Implied Probability — Turning Odds Into Percentages
Implied Probability
Odds and probability are two ways of saying the same thing. "Implied probability" is just the percentage chance baked into a price. Once you can pull it out, you can compare a sportsbook's opinion directly against a model's — which is the whole game.
The two formulas
For a minus (favorite) price, ignore the sign and use:
implied % = odds ÷ (odds + 100)
So -200 → 200 ÷ 300 = 66.7%.
For a plus (underdog) price:
implied % = 100 ÷ (odds + 100)
So +150 → 100 ÷ 250 = 40%.
A handy anchor: -110 (the standard price) works out to about 52.4%.
Why it's useful
Say a model thinks a team wins 58% of the time, and the sportsbook is offering +120 on that team. Convert the price: 100 ÷ 220 = 45.5%. The model says 58%, the market is charging like it's 45.5% — that gap is value, and it's exactly what our edge number measures.
If instead the price were -150 (implied 60%), the model's 58% would be below the market — no value, even if the team is likely to win. A good team at a bad price is a bad bet.
The catch: the percentages don't add up
Add the implied probabilities of both sides of a real game and you'll get more than 100% — often 104–110%. That extra slice is the sportsbook's built-in margin, the vig. It means the raw implied probabilities are slightly inflated on every side. To compare fairly against a model, you first have to strip that margin out, which we cover in Vig-Free Probability.
On FreezyPicks
When you see a pick listed as "61% win," that's a probability our models produced. When you see "+2.5% edge," we've converted the market price to its vig-free implied probability and subtracted it. Implied probability is the bridge between a number on a sportsbook and a number in a model — and now you can walk across it yourself.
FreezyPicks aggregates independent models, sharp-money data, and our own Iceberg simulation into free, graded picks — for entertainment, not betting advice. See today's picks or the full disclaimer. 21+ and where legal.