Implied probability is the percentage hiding inside a price. When a bookmaker quotes 2.10 for a home win, or an options market shows a call with a 0.50 delta, each number already contains an estimate of how often the outcome happens, and the whole job of implied probability is to decode that number back into a percentage you can compare against your own estimate. The Implied Probability Calculator does that decoding in two directions: it turns any quoted probability into decimal, American and fractional odds, and it takes an option delta, a strike and a premium, then reads off the chance the option finishes in the money plus the breakeven price on both sides.
What implied probability is: the percentage hidden inside the price
A decimal quote is a probability in disguise. The conversion is one line of arithmetic: divide 100 by the decimal odds. At 1.67 the implied probability is 100 divided by 1.67, which is 59.88 percent. At even money, 2.00, the market is saying exactly 50 percent. At 3.00 the chance is 33.33 percent, at 4.50 it is 22.22 percent, at 1.20 it is 83.33 percent, and at a long-shot 10.00 the market prices the outcome at 10 percent. The longer the decimal odds, the smaller the probability, and the relationship is inverse, so doubling the price exactly halves the implied chance.
Reading it back is where the number becomes useful. A quote of 1.67 is not really a payout table, it is the market's statement that the event happens roughly six times out of ten. Once every quote is on the same percentage scale, a home win at 2.10, a draw at 3.40 and a call at 0.50 delta can all be compared side by side, and that comparison is the entire point of implied probability. The reverse direction, starting from a probability and producing the odds, is covered in the next section.
Converting each odds format to a percentage
Decimal odds are the simplest because the formula is the direct one, 100 divided by the odds. The other two formats both reduce to a decimal first. American odds carry a sign. A negative American quote of minus 150 means the bettor must risk 150 to win 100, and the implied probability is 150 divided by 250, which is 60 percent. A positive quote of plus 150 pays 150 on a 100 stake, so the probability is 100 divided by 250, or 40 percent. The same rule gives minus 400 an implied probability of 80 percent and plus 1000 a long-shot 9.09 percent.
Fractional odds a to b convert with 100 times b over a plus b. The classic 2/3 quote becomes 300 over 5, which is 60 percent. A 5/2 quote is 200 over 7, which is 28.57 percent. A 10/1 long shot is 100 over 11, which is 9.09 percent, and a 4/1 quote is 100 over 5, or 20 percent. The full odds conversion walkthrough, including how the American rounding rules behave at the plus-minus boundary, is written out step by step in the companion article, and the odds converter tool does the three-way conversion in one click for any value you have on hand.
The tool works the other way too, from a probability to all three formats. Enter 60 and it returns 1.67 in decimal, minus 150 in American and 2/3 in fractional, where the 40 to 60 ratio is reduced by its greatest common divisor. Enter 40 and you get 2.50, plus 150 and 3/2. Enter 50 and the three formats agree at even money: 2.00, plus 100 and 1/1. Enter 80 and you get the heavy favorite 1.25, minus 400 and 1/4. Enter 25 and you get 4.00, plus 300 and 3/1, and enter 10 for the long shot 10.00, plus 900 and 9/1. The American column is always rounded to a whole number, which is why a probability of 33.33 displays as plus 200 rather than plus 199.9 or plus 200.1.
Stripping the bookmaker margin: fair probabilities from a three-way market
Here is the part that trips people up. Take a soccer match priced at 2.10 for a home win, 3.40 for a draw and 3.60 for an away win. Convert each one: 100 over 2.10 is 47.62 percent, 100 over 3.40 is 29.41 percent, and 100 over 3.60 is 27.78 percent. Add them together and you get 104.81 percent, which is impossible for a market with exactly one winner. The extra 4.81 percentage points are the bookmaker's margin, the overround, and it is the fee the market charges on top of the true probabilities.
The fix is normalization. Divide each implied probability by the total of 104.81: the home win drops from 47.62 to 45.43 percent, the draw from 29.41 to 28.06, and the away win from 27.78 to 26.50. The three now sum to exactly 100 percent, and they are the fair probabilities hidden under the margin. The margin itself can be quoted two ways, as 4.81 points of overround, or as 4.59 percent of the total pool, which is 4.81 divided by 104.81. Both describe the same fee, one in absolute points and one relative to the market size.
Why bother? Because bookmakers quote the same match at slightly different prices, and the raw odds do not make the comparison visible. One book's 2.10 and another book's 2.12 look nearly identical, but after normalization the fair home probability differs by almost half a point, and that half point is the entire difference in value between the two quotes. Normalized probabilities are also the right baseline for the next two sections, because when you compare your own estimate against the market, you want to compare it against the market's fair view, not against a view padded by the margin.
Implied probability in options: delta as the chance of finishing in the money
Implied probability has a second home in options, where it answers a different question: what are the odds this option expires in the money? The standard shortcut is delta. Option delta is commonly used as a rough approximation of the probability that the option expires in the money, so a call with a 0.50 delta carries a 50 percent implied probability, a 0.30 delta carries 30 percent, a 0.90 delta carries 90 percent, and a 0.10 delta is the 10 percent lottery ticket. The calculator's delta table attaches plain-language labels to the same scale: 0.10 is very unlikely, 0.30 unlikely, 0.50 even, 0.70 likely, 0.90 very likely, and 0.95 almost certain.
The delta mode asks for three inputs and returns four numbers. Enter a delta of 0.50, a strike of 100 and a premium of 3, and it reports 50 percent in the money, 50 percent out of the money, a call breakeven of 103 and a put breakeven of 97. The breakeven arithmetic is one line each way: a call needs the price to cover the premium, so breakeven is strike plus premium, 100 plus 3 equals 103, and the underlying must rise 3.00 percent just to give back the cost of the option. A put needs the price to fall by the premium, so 100 minus 3 is 97, a 3.00 percent drop. A second example, delta 0.30 with a strike of 110 and a premium of 2.50, gives 30 percent in the money and a call breakeven of 112.50, which is a 2.27 percent move on the 110 strike.
The approximation is tightest for at-the-money options well away from expiration, where the pricing model and the delta reading line up. It gets shakier as expiration approaches, because a 0.50 delta option in the last few days is a coin flip that snaps to zero or certain as the price crosses the strike, and the delta number alone no longer tells you which way the snap is coming. The Black-Scholes calculator is where the exact probability lives, and the article on calculating a Black-Scholes option price walks through the normal distribution step that turns delta into a defensible probability rather than a shortcut.
From probability to edge: expected value and bet sizing
Implied probability tells you what the market thinks. It does not tell you whether the market is right, and that distinction is where money is made or lost. The test is expected value: multiply your own probability by the decimal odds and subtract 1. At 2.10 the break-even probability is 47.62 percent, so you need to be right more often than that just to wash. If your honest estimate of the home win is 55 percent, the expected value is 0.55 times 2.10 minus 1, which is plus 15.5 percent per unit staked, or 15.50 dollars of expected profit on a 100 dollar bet.
The mirror image is just as important. If your estimate is 45 percent, the same 2.10 quote gives 0.45 times 2.10 minus 1, which is minus 5.5 percent, a negative edge that loses money the more you bet, no matter how confident the story feels. And the benchmark for your estimate should be the fair probability from the last section, 45.43 percent after the margin is stripped, not the raw 47.62. An estimate of 55 percent sits 9.6 points above the market's fair view, and that gap, not the raw odds, is the edge you would be paid for.
Size follows the edge. The Kelly criterion, covered in the Kelly criterion article and computed in the Kelly criterion calculator, gives the fraction of bankroll to risk on a positive-edge bet. With decimal odds of 2.10 the net payoff b is 1.10, and with a 55 percent win probability the full Kelly fraction is 0.55 times 1.10 minus 0.45, all over 1.10, which works out to 14.09 percent of bankroll. Most practitioners run half that, 7.05 percent, because the edge estimate is itself uncertain and half Kelly cuts the volatility roughly in half while giving up only a modest slice of the long-run growth.
Using the calculator and the honest limits
The calculator has two modes. The option delta mode takes a delta between 0 and 1, a strike price and a premium, and returns the in-the-money and out-of-the-money probabilities plus both breakeven prices, with a reference table of common delta values from 0.10 to 0.95. The probability mode takes a percentage between 0 and 100 and returns the three odds formats side by side, with the overround shown as 100 percent for a single-outcome quote. Both directions are the same arithmetic, one line each way, and the tool exists so you do not have to re-derive it under pressure.
Four limits are worth stating plainly. First, delta is an approximation, and the tool says so in its own note: the actual probability depends on the pricing model and the assumptions about volatility. Away from expiration and near the money it is close; near expiration it can be wrong by a wide margin. Second, the overround normalization assumes the market is roughly efficient, so if the quoted odds are mispriced, the normalized probabilities inherit the mispricing and the margin strip only removes the fee, not the error. Third, the display rounds: American odds to whole numbers and decimal odds to two places, so 33.33 percent shows as plus 200 and 1.6667 shows as 1.67, which is fine for reading the market but should not be fed back into a precision calculation. Fourth, a single probability is not a distribution. If you want the shape of the outcome, the range of possible prices at expiration, or the path of a portfolio over time, the Monte Carlo stock simulator is the right instrument, and the implied probability is one input into it rather than the whole answer.
The arithmetic of implied probability fits on one index card: divide 100 by the odds to read the market, divide by the overround to strip the margin, subtract 1 from your probability times the odds to find the edge, and size the position to that edge. The calculation is the easy part. The estimate you compare it against is where the skill, and the risk, actually live.