Binomial Probability Calculator
Exact, at-most and at-least binomial probabilities with the mean and standard deviation.
What this page computes
Type the number of trials n, the number of successes k and the success probability p. The point probability is C(n,k)p^k(1 − p)^(n−k), evaluated through the log gamma function so large n does not overflow; the page adds the lower tail P(X ≤ k) and the upper tail P(X ≥ k), labels which tail each is, and prints the mean np and standard deviation √(np(1 − p)). Figures print to six significant figures.
The model assumes n independent trials with the same p, so the page is not a description of a real experiment that breaks those assumptions. A probability of exactly 0 or 1 is refused, because the outcome is then certain, and k outside 0 to n is refused with the bound.
The sums are computed in this tab, so the same three numbers always print the same probabilities.
The page evaluates the binomial distribution for n independent trials that share one success probability: P(X = k) = C(n, k)·p^k·(1 − p)^(n − k). Ten trials at p = 0.2 give P(X = 3) = 0.201327, with P(X ≤ 3) = 0.879126 below it and P(X ≥ 3) = 0.3222 above it, and the mean np = 2 and standard deviation √(np(1 − p)) = 1.26491 printed beside them. It needs p strictly inside 0 … 1, so a certain or impossible event is refused instead of being answered with a degenerate distribution.
Compute binomial point and tail probabilities for up to 100 000 trials. Type the trials, the successes and p; every probability names its tail.
How to use it
Type the number of trials n, the number of successes k and the success probability p. n runs from 1 to 100 000 and k from 0 to n; a k outside that range is refused by name, and so is a p of exactly 0 or exactly 1. Every printed probability says which tail it is: P(X ≤ k) is the lower tail and P(X ≥ k) is the upper one, so no line is a bare number whose meaning has to be guessed.
What the three probabilities mean
The three probabilities come from one distribution, so P(X ≤ k) and P(X ≥ k) together count the outcome k twice and their sum is 1 + P(X = k); that overlap is deliberate and is why neither tail is the complement of the other. The mean np and the standard deviation √(np(1 − p)) describe the whole run of n trials, not the single value k. Nothing is looked up in a table and nothing is simulated: each probability is the closed form evaluated directly.
What to Expect
Best for
- Checking a binomial probability by hand before an exam or a report
What this tool keeps
- Nothing leaves this tab: the arithmetic runs in the page you are reading, and a reload discards both the input and the result.
Sample output
P(X = 3) = 0.201327 P(X ≤ 3) = 0.879126 (lower tail) P(X ≥ 3) = 0.3222 (upper tail) mean = np = 2, sd = √(np(1 − p)) = 1.26491 n = 10 trials, p = 0.2
Frequently asked questions
Which tail is which in the output?
P(X ≤ k) is the lower tail — k successes or fewer — and P(X ≥ k) is the upper tail. They overlap at exactly k, so their sum is 1 + P(X = k) rather than 1.
Is anything uploaded?
No. The three probabilities are computed by JavaScript in this page from the numbers you typed, and nothing is posted to a server or stored.
What inputs are refused?
Trials below 1 or above 100 000, successes outside 0 … n, and a success probability of exactly 0 or exactly 1. Each refusal names the value and the range it missed.
Build custom tools for your store
Join the SDK program to create useful ecommerce utilities for your store.