What Are Odds?
Odds and probability answer the same question with different bookkeeping. Odds set one class of outcome against the other; probability sets one class against the total. Each converts into the other without any loss of information, so the choice between them is convention — betting and epidemiology kept odds, everyday statistics went with probability.
Odds and Probability Are Not the Same Number
Picture a bag holding four tickets, three of them winners. The event happens three ways and fails one way, so the odds in favor are 3 to 1. The probability is three winners out of four tickets, or 75%. Nothing about the situation changed between those two sentences; only the number under the line did. The trap is that a written ratio looks like a fraction and reads like one, and it is not.
| Written as odds | If misread as a fraction | What it actually means |
|---|---|---|
| 3 : 1 | 3/1 = 300% | 75.0% — three winning outcomes per one losing outcome |
| 1 : 3 | 1/3 = 33.3% | 25.0% — one winning outcome per three losing outcomes |
| 1 : 1 | 1/1 = 100% | 50.0% — one winning outcome per one losing outcome |
| 1 : 5 | 1/5 = 20.0% | 16.7% — one winning outcome per five losing outcomes |
The middle column is wrong in every row, and it is the reading people slide into. The last row is the most instructive: 1 : 5 and one-in-five are different chances, because 1 : 5 describes six outcomes in total and one-in-five describes five.
Odds In Favor, Odds Against, and Which One You Are Holding
The first field takes the number of ways the event happens; the second takes the number of ways it fails. Enter 3 and 1 and the panel returns 75.0%. Enter 1 and 3 and it returns 25.0%. Both are correct answers to the question that was actually typed, and nothing on screen can tell you which question you meant to ask.
British fractional odds make this live. A horse quoted at 7/2 pays 7 units of profit for every 2 staked, so the 7 is counting losing outcomes — those are odds against. Entered as odds in favor, 7/2 becomes 2 then 7, and the calculator returns 22.2% probability with decimal odds 4.50 and an American line of +350. Enter it the other way round, 7 then 2, and you get 77.8% and −350: a different horse entirely, in a different race.
The fourth line of the panel exists for exactly this. For 2 and 7 it carries, word for word, “Odds AGAINST just swap the numbers: 7 : 2 against = the same 22.2% chance” — and if that swapped version is the one you meant, re-enter with the fields reversed.
Odds are a ratio before they are anything else, so reducing one, scaling it, or splitting a total in the same proportion belongs in the Ratio Calculator.
The Four Formats You Will Meet
Four notations are in circulation for the same quantity, and a web page or a betting slip will rarely tell you which one it has chosen.
- Ratio or fractional — two counts separated by a colon or a slash, such as 3 : 1 or 7/2. This is what the calculator's two number fields take. Colon form is usually quoted in favor; slash form in British betting is quoted against.
- Percentage probability — a single figure between 0 and 100. This is the calculator's other input, and the headline it prints on the odds side.
- Decimal, the European default — the total return per unit staked, stake included. Decimal 4.00 returns 4 units on a 1-unit bet: 3 of profit plus the original stake back. Its implied probability is 100 ÷ 4.00 = 25.0%.
- American, also called moneyline — a minus figure is the stake needed to win 100; a plus figure is the profit on a stake of 100. The two signs are not mirror images: −300 implies 75.0% and +300 implies 25.0%.
The calculator reads the first two formats and prints the last two. There is no field for a decimal or an American line, so a posted price has to be turned into a percentage before it can be typed — the arithmetic for that is a few paragraphs down.
How Do You Calculate Odds?
Each direction is one short piece of arithmetic. Odds to probability divides the winning count by the total. Probability to odds subtracts instead: the ratio is P against 100 − P, because whatever is not the chance of happening is the chance of not happening.
The Odds Formula, Written Out
Four lines cover every figure this page produces, in both directions and in both betting notations.
Odds A : B in favor → Probability = A ÷ (A + B)
Probability P% → Odds = P : (100 − P)
Decimal odds = (A + B) ÷ A which is the same as 100 ÷ P
American odds = −(P ÷ (100 − P)) × 100 when P is 50 or more
= +((100 − P) ÷ P) × 100 when P is under 50
then rounded to the nearest whole number
The American rule is where the code makes a choice worth knowing about. Its test is “is the probability at least 50?”, and at exactly 50 that test passes — so even money prints as −100 rather than +100. Both notations describe the identical price: risk 100 to win 100.
Step by Step
- Set the Convert control. It is a two-button segmented switch and it opens on Odds → Probability.
- Converting odds: put the count of ways the event happens in the first field, and the count of ways it fails in the second.
- Converting a probability: put the percentage in the first field and leave the second one empty — that direction ignores it entirely.
- Press Calculate. Both number fields start blank, so the panel shows a dash until you do.
- Read the headline first, then the three supporting lines. Copy result puts all four on the clipboard, though any line the panel has laid out as a label and a value arrives with the two halves run together.
- Press Reset to clear the fields and return the panel to its dash.
Nothing recalculates as you type. Change a value and press Calculate again to refresh the panel.
Worked Example: Odds of 3 to 1
Back to the bag of four tickets — three winners, one blank. The event happens three ways and fails one, which is 3 to 1 in favor. With the switch on Odds → Probability, type 3 into the first field and 1 into the second.
- Odds → Probability, first field 3, second field 1
- 75.0% probability
Beneath the headline the panel prints three lines: “Odds 3 : 1 in favor → 3 ÷ (3 + 1)”, then “Decimal odds 1.33 · American −300”, then the swap reminder. Decimal 1.33 is the total return on a one-unit stake — about 0.33 of profit and the unit itself back. The American −300 says the same thing in stake terms: risk 300 to win 100, returning 400 on 300, which is 1.33 per unit again.
The Same Sum in Reverse: 25% Back to Odds
Switch to Probability (%) → Odds, type 25 into the first field and press Calculate. The headline becomes the ratio instead of the percentage.
- Probability (%) → Odds, first field 25
- 1 : 3 in favor
Its supporting lines read “25% probability → odds for : against” and “Decimal odds 4.00 · American +300”. That decimal is worth holding on to, because 4.00 × 0.25 = 1.00: decimal odds and implied probability are reciprocals by construction. And +300 is the exact counterpart of the −300 above — the two prices are the two sides of one 3-to-1 event.
Turning a Posted Line Back Into a Percentage
Since neither a decimal nor an American line can be typed in, converting a price you have been quoted is arithmetic you do first.
Decimal d → Probability = 100 ÷ d Negative line −L → Probability = L ÷ (L + 100) × 100 Positive line +L → Probability = 100 ÷ (L + 100) × 100
A −150 favorite is 150 ÷ 250 = 60.0%. Feed 60 into the probability side and the calculator hands back 3 : 2 in favor with American −150 beside it, closing the loop. A +150 underdog is 100 ÷ 250 = 40.0%, which returns 2 : 3 in favor and American +150. Those two happen to add to exactly 100 because the pair was chosen to be fair; a market you actually find will not.
What Happens With a Decimal Percentage
The probability side reduces its ratio only when the percentage is a whole number. Type 25 and you get 1 : 3, because 25 and 75 share a factor of 25. Type 65 and you get 13 : 7. Type 65.5 and the reduction is skipped: the panel prints 65.5 : 34.5 in favor.
The betting figures move with it. A probability of 16.7% returns decimal odds 5.99 and American +499, where a true one-in-six — entered on the odds side as 1 and 5 — returns 6.00 and +500. That gap is not an error: one in six is 16.666…%, so 16.7% is a marginally better chance than one in six, and the prices reflect it. When the clean ratio is what you are after, use the odds side.
Reducing 65 : 35 down to 13 : 7 by hand is a greatest-common-factor problem, which is the job of the GCF Calculator.
Odds Chart: Every Format Side by Side
Each row here came out of the calculator's Odds → Probability side, with the ratio in the left column typed into the two number fields. Probabilities appear exactly as the panel prints them, to one decimal place.
| Odds in favor | Probability | Decimal odds | American line |
|---|---|---|---|
| 1 : 9 | 10.0% | 10.00 | +900 |
| 1 : 4 | 20.0% | 5.00 | +400 |
| 1 : 3 | 25.0% | 4.00 | +300 |
| 1 : 2 | 33.3% | 3.00 | +200 |
| 1 : 1 | 50.0% | 2.00 | −100 |
| 3 : 2 | 60.0% | 1.67 | −150 |
| 2 : 1 | 66.7% | 1.50 | −200 |
| 3 : 1 | 75.0% | 1.33 | −300 |
| 4 : 1 | 80.0% | 1.25 | −400 |
| 9 : 1 | 90.0% | 1.11 | −900 |
Even money is the hinge of the table. Decimal odds cross 2.00 there and the American sign flips, and the row is symmetric in a way no other row is — read 1 : 1 forwards or backwards and it is still 50.0%. Notice too that the decimal column is the reciprocal of the probability column all the way down: 10.00 against 10.0%, 1.25 against 80.0%.
Odds for Dice, Cards and Wheels
Counting the outcomes is the whole job; the conversion afterwards is mechanical. Every row below was entered as the two counts shown in its second column.
| Event | Odds in favor | Probability | Decimal | American |
|---|---|---|---|---|
| A coin lands heads | 1 : 1 | 50.0% | 2.00 | −100 |
| One die shows a chosen face | 1 : 5 | 16.7% | 6.00 | +500 |
| Two dice total 7 (6 of 36 rolls) | 1 : 5 | 16.7% | 6.00 | +500 |
| Two dice total 12 (1 of 36 rolls) | 1 : 35 | 2.8% | 36.00 | +3500 |
| Cutting a spade from a full deck | 1 : 3 | 25.0% | 4.00 | +300 |
| Cutting an ace from a full deck | 1 : 12 | 7.7% | 13.00 | +1200 |
| One number on a 37-pocket wheel | 1 : 36 | 2.7% | 37.00 | +3600 |
| One number on a 38-pocket wheel | 1 : 37 | 2.6% | 38.00 | +3700 |
| Dealt a pair of aces (6 of 1,326 hands) | 1 : 220 | 0.5% | 221.00 | +22000 |
The last row shows the rounding at work: a pair of aces is 6 starting hands out of 1,326, which is 0.4525%, and the panel prints 0.5%. The ratio 1 : 220 is exact; the percentage beside it is a display. The two wheel rows give the clearest picture of a house edge anywhere in the table — a straight-up number pays 35 to 1 under the standard payout schedule, while the true odds against it are 36 to 1 on the 37-pocket wheel and 37 to 1 on the 38-pocket one.
How to Read Your Result
Whenever the inputs pass, four lines come back in the same order, whichever direction you converted.
What Each of the Four Lines Says
- The headline is the converted figure. Odds → Probability prints a percentage to one decimal, such as “75.0% probability”. Probability → Odds prints the ratio, such as “1 : 3 in favor”.
- The second line names what was converted. On the odds side it is the division itself, so the sum can be checked on paper: for 3 and 1 it reads “Odds 3 : 1 in favor → 3 ÷ (3 + 1)”. On the probability side there is no division to show, and for 25 it reads “25% probability → odds for : against”.
- The third line carries both betting formats, separated by a dot: “Decimal odds 1.33 · American −300”.
- The fourth is a standing note. On the odds side it restates your entry as odds against; on the probability side it repeats what separates odds from probability.
There is no colored gauge on this page, and deliberately so — a probability has no good or bad band to shade. Where one of those lines carries an equals sign, or a colon that is not part of a ratio, the panel lays it out as a label on the left and a value on the right rather than as one sentence. The words are the same either way; only the separator disappears.
When the Panel Refuses
Both directions guard their inputs, and a failed guard replaces the four lines with a two-line amber notice.
- Odds → Probability needs a positive number in both fields. Zero, a negative, or a blank all produce: “Enter both parts of the odds (e.g., 3 and 1 for 3:1)”.
- Probability (%) → Odds accepts anything strictly above 0 and strictly below 100. Outside that range it produces: “Enter a probability between 0 and 100 (exclusive)”.
Those bounds are looser than they sound. 0.5 and 99.99 both pass, returning 0.5 : 99.5 and 99.99 : 0.01 respectively. The odds side takes decimals too — 2.5 and 1 is a legal entry and returns 71.4%, the same answer as 5 and 2. Neither number field carries a browser minimum or maximum, and the form is submitted with validation switched off, so the formula's own test is the only thing checking you.
Three Checks That Catch a Wrong Entry
The conversion cannot be wrong, which means every error on this page is an input error. These three catch most of them before they travel any further.
- Check the sign against the halfway mark. Under 50% the American figure carries a plus; at 50% or above it carries a minus. A minus line on something you consider an underdog means the two counts went into the wrong fields.
- Check the total. Odds of A : B describe A + B outcomes, not A of them and not B of them. 1 : 5 is six outcomes and 16.7%; 1 : 4 is five outcomes and 20.0%. An off-by-one here moves the answer a full step.
- Read the fourth line back to yourself. It spells your entry out reversed — and if that reversed sentence is the one matching your situation, the fields need swapping.
Why Bookmaker Odds Add Up to More Than 100%
Convert both sides of a real two-way market and the implied probabilities will not total 100. They total more, and the excess is the price of being allowed to bet.
Working the Margin Out
Take a market priced at −110 on both sides. A −110 line implies 110 ÷ 210 = 52.381%. Both sides carry it, so the market as a whole implies 104.762%. The 4.762 points above 100 are the overround — the amount by which the two prices together sit above a fair book.
Implied % from a negative line −L = L ÷ (L + 100) × 100 Implied % from a positive line +L = 100 ÷ (L + 100) × 100 Overround = the implied percentages of every outcome, added up Fair % = one outcome's implied % ÷ overround × 100
Type 52.4 into the probability side and the calculator returns American −110, confirming the conversion running the other way. Now divide each side's 52.381% by the market's 104.762% total: both come back to exactly 50.0%. That is what the market was actually saying — a coin flip — and the −110 pair was the wrapper around it.
A Lopsided Market, Deflated
The arithmetic earns its keep when the two sides are unequal. A market at −180 and +150 implies 64.286% and 40.000%, for a total of 104.286% and an overround of 4.286 points. Dividing each figure by that total gives 61.6% and 38.4%, which do add to 100.
Both go straight back into the calculator. Typing 61.6 returns 61.6 : 38.4 in favor with American −160; typing 38.4 returns 38.4 : 61.6 in favor with American +160. So the margin-free version of a −180 / +150 market prints as −160 / +160, and the distance between the posted pair and that one is what the bet costs before anything has happened on the field.
Deflating a line does not make it true. It strips the margin and leaves the market's own opinion — a crowd estimate with money behind it, which is a reasonable starting point and not a fact. Backing every outcome of a 104.762% market in proportion to its implied probability returns 100 ÷ 104.762 = 95.5 units per 100 staked, no matter which side wins.
Odds Ratios in Research and Health Reporting
Odds survive in one more place outside betting: medical and social research, where the odds ratio is a standard way to report an association. It is also why a headline can announce that something “doubles the odds” of an outcome that has barely moved — and, less often noticed, why another can overstate a common outcome badly.
What an Odds Ratio Is
An odds ratio compares the odds of an outcome in an exposed group against the odds of the same outcome in an unexposed group. StatPearls sets it out as the odds of the event in the exposure group divided by the odds in the non-exposure group, which on a two-by-two table simplifies to ad ÷ bc. An odds ratio of 1 means exposure did not shift the odds; above 1 means higher odds with exposure, and below 1 means lower.
Nothing in that definition mentions probability, and that is the point. The measure is built out of odds at every step, so it inherits the property that made odds awkward in the first place: it stretches as outcomes become likelier.
Why an Odds Ratio Is Not a Risk Ratio
Put two probabilities through the calculator and the stretch becomes visible. A 75% outcome converts to 3 : 1 in favor; a 25% outcome converts to 1 : 3. The odds ratio between those groups is 3 ÷ (1/3) = 9.0, while the risk ratio is simply 75 ÷ 25 = 3.0. Two groups, one comparison, and the odds version is three times the size of the risk version.
Now make the outcome rare. A 2% rate converts to 1 : 49 and a 1% rate to 1 : 99, so the odds ratio is (1/49) ÷ (1/99) = 2.02 against a risk ratio of 2.00 — close enough to be interchangeable in a sentence. That is precisely the rule StatPearls states: where the event is rare the two measures are comparable, but the odds ratio will overestimate the risk where the event is more common, and there the relative risk is the more accurate figure.
So “doubles the odds” is a fair paraphrase when the outcome is rare and a misleading one when it is not. The figure to hunt for in the article is the base rate — what share of the unexposed group had the outcome at all. Without it, an odds ratio cannot be turned into anything a reader can picture.
Turning a base rate into a share of a group, or one group into a share of another, is ordinary percentage work handled by the Percentage Calculator.
Limits: When This Calculator Does Not Apply
The conversion itself is exact arithmetic and cannot be wrong. Everything that can go wrong sits on one side of it or the other.
- It converts; it does not estimate. Entering 3 : 1 establishes that 3 : 1 means 75.0%, not that the event has a 75% chance. The odds still have to come from somewhere — counted outcomes, a model, or a market.
- There is no input for a decimal or an American line. The fields take a ratio or a percentage and nothing else, so a quoted price must be converted by hand first.
- Probability output is rounded to one decimal place. 1 : 220 prints as 0.5% when the exact value is 0.4525%, and the longer the shot, the more of the figure that rounding hides.
- The American line is rounded to a whole number. A probability of 16.7% works out at +498.80 and prints as +499; 61.6% works out at −160.42 and prints as −160. Working the formula above by hand can therefore land a fraction away from the figure on screen.
- The ratio is reduced only for whole-number percentages. 65 returns 13 : 7; 65.5 returns 65.5 : 34.5, which is correct but not tidy.
- Nothing checks direction. Odds against, typed in as odds in favor, produce a perfectly valid answer to a completely different question.
- Implied probability is not probability. A converted betting line still carries the operator's margin until you deflate it, and a deflated line is still an estimate made by other people.
- The arithmetic runs in your browser, but the page around it loads analytics and advertising scripts like any other page on the web.
One caution sits outside the mathematics. Converting a line into a percentage makes a bet feel calculable, and the overround guarantees it is not: back every outcome of a market priced at 104.762% in proportion and you get 95.5 units back per 100 staked, every single time. If gambling has stopped being a hobby, the National Council on Problem Gambling runs a free, confidential helpline on 1-800-MY-RESET, also reachable as 1-800-697-3738.