Visual Tools
Calculators
Tables
Mathematical Keyboard
Converters
Other Tools

How to Find a Binomial Probability

A binomial variable counts the successes in a fixed number of independent trials that all have the same success probability. Write the event as a list of counts kk, find each one with the PMF, and add them; for "at least" questions, subtract the small side from 11.

Recognize the task

The problem fixes the number of trials in advance and asks how many of them succeed: "1010 free throws, P(at least 8 made)P(\text{at least } 8 \text{ made})", "66 coin tosses, exactly 22 heads", "2020 parts, at most 11 defective". Test the four conditions: a fixed nn, two outcomes per trial, independent trials, the same pp on every trial. If all four hold, X∼Bin⁡(n,p)X \sim \operatorname{Bin}(n, p).
The model and its conditions are set out in the binomial distribution lesson.

Ways to do it

Add the PMF termsstandard
P(a≤X≤b)=∑k=ab(nk)pk(1−p)n−kP(a \le X \le b) = \sum_{k=a}^{b} \binom{n}{k} p^k (1-p)^{n-k}
Evaluate the binomial PMF at each count in the event and add. Best when the event holds only a few values.
Use the complementat least
P(X≥k)=1−P(X≤k−1)P(X \ge k) = 1 - P(X \le k-1)
When the event holds most of the counts, add the few counts outside it and subtract from 11 (complement rule).
Approximate a large binomiallarge n
Bin⁡(n,p)≈Poisson⁡(np) or N(np, np(1−p))\operatorname{Bin}(n, p) \approx \operatorname{Poisson}(np) \ \text{or} \ N\big(np,\, np(1-p)\big)
With many trials and a long sum, use the Poisson approximation for small pp or the normal approximation otherwise.

Steps

StepOn 55 free throws, p=0.7p = 0.7, P(X≥4)P(X \ge 4)
1Check the binomial conditions. Fixed number of trials, two outcomes, independent trials, the same success probability each time.55 shots, made or missed, independent, p=0.7p = 0.7 ✓
2Name nn, pp and the success. Define XX as the number of successes. The failure probability is 1−p1 - p.X∼Bin⁡(5, 0.7)X \sim \operatorname{Bin}(5,\, 0.7), 1−p=0.31 - p = 0.3
3Turn the event into a list of counts. "At least 44" is {4,5}\{4, 5\}, "more than 44" is {5}\{5\}, "at most 11" is {0,1}\{0, 1\}. If the list is long, take the complement instead.P(X≥4)=P(X=4)+P(X=5)P(X \ge 4) = P(X = 4) + P(X = 5)
4Evaluate the PMF at each count. Use (nk)pk(1−p)n−k\binom{n}{k} p^k (1-p)^{n-k}. The two exponents always add up to nn.(54)(0.7)4(0.3)=0.36015\binom{5}{4}(0.7)^4(0.3) = 0.36015,  (0.7)5=0.16807\ (0.7)^5 = 0.16807
5Add, or subtract from 11. Add the terms of the event. With the complement, subtract their sum from 11. Round at the end.P(X≥4)=0.52822≈0.5282P(X \ge 4) = 0.52822 \approx 0.5282

Worked examples

Example 1 · exactly k successes

one term
66 fair coin tosses, P(exactly 2 heads)P(\text{exactly } 2 \text{ heads})
X∼Bin⁡(6,12)X \sim \operatorname{Bin}(6, \frac{1}{2})
(62)(12)2(12)4=1564\binom{6}{2}\left(\frac{1}{2}\right)^2\left(\frac{1}{2}\right)^4 = \frac{15}{64}
(62)=15\binom{6}{2} = 15, 26=642^6 = 64
P(X=2)=1564≈0.2344P(X = 2) = \frac{15}{64} \approx 0.2344

Example 2 · at most k successes

short sum
88 quiz guesses, p=14p = \frac{1}{4}, P(X≤2)P(X \le 2)
counts 0,1,20, 1, 2
(34)8+8⋅14(34)7+28⋅116(34)6\left(\frac{3}{4}\right)^8 + 8 \cdot \frac{1}{4}\left(\frac{3}{4}\right)^7 + 28 \cdot \frac{1}{16}\left(\frac{3}{4}\right)^6
one PMF term per count
=6561+17496+2041265536=4446965536= \frac{6561 + 17496 + 20412}{65536} = \frac{44469}{65536}
common denominator 484^8
P(X≤2)≈0.6785P(X \le 2) \approx 0.6785

Example 3 · at least one success

complement
1010 parts, each defective with p=0.1p = 0.1, P(X≥1)P(X \ge 1)
nine counts in the event
P(X≥1)=1−P(X=0)P(X \ge 1) = 1 - P(X = 0)
one count outside it
1−(0.9)10=1−0.348681 - (0.9)^{10} = 1 - 0.34868
P(X=0)=(1−p)nP(X = 0) = (1-p)^n
P(X≥1)≈0.6513P(X \ge 1) \approx 0.6513
Check any binomial probability in the discrete distributions calculator.

See it happen

Each way to get exactly kk successes is one ordering of kk successes and n−kn - k failures, and every ordering has the same probability pk(1−p)n−kp^k (1-p)^{n-k}. The coefficient (nk)\binom{n}{k} counts the orderings. Adding over the counts in an event adds the heights of the bars at those counts. Move nn and pp and watch the bars shift in the binomial distribution visualizer.

Common mistakes

Dropping the binomial coefficient

✗ P(X=4)=(0.7)4(0.3)=0.07203P(X = 4) = (0.7)^4(0.3) = 0.07203
✓ there are (54)=5\binom{5}{4} = 5 orders: 5⋅0.07203=0.360155 \cdot 0.07203 = 0.36015

Reading "more than" as "at least"

✗ P(X>4)=P(X=4)+P(X=5)P(X > 4) = P(X = 4) + P(X = 5)
✓ P(X>4)=P(X=5)=0.16807P(X > 4) = P(X = 5) = 0.16807; only "at least" includes 44

Wrong complement

✗ P(X≥3)=1−P(X≤3)P(X \ge 3) = 1 - P(X \le 3)
✓ P(X≥3)=1−P(X≤2)P(X \ge 3) = 1 - P(X \le 2); the value 33 belongs to the event

Binomial for draws without replacement

✗ 33 people from 44 women and 66 men: Bin⁡(3,0.4)\operatorname{Bin}(3, 0.4)
✓ the chance changes after each draw, so use the hypergeometric model