331+ free tools. No sign-up. No hidden fees. Explore the toolkit

Probability Calculator

Probability is the mathematics of chance: how likely is an event, what happens when two events combine, and how do repeated tries change the odds? This calculator covers all three questions and shows every answer as a fraction, a decimal and a percentage, so the result is easy to read however you think about chance. If you are new to the idea, start with what probability means.

Work out the chance of an event happening — a single coin toss, two events together, or the same trial repeated many times — with the probability shown as a percentage, a decimal and odds.

Choose what you want to work out.

Outcomes that count as the event happening.

All equally likely outcomes.

Probability

Save to Pinterest
Note: Every probability is a number between 0 (impossible) and 1 (certain). Answers are shown as a fraction, a decimal and a percentage; fractions are approximations when the exact value needs a very large denominator.

How It Works

Probability measures chance on a scale from 0 to 1. A single event is the simplest case: if every outcome is equally likely, divide the favourable outcomes by the total outcomes. With 3 winning tickets among 20, P(win) = 3/20 = 0.15 = 15%.

Two events combine in three useful ways. The intersection, P(A and B), is the chance both happen. The union, P(A or B), is the chance at least one happens: add the two probabilities, then subtract the overlap so it is not counted twice — P(A or B) = P(A) + P(B) − P(A and B). The conditional probability P(A|B) is the chance A happens given that B has happened: P(A|B) = P(A and B) / P(B). For independent events the intersection is simply P(A) × P(B); for dependent events you enter the overlap yourself. Bayes' theorem turns a conditional probability around: it works out P(A|B) from P(B|A).

Repeated trials use the complement rule. The chance of at least one success in n independent tries is 1 minus the chance of failing every time: P(at least one) = 1 − (1 − p)ⁿ. It grows quickly at first — four tries at 25% each give a 68.36% chance of at least one success — but it never quite reaches 100%.

Worked Example

Single event: 3 winning tickets out of 20 gives P(win) = 3/20 = 0.15 = 15%. Two independent events: with P(A) = 0.6 and P(B) = 0.5, the intersection is 0.6 × 0.5 = 0.3, the union is 0.6 + 0.5 − 0.3 = 0.8, and P(A|B) = 0.3/0.5 = 0.6. Two dependent events: with the same P(A) and P(B) but an overlap of 0.2, the union is 0.6 + 0.5 − 0.2 = 0.9, P(A|B) = 0.2/0.5 = 0.4, and P(B|A) = 0.2/0.6 = 1/3 ≈ 33.33%. Repeated trials: a 25% chance per try over 4 tries gives 1 − 0.75⁴ = 1 − 0.3164 = 0.6836 = 68.36% (exactly 175/256).

Assumptions

Sources: No external data is used on this page — every figure is calculated from the numbers you enter. Last reviewed: 21 November 2026.

Built and tested by SumKitTool — see how we check our numbers.

How to Use This Calculator

Pick a mode, fill in the numbers, and the results update live as you type.

  1. Single event: enter the favourable outcomes and the total outcomes — for example, 1 and 6 for rolling a 6 on a die.
  2. Two events: enter P(A) and P(B) as decimals. If the events are dependent, switch the relationship to “Dependent” and enter the overlap P(A and B).
  3. Repeated trials: enter the success chance per try and the number of tries to see the chance of at least one success.

Every answer appears as a fraction, a decimal and a percentage. The Reset button restores the worked examples.

Tips and Practical Advice

The Birthday Problem: Why Shared Birthdays Are So Common

Probability often disagrees with gut feeling, and the birthday problem is the classic example. In a group of just 23 people, the chance that at least two share a birthday is 50.73% — better than even. With 30 people it is 70.63%, and with 50 people it reaches 97.04%. Most people guess far lower, because they picture their own birthday and ask whether someone else matches it — a much rarer event, about 5.86% in a group of 23. The birthday problem asks a different question: does any pair in the room match?

The reason is the number of pairs. With 23 people there are 23 × 22 / 2 = 253 different pairs, and each pair is a separate chance to match. The calculation uses the complement rule, the same idea behind the repeated-trials mode above: work out the chance that nobody shares a birthday, then subtract from 1. The first person can have any birthday (365/365), the second must avoid it (364/365), the third must avoid both (363/365), and so on down to 343/365 for the 23rd person. Multiply those 23 fractions and the chance of no match is 49.27% — so the chance of at least one match is 1 − 0.4927 = 50.73%.

The same complement thinking answers “at least one” questions everywhere. Rolling a six at least once in six throws of a die is not a certainty — each throw is independent, so the chance is 1 − (5/6)⁶ = 66.51%. Try p = 0.1667 and n = 6 in the repeated-trials mode above and you will see 66.52%, the tiny difference coming from rounding 1/6 to 0.1667. Adding the probabilities (6 × 1/6 = 100%) is the classic mistake: probabilities of overlapping events do not add that way. And a run of bad luck changes nothing — after five tails in a row, the next coin toss is still 50/50, because the coin has no memory. That is the gambler's fallacy, and it is the mirror image of the birthday problem: in one case coincidence is more likely than it feels, in the other a pattern feels due when it is not.

Before trusting any probability answer, check the assumptions behind it. The birthday calculation assumes birthdays are spread evenly across 365 days (ignoring 29 February) and that people's birthdays are independent — both close enough for a classroom, though real birthdays cluster slightly by season. The repeated-trials formula needs the same p on every try and no influence between tries. When those assumptions hold, the complement rule turns a hard “at least one” question into an easy multiplication — and the answer is usually more surprising than intuition suggests.

Frequently Asked Questions

What is the probability of an event?

It is the chance the event happens, written as a number between 0 (impossible) and 1 (certain). If 3 of 20 raffle tickets win a prize, the probability of winning is 3/20 = 0.15 = 15%. This calculator shows every answer as a fraction, a decimal and a percentage.

What is the difference between independent and dependent events?

Independent events do not affect each other — the chance of rolling a 6 is 1/6 no matter what you rolled before — so P(A and B) = P(A) × P(B). Dependent events do affect each other, such as drawing cards without replacing them, so you must enter the overlap P(A and B) yourself and the calculator works out the rest.

How do I calculate the probability of at least one success?

Work out the chance of failing every time, then subtract from 1: P(at least one) = 1 − (1 − p)ⁿ. With a 25% chance per try and 4 tries, that is 1 − 0.75⁴ = 1 − 0.3164 = 0.6836, or 68.36% — much better than a single try, but still not a guarantee.

Why can't a probability be more than 1?

A probability of 1 means the event is certain, so there is nothing beyond it — 100% is the ceiling. If your calculation gives a number above 1, one of the inputs is wrong: check that each probability is between 0 and 1, and that the overlap P(A and B) is no larger than either single probability.

What is the birthday paradox?

It is the surprising fact that in a group of just 23 people, there is a 50.73% chance that at least two share a birthday — better than even. With 30 people the chance is 70.63%, and with 50 people it is 97.04%. It feels wrong because 23 people make 253 different pairs, and each pair is a chance to match. It is not a true paradox, just probability disagreeing with intuition.

If I flip heads five times in a row, is tails more likely next?

No. Each coin toss is independent, so the next flip is still 50/50 no matter what came before — the coin has no memory. The 1 in 32 chance (0.5⁵ = 3.125%) is the probability of five heads in a row before you start flipping, not the probability of the next flip after four heads have already happened. Believing a result is “due” after a run is called the gambler's fallacy.

Last updated: