has to be better than the market's. The market's number has absorbed every injury report and every dollar of sharp money. The default assumption should be that it is right and you are not.
• The fee doesn't disappear because you removed it from the math. You still transact at the posted price, vig included. De-vigging changes what you measure, not what you pay.

This is the same posture we take on our own models. Stated plainly on our methodology page: no Blitzen model clears the vig against the close yet — the signal is real on moneylines, it just doesn't cover the juice on its own. We publish that because a number you can't check is worth less than no number.

Do it yourself

All of this is arithmetic, and all of it is free and public on Blitzen — no account needed:

• No-vig calculator — both sides in, fair probabilities and fair American/decimal odds out, plus the book's margin.
• Implied probability converter — American ↔ decimal ↔ percent, formulas shown.
• Hold calculator — overround and theoretical hold from a two-way price pair, so you stop conflating them.
• EV calculator — your probability and the offered price → expected value per $100 staked, and the break-even probability the price implies.

Learn the arithmetic once and you'll never look at a posted price the same way again. That's the entire point: read the market, then check its work.

Blitzen is an analytics and research tool, not betting advice or a picks service. Nothing here predicts a financial outcome. Sports betting carries real financial risk — wager only what you can afford to lose, and only where it is legal. Must be 21+. If gambling stops being fun, help is available: call 1-800-GAMBLER.