The Calculation
Implied probability is one divided by the decimal price. Everything else on this page is bookkeeping around that single step.
At -110 the decimal price is 1.9091, so the implied probability is 1 ÷ 1.9091 ≈ 52.38%. At +250 the decimal is 3.50, so it's 1 ÷ 3.50 ≈ 28.57%.
Working straight from American odds, without the detour through decimal: for a negative price, probability = |odds| ÷ (|odds| + 100). For a positive price, probability = 100 ÷ (odds + 100). So -150 gives 150 ÷ 250 = 60%, and +150 gives 100 ÷ 250 = 40%.
Implied Probability Is Also Your Break-Even Rate
The same percentage answers a second question: how often do you actually need to be right to stop losing money.
At -110, 52.38% is both what the price implies and the win rate required to break even. Win 52 of 100 bets at -110 and you finish slightly down. Win 53 and you're slightly up. The gap between 50% and 52.38% is, in a sentence, the entire business model of a sportsbook.
Written as a record instead of a percentage, it gets more concrete. "You need 52.38%" is abstract. "You need 52-48, every hundred bets, forever" is the actual requirement, and it reads as considerably harder than the percentage did.
| Price | Implied | Record needed / 100 |
|---|---|---|
| -200 | 66.67% | 67-33 |
| -150 | 60.00% | 60-40 |
| -110 | 52.38% | 52-48 |
| +100 | 50.00% | 50-50 |
| +150 | 40.00% | 40-60 |
| +250 | 28.57% | 29-71 |
Why Both Sides Add Up To More Than 100%
Add the two sides of any market together and the implied probabilities total more than one. That surplus is the book's margin, and ignoring it is how a reasonable sounding bet turns out to have never really been live.
A -110 pair implies 52.38% twice, or 104.76% combined. The extra 4.76% is what the industry calls "the juice"or "vig". Split it out proportionally and each side's fair probability lands exactly at 50%, which is what a coin flip should be.
This matters most when the two sides aren't symmetric. Take -140 on one side and +120 on the other: raw implied probabilities of 58.33% and 45.45%, totaling 103.79%. The favorite's fair probability works out to 58.33 ÷ 103.79 ≈ 56.2%. So the fair price is closer to -128 than to -140. If your own read on the game is 57%, that's a real edge at -140 but barely one, and it disappears entirely if you rounded your own number even slightly high.
What To Actually Do With The Number
Once a price becomes a percentage, a few genuinely useful comparisons open up.
- Compare it to your own read. A possibly live bet exists when your estimate is meaningfully higher than the fair implied probability. Not just a point or two higher, since that's well within normal estimation error for most people handicapping by hand.
- Compare it across books. -105 and -115 on the identical side are 51.22% and 53.49% More than two full points apart on the same opinion. Shopping for the better number costs nothing and adds up over a season.
- Compare it to what actually happened. This is the step most services skip, and it's the whole reason Line Theory posts a public, timestamped record instead of a highlight reel. A probability estimate only means something once you can check it against real results, wins and losses both.
Frequently Asked Questions
How do I calculate implied probability from betting odds?
Divide 1 by the decimal price. Directly from American odds: negative prices are |odds| ÷ (|odds| + 100); positive prices are 100 ÷ (odds + 100). So -150 implies 60% and +150 implies 40%.
What win rate do I need to break even at -110?
52.38%, or roughly 52-48 across every hundred bets. Winning exactly half your bets at -110 loses money steadily over time. That gap is the sportsbook's edge.
Why do both sides of a market add up to more than 100%?
Because the book's margin is baked into both prices. A -110 pair totals 104.76% of implied probability; the surplus is the overround, and it represents roughly what the book holds on balanced action.
What's the difference between implied probability and fair probability?
Implied probability comes straight from one price and includes the vig. Fair (no-vig) probability divides that by the total of both sides, removing the book's edge. It's the more honest number to compare your own estimate against.