Connect via MCP →

Enter Calculation

A probability in [0,1] as a decimal (e.g. 0.1667) or fraction (e.g. 1/6) - NOT a percent.

Formula

Show calculation steps (1)
  1. Cumulative & Distribution Stats

    Cumulative & Distribution Stats: Dice Roll Probability Calculator (Binomial Distribution)

    At-most and at-least probabilities sum the pmf; mean and variance summarize the distribution.

Advertisement

Results

Probability of exactly 1 successes P(X = 1)
0.32301117
= 32.3011%
P(X ≤ m) cumulative 0.48451675
P(X ≥ m) 0.83849442
Expected number of successes (mean = n·p) 1.666667
Variance (n·p·(1−p)) 1.388889

Probability for each number of successes

m P(X = m) % P(X ≤ m)
0 0.16150558 16.1506% 0.16150558
1 0.32301117 32.3011% 0.48451675
2 0.29071005 29.071% 0.7752268
3 0.15504536 15.5045% 0.93027216
4 0.05426588 5.4266% 0.98453803
5 0.01302381 1.3024% 0.99756184
6 0.00217064 0.2171% 0.99973248
7 0.00024807 0.0248% 0.99998055
8 0.00001861 0.0019% 0.99999916
9 0.00000083 0.0001% 0.99999998
10 0.00000002 0% 1

What this calculator does

This is a pure-math probability tool based on the binomial distribution. It tells you how likely a specific outcome is to occur exactly m times across n independent trials, where each trial succeeds with probability p. The classic example is rolling a fair six-sided die: a single chosen face appears with probability \(p = 1/6\) on each roll. The math is universal, though - any per-trial success probability between 0 and 1 works (coin flips, free-throw makes, defect rates, and more).

Bar chart of a binomial probability distribution with one highlighted bar
A binomial distribution gives the probability of each possible number of successes m.

How to use it

Enter the number of trials n (1 to 500), the probability per trial p, and a target number of successes m. You may type p as a decimal such as 0.1667 or as a fraction such as 1/6; the calculator evaluates the fraction to a decimal automatically. Importantly, p is a probability in the range 0 to 1, not a percentage - enter 1/6, not 16.67. The result shows \(P(X = m)\) plus a full table of \(P(X = m)\) for every m from 0 to n, the cumulative \(P(X \le m)\) and \(P(X \ge m)\), the mean, and the variance.

The formula explained

The probability of exactly m successes is $$P(X = m) = \binom{\text{Trials }n}{\text{Successes }m} \, p^{\,m} \left(1-p\right)^{n-m}$$ where \(C(n, m) = n! / (m! (n - m)!)\) is the binomial coefficient ("n choose m"). The mean is \(\mu = n\cdot p\) and the variance is \(\sigma^{2} = n\cdot p\cdot(1 - p)\). For large n the calculator works in log-space using the log-gamma function to avoid factorial overflow.

Advertisement
Diagram showing n trials with m successes and the rest failures using dice
Each formula factor counts the ways to choose m successes among n trials.

Worked example

Roll a die \(n = 10\) times; what is the chance a chosen face shows up exactly \(m = 2\) times? Here \(p = 1/6\). \(C(10, 2) = 45\), \(p^{2} = 1/36 \approx 0.027778\), and \((5/6)^8 \approx 0.232557\). So $$P(X = 2) = 45 \times 0.027778 \times 0.232557 \approx 0.29071,$$ about 29.07%. The mean is \(10/6 \approx 1.667\) and the variance is \(10 \times (1/6) \times (5/6) \approx 1.389\).

FAQ

Is p a percentage? No. p is a probability between 0 and 1. For a die face use 1/6 or about 0.1667, not 16.67.

How do I get "at least once"? Use \(P(X \ge 1) = 1 - P(X = 0)\). For 3 dice and a specific face: \(1 - (5/6)^{3} = 1 - 0.578704 \approx 0.4213\), or about 42.13%.

Why do the table probabilities add to 1? Every trial must produce some number of successes from 0 to n, so the probabilities of all mutually exclusive outcomes sum to exactly 1 - a handy sanity check.

Last updated: