Salam Academy
Poisson and binomial distributions · Subtopic 1 of 4

The Binomial Distribution

Count the successes in nn independent trials, each with the same chance pp. The probability of exactly rr successes is (nr)pr(1−p)n−r\binom nr p^r(1-p)^{n-r}: choose which trials succeed, then multiply.

Cambridge 9709 Paper 5Edexcel 9MA0 Paper 3Edexcel 9FM0 OptionAQA 7357 Paper 3≈ 40 minRevision 6 · updated 2026-10-27
Section 1 of 14

Learning objectives

  • State the conditions for a binomial model, X∼B(n,p)X\sim B(n,p).
  • Calculate P(X=r)=(nr)prqn−rP(X=r)=\binom nr p^r q^{n-r}, where q=1−pq=1-p.
  • Find cumulative probabilities such as P(X≤2)P(X\le2) and P(X≥1)P(X\ge1).
  • Use E(X)=npE(X)=np and Var(X)=npq\text{Var}(X)=npq.
Pause and think

A student guesses all 44 answers on a multiple-choice quiz, each with 44 options. Why is the probability of getting exactly 11 right not just 14×(34)3\tfrac14\times\left(\tfrac34\right)^3?

Section 2 of 14

Core idea

Start hereTen free throws

A basketball player makes 30% of free throws. In 10 attempts, what's the chance of exactly 3 baskets? Each throw is a yes/no trial with the same chance, independent of the others. That pattern is so common it has its own distribution: the binomial.

The big idea

X∼B(n,p)X\sim B(n,p) counts successes in nn independent trials with constant success probability pp. P(X=r)=(nr)pr(1−p)n−rP(X=r)=\binom nrp^r(1-p)^{n-r}; E(X)=npE(X)=np; Var⁡(X)=np(1−p)\operatorname{Var}(X)=np(1-p).

1. Conditions

  • Fixed number of trials nn.
  • Each trial: success or failure.
  • Constant probability pp.
  • Independent trials.

2. The formula

P(X=r)P(X=r)
  1. one order (e.g. SSS FFFFFFF): pr(1−p)n−r\text{one order (e.g. }SSS\,FFFFFFF):\ p^r(1-p)^{n-r}
    Independent: multiply.
  2. number of orders=(nr)\text{number of orders}=\binom nr
    Choose which rr trials succeed.
  3. P(X=r)=(nr)pr(1−p)n−rP(X=r)=\binom nrp^r(1-p)^{n-r}
    Each order equally likely; add them.
B(10, p): the shape as p changes
2468100.10.20.30.40number of successes rP(X = r)mean np
Small pp: the bars pile up near 00. p=0.5p=0.5: symmetric. Large pp: piled up near 1010. The peak is always near the mean npnp.
Worked exampleX∼B(10,0.3)X\sim B(10,0.3)

Find P(X=3)P(X=3), P(X≤2)P(X\le2), P(X≥4)P(X\ge4), mean and variance.

Goal
Values.
Strategy
Formula; complement for "at least".
  1. P(X=3)=120(0.027)(0.08235)=0.267P(X=3)=120(0.027)(0.08235)=0.267
    (103)=120\binom{10}3=120.
  2. P(X≤2)=0.0282+0.1211+0.2335=0.383P(X\le2)=0.0282+0.1211+0.2335=0.383
    r=0,1,2r=0,1,2.
  3. P(X≥4)=1−(0.383+0.267)=0.350P(X\ge4)=1-(0.383+0.267)=0.350
    Complement of X≤3X\le3.
  4. E=3, Var⁡=2.1E=3,\ \operatorname{Var}=2.1
    npnp, np(1−p)np(1-p).
Answer0.2670.267; 0.3830.383; 0.3500.350; 33; 2.12.1.
✓ CheckThe most likely value is 3=E(X)3=E(X) ✓.
Worked exampleFinding n

How many rolls of a die so that P(at least one six) > 0.9?

Goal
Least nn.
Strategy
Complement; logs.
  1. 1−(56)n>0.9⇒(56)n<0.11-(\tfrac56)^n>0.9\Rightarrow(\tfrac56)^n<0.1
    Rearrange.
  2. n>ln⁡0.1ln⁡(5/6)=12.6n>\frac{\ln0.1}{\ln(5/6)}=12.6
    ln⁡56<0\ln\frac56<0: flip.
Answer1313 rolls.
✓ Check(56)13=0.093<0.1(\frac56)^{13}=0.093<0.1 ✓; (56)12=0.112(\frac56)^{12}=0.112 ✗.
Common misconceptionWithout replacement Drawing cards without replacement changes pp: not binomial.
Pause and think

Why isn't "number of aces in 5 cards from a pack" binomial?

Summary

  • Four conditions.
  • (nr)prqn−r\binom nrp^rq^{n-r}.
  • npnp, npqnpq.
Key words
Trial
One repetition.
Binomial distribution
Number of successes in n independent trials.
X ~ B(10, 0.3)
2468100.050.10.150.20.250.30number of successes rP(X = r)most likely: r = 3
Each bar is P(X=r)=(10r)(0.3)r(0.7)10−rP(X=r)=\binom{10}{r}(0.3)^r(0.7)^{10-r}. The distribution peaks near the mean np=3np=3 and is skewed to the right because p<0.5p<0.5.
HistoryBernoulli's Art of Conjecturing. Jacob Bernoulli worked for twenty years on Ars Conjectandi ("The Art of Conjecturing"), published in 1713, eight years after his death. It analysed repeated success-or-failure trials (now called Bernoulli trials) and proved the first law of large numbers: over many trials, the proportion of successes settles down to pp.
Fun factThe bean machine. Francis Galton built a board in which balls bounce off rows of pins, going left or right with equal probability at each pin. The slot a ball lands in follows a binomial distribution, and thousands of balls pile up into the familiar hump shape. Desktop versions are sold as toys.
Keep reading for free

Create a free account to read the rest of this lesson: key formulae, worked examples, explore the graph, common mistakes, exam tip, guided practice and more. Your answers are checked and your progress is saved.