What this calculator does
The binomial distribution describes repeated yes-or-no trials: ten coin flips, twenty sales calls, five free throws. Each trial has two outcomes, the chance of success stays the same, and the trials do not influence each other. Under those conditions the distribution gives exact probabilities for any count of successes, plus a mean and a spread. This page computes the point probability, the cumulative probability, and both summary statistics, and it explains when the binomial model fits and when it does not.
How to use it
- Enter n, the number of trials, as a whole number.
- Enter p, the probability of success on one trial, between 0 and 1.
- Enter k, the number of successes you want the probability for.
- Read P(X = k) and the cumulative P(X ≤ k).
- Check the mean and standard deviation for the expected count and spread.
How it works
Example calculations
Example 1: n = 10, p = 0.5, k = 5
P(X = 5) = 252 x 0.5^10 = 0.2461, about 24.6%. The mean is 5 and the standard deviation is about 1.58. The chance of five or fewer successes is 0.6230, just under two thirds.
Example 2: n = 20, p = 0.3, k = 6
P(X = 6) = 38,760 x 0.3^6 x 0.7^14 = about 0.1916, or 19.2%. The mean is 6 and the standard deviation is about 2.05. The chance of six or fewer is 0.6080, so a result at the mean still covers most of the lower tail.
Example 3: n = 5, p = 0.8, k = 4
P(X = 4) = 5 x 0.8^4 x 0.2 = 0.4096, or 41.0%. The mean is 4 and the standard deviation is about 0.89. The chance of four or fewer is 0.6723, because the only way to exceed four is to succeed on all five trials.
What the result means
The point probability answers a narrow question: exactly k successes. The cumulative figure usually answers the practical one: at most k, or, by subtracting from 1, at least k. Read the mean as the long-run average count if you repeated the whole set of trials many times, not as a prediction for the next set. The standard deviation tells you how far a typical result tends to stray from the mean; a small standard deviation means results cluster, a large one means they scatter. When n is large and p is not near 0 or 1, the distribution looks roughly bell-shaped, and a normal approximation with the same mean and standard deviation becomes reasonable. When p is extreme, expect a skewed distribution and many zero-success outcomes. Always check that the model's assumptions actually hold before trusting a probability.
Common mistakes
- Entering 50 instead of 0.5 for a 50% success probability.
- Using a k larger than n, which has probability zero.
- Treating trials as independent when they are drawn without replacement.
- Reading the point probability when the question was at least k successes.
- Assuming p stays constant when the process changes between trials.
Assumptions and limits
- The number of trials n is fixed in advance.
- Each trial has exactly two outcomes, success or failure.
- The success probability p is the same on every trial.
- Trials are independent of one another.
- k is a whole number between 0 and n.
Frequently asked questions
What is the difference between P(X = k) and P(X ≤ k)?
P(X = k) is the chance of exactly k successes. P(X ≤ k) adds the probabilities for 0 through k, so it covers k or fewer. For at least k, calculate 1 minus P(X ≤ k - 1).
What does the mean mean here?
It is the expected number of successes, n x p. If you ran the same number of trials many times, the average count would approach that value. It does not mean you will see exactly that many successes next time.
When can I use a normal approximation?
A common guideline is that both n x p and n x (1 - p) should be at least 5 or 10, based on the source you follow. When they are, a normal curve with the same mean and standard deviation gives a close approximation, especially with a continuity correction.
Does this work for drawing without replacement?
Not exactly. Sampling without replacement changes the probability after each draw, which is the hypergeometric distribution. If the sample is small relative to the population, the binomial is usually a close approximation.
Can p be exactly 0 or 1?
Yes, though the result is trivial. If p = 0 the only possible count is zero successes; if p = 1 every trial succeeds. The formula still works as long as you avoid 0^0 in the arithmetic.
Why does the cumulative probability jump at the mean?
Because the distribution piles up around its most likely values. For n = 10 and p = 0.5 the single outcome k = 5 carries about a quarter of all the probability, so the cumulative total rises sharply as you pass through it.
Sources
2026 update
How to use the professional output
This page is not just a formula box. The result is designed to be checked, compared, and taken away.
- Professional worksheet. The result includes the inputs used, the method and scope, and a one-variable sensitivity check for finance calculators.
- A/B scenarios. Save two input sets in the browser and compare the primary, secondary, and breakdown figures side by side.
- Print / PDF and CSV. Print a branded worksheet, or export the year-by-year table when the calculator produces one.
The formulas and assumptions above still govern the number. The worksheet makes the estimate auditable; it does not make it a lender quote, tax return, or personalized advice.
Go deeper
