Down a goal with ten minutes left, a hockey team sits at +300 on the live moneyline. That price is a claim: it says the comeback should happen about one time in four. Whether that claim is too low, too high or about right is the only question that matters for the bet, and it cannot be judged until the price is turned into a percentage.
The implied probability calculator above does that conversion. Enter a price in American, decimal or fractional form, and it returns the implied probability, the number of wins needed per 100 bets to break even, and the same price in American and decimal formats.
How to use the implied probability calculator
Type a price into the single input. Signed numbers and unsigned numbers of 100 or more are read as American, fractions as fractional, and anything else above 1 as decimal.
The headline number is the implied probability, rounded to one decimal place. It is the most useful single figure on a betting slip, and also the one most slips leave off. Below it sits the same figure written as wins per 100 bets, which is easier to picture: 25.0 means a quarter of such bets have to win just to break even. The American and decimal versions of the price are shown too, which catches typing mistakes before they turn into a bad decision.
The formula, in every format
All three formats lead to the same answer. The table shows the shortcut for each.
| Format | Implied probability |
|---|---|
| Decimal (D) | 1 ÷ D |
| American, positive (+A) | 100 ÷ (A + 100) |
| American, negative (−A) | A ÷ (A + 100) |
| Fractional (a/b) | b ÷ (a + b) |
At +300 the calculation is 100 ÷ 400 = 25.00%. At -200 it is 200 ÷ 300 = 66.67%. A fractional 3/2 gives 2 ÷ 5 = 40.00%, the same as +150 and 2.50 in decimal.
Worked example: price against estimate
Implied probability is half of a comparison. The other half is an estimate of the real chance, and the gap between the two is the edge, if there is one.
Expected value per dollar staked is your estimated probability times the decimal odds, minus 1. Take three cases.
| Price | Decimal | Implied | Estimate | Expected value per $1 |
|---|---|---|---|---|
| +150 | 2.50 | 40.00% | 45% | 0.45 × 2.50 − 1 = +$0.125 |
| +300 | 4.00 | 25.00% | 22% | 0.22 × 4.00 − 1 = −$0.12 |
| -110 | 1.9091 | 52.38% | 54% | 0.54 × 1.9091 − 1 = +$0.031 |
The first row is a clear edge on paper: a $100 bet at +150 with a true 45% chance is worth $12.50 on average. The second row is the hockey comeback. If the real chance is 22%, the +300 price is too short, and each dollar bet loses 12 cents on average. The third row shows how thin the margin is at standard prices. Even a 54% estimate, a strong edge by most standards, earns only about 3 cents per dollar.
The weak link is always the estimate. A few percentage points decide the sign of the result, and few people can estimate a live comeback to within a few points. Treat the implied figure as the bar the estimate has to clear, and be honest about how wide the estimate’s own error is.
Turning an estimate back into a price
The formula also runs the other way, and that direction is often more useful. Start from an estimate and work out the price it deserves. For a probability below 50%, the fair American price is (1 − p) ÷ p × 100. For a probability above 50%, it is −p ÷ (1 − p) × 100.
An estimate of 45% gives (0.55 ÷ 0.45) × 100 = +122. An estimate of 70% gives −(0.70 ÷ 0.30) × 100 = -233. Those are prices with no margin at all, so they set a floor: a bet is only worth placing at a price longer than the one your estimate produces.
Writing the floor down before a game is a good habit for live betting in particular. If you have decided that a trailing team at halftime is worth a bet only at +250 or longer, the in-play decision becomes a quick check of the board instead of a judgment made while the market moves. The implied probability calculator supplies the other half of the check, turning whatever is posted into a percentage that can be set against the plan.
Break-even rates at common prices
The table lists the share of bets that has to win at each price. It is the same figure the calculator shows, collected in one place.
| Price | Implied probability | Wins per 100 to break even |
|---|---|---|
| -200 | 66.67% | 66.7 |
| -150 | 60.00% | 60.0 |
| -120 | 54.55% | 54.5 |
| -110 | 52.38% | 52.4 |
| +100 | 50.00% | 50.0 |
| +150 | 40.00% | 40.0 |
| +200 | 33.33% | 33.3 |
| +300 | 25.00% | 25.0 |
The steps between rows are uneven. Moving from -110 to -120 adds more than two points to the break-even rate, while moving from +200 to +300 takes away eight. Small changes near even money matter a lot for a bettor whose edge is only a point or two, and the table makes that visible in a way the American numbers do not.
Implied probability in live betting
Live prices change with every play, and implied probability is the cleanest way to read what each change means. American odds hide the size of a move. A team going from +150 to -150 looks like a 300-point swing; in probability it is 40.00% to 60.00%, and that is the number worth thinking about.
Broadcasts and apps often display their own live win-probability figures. Those come from models that may weigh the game state differently from the sportsbook’s model, and neither one is a source of truth. When a broadcast shows 30% for the trailing team and the price is +300 (25.00% implied), the two disagree by five points. That gap is worth a look, not an automatic bet: the broadcast model carries no margin and takes no bets.
Timing changes the picture too. A price posted right after a goal or a turnover reflects the new state of the game, while one captured just before it reflects the old one. If you take a percentage from the screen while the market is suspended, the number may describe a moment that has already passed. Read the price when the market reopens, convert it, and only then compare it with a model or an estimate.
A price is a snapshot. Treat it like one.
Short live prices are where implied probability matters most. A team leading by three scores late might be -1000, which implies 90.91%. The bet only breaks even if the team wins more than ten of every eleven such games. One collapse erases ten winning bets, and late collapses are exactly the rare events that a gut feeling tends to discount.
Adding up both sides of a market
Implied probability for one price always includes the sportsbook’s margin. The quickest way to see it is to add the sides of one market. A standard -110 / -110 total adds up to 104.76%; a live moneyline at -400 / +300 reaches 105.00%. Anything above 100% is the margin.
That excess means the implied probability of a single side overstates the market’s real view. At -110 the implied chance is 52.38%, while the fair chance after removing the margin is 50.00%. For any comparison that depends on a few points, run both sides through the no vig calculator first and compare the estimate with the fair figure.
A side-track worth knowing: futures boards push this to extremes. When every team in a league is listed to win the title, the implied probabilities of all the prices together usually sum to far more than 100%, because each longshot carries its own slice of margin. Adding up a whole futures board is a fast way to see how expensive that market is.
The same sum works for any market with more than two outcomes. A soccer match priced at home +150, draw +240 and away +190 implies 40.00%, 29.41% and 34.48%, which comes to 103.89%. A next-goal market in the same game, with three or four outcomes and a shorter time to settle, can carry a wider total, and the sum is how to tell. Comparing the totals of two markets on the same match shows which one charges more for the same game.
Pitfalls when reading implied probability
The first pitfall is treating the number as a forecast. It is a price. The sportsbook sets it to balance risk and to keep a margin, so it reflects the book’s model plus the money bet so far, not a neutral prediction.
The second is false precision. The calculator shows one decimal place, but the inputs rarely deserve that much confidence. A personal estimate of “about 45%” is a range, perhaps 40% to 50%, and a range that wide straddles the 40.00% implied by +150.
The third is mixing prices from different moments. In a live game, an estimate made before a timeout and a price posted after it are about two different situations. Compare a price with an estimate formed at the same moment.
The fourth is underrating how much a single point costs at the extremes. A 1% move matters far more at -1000 than at +100. At -1000 the bet risks $1,000 to win $100, so a true chance of 89.9% instead of 90.9% costs about $11.00 per bet in expected value, while the same one-point gap at +100 costs $2.00 on a $100 stake.