Parlay Calculator
See your parlay’s combined odds and payout, remove the vig to estimate each leg’s true probability, and apply a correlation adjustment when legs come from the same game.
Your parlay legs
2 / 6 legsPut two legs in the same game to apply a correlation adjustment.
Combined book price
Per-leg de-vig
| Leg | Odds | Book implied | True (de-vig) | Method |
|---|---|---|---|---|
| 1 | -110 | 52.4% | 49.8% | haircut |
| 2 | +150 | 40.0% | 38.0% | haircut |
“two-sided” = both sides supplied, normalized proportionally. “haircut” = one side only, graduated longshot haircut. Both are estimates.
If the legs were independent
The book's price is below fair — roughly 9.7% margin on this parlay.
How this calculator works
- Enter 2 to 6 legs as American odds (for example -110 or +150), and pick a game for each leg.
- We convert each price to an implied probability and remove the vig to estimate each leg’s true probability.
- We multiply the true probabilities for an independent estimate, then adjust for correlation between legs that share a game.
What is the vig — and how we remove it
A sportsbook’s prices include the vig (also called juice) — a built-in margin that makes the implied probabilities across a market add up to more than 100%. To estimate a fair price we strip that margin out. If you supply both sides of a market, we normalize the two implied probabilities proportionally — p / (p + q), the multiplicative method. If you supply only one side, we apply a graduated haircut: about 5% at standard prices, rising toward 15–20% at longshot prices, mirroring how the vig is concentrated on longshots. Every figure here is an estimate derived from the prices you enter, not a measured probability.
Why correlation changes parlay math
Multiplying leg odds assumes every leg is independent. Legs from the same game often are not — a pitcher recording a lot of strikeouts and his team winning tend to move together. When legs are positively correlated, the true combined probability is higher than the naive product, so the parlay is worth more than multiplied odds suggest; when they conflict, it is worth less. This is exactly why sportsbooks reprice same-game parlays instead of simply multiplying the leg prices. The worked example below uses a simplified correlation coefficient (φ); Bettorly’s app measures actual historical co-occurrence between legs instead.
| Measure | Value |
|---|---|
| Legs entered | -110 and +150 |
| Combined book price | +377 (decimal 4.77) — a $100 stake returns $477.27 |
| Independent true probability | 18.9% → fair price +429 ($528.83) |
| Same game, moderately correlated (φ = +0.30) | 26.2% → fair price +282 ($381.82) |
These figures use φ = +0.30 (moderately correlated) for the two same-game legs. They are estimates, not predictions.
Frequently asked questions
How do you calculate parlay odds from American odds?
We convert each leg from American odds to decimal odds, then multiply the decimal odds together to get the combined parlay price. Multiplying that combined price by your stake gives the total potential payout. This is the standard naive parlay price a sportsbook shows before any correlation adjustment.
What is the vig and what does "de-vig" mean?
The vig (or juice) is the margin a sportsbook builds into its prices, which makes the implied probabilities of a market add up to more than 100%. De-vigging removes that margin to estimate a leg’s fair, true probability. If you enter both sides of a market we normalize them proportionally; with only one side we apply a graduated haircut. These are estimates, not certainties.
Why do correlated legs change a parlay’s value?
When two legs come from the same game they can move together — if one hits, the other can become more or less likely. Multiplying leg odds assumes the legs are independent, which overstates or understates the true combined probability for same-game parlays. Our correlation adjustment applies a simplified assumed relationship so the estimate is more realistic.
Are these probabilities guaranteed to be accurate?
No. Every number here is an estimate derived from the market prices you enter and a simplified correlation model. Past frequencies and market-implied probabilities never guarantee an outcome. Treat the results as analytics to inform your own judgment, not a prediction.
Does Bettorly take bets?
No. Bettorly is an analytics tool — we don’t take bets and we don’t offer betting picks, and we’re not affiliated with any sportsbook. We help you understand how the legs in a parlay relate to each other.
Ready for the real thing?
This calculator uses a simplified, assumed correlation. Bettorly’s app analyzes how the legs of a live slate actually correlate, using real hit-rate data.
Analyze a real slate in Bettorly