Is the geometric distribution on the AP Stats exam?

Q: Do I need the geometric probability formula for the May 2027 exam?

No. You need the binomial formula, P(X = x) = C(n, x)p^x(1-p)^n-x, which is topic 2.10 and is printed on the formula sheet. The geometric formula is not on the formula sheet and is not examined.

By Jude Wallis · Published August 11, 2026

No. The geometric distribution was removed when College Board redesigned AP Statistics for Fall 2026, first examined in May 2027. It was topic 4.12 in the old nine-unit course. The binomial distribution stayed, as topic 2.10, and it is the one the exam still asks about.

AP Statistics: Unit 2 (topics 2.10 The Binomial Distribution). The binomial distribution is Unit 2 topic 2.10 of the Fall 2026 AP Statistics course. The geometric distribution was topic 4.12 in the previous course and is listed as removed on College Board's AP Statistics future revisions page, so it is not assessed from the May 2027 exam onward.

No, the geometric distribution was removed

The redesigned AP Statistics course takes effect in Fall 2026 and is first examined in May 2027, and the geometric distribution is not in it. College Board names it directly in the list of cut topics on its AP Statistics future revisions page, where it was topic 4.12 of the course this one replaces.

The geometric distribution is the cleanest case on that list. It is simply gone, and nothing in the five-unit framework takes its place. If you are studying for the May 2027 exam or later, you do not need it.

It is one of five entries on that list. What was removed from AP Statistics covers the other four, including the one about departures from linearity that is routinely over-read.

What a geometric random variable is

A geometric random variable counts how many independent trials you run up to and including the first success, when every trial has the same success probability $p$ .

If $X$ is the trial the first success lands on, then $X$ can be $1, 2, 3, \ldots$ with no upper limit, and

P(X = x) = (1-p)^{x-1}\, p

because you need $x-1$ failures first, each with probability $1-p$ , then one success. The mean is $\mu_X = \frac{1}{p}$ and the standard deviation is $\sigma_X = \frac{\sqrt{1-p}}{p}$ . Those are the standard college forms; the AP formula sheet does not list them.

The short version: a geometric setting asks how long you wait. The geometric distribution glossary entry works through a die example if you want the fuller picture.

What the exam asks instead: the binomial distribution

The binomial distribution is very much on the exam. It is topic 2.10, The Binomial Distribution, inside Unit 2, Probability, Random Variables, and Probability Distributions.

A binomial setting fixes the number of trials at $n$ ahead of time and counts how many successes occur. For exactly $x$ successes,

P(X = x) = \binom{n}{x} p^x (1-p)^{n-x}

with mean $\mu_X = np$ and standard deviation $\sigma_X = \sqrt{np(1-p)}$ . That version is printed on the formula sheet you get during the exam.

So the pair of distributions that used to bracket each other on the old exam is now a single distribution. Where a question once might have asked you to decide between them, it now asks you to recognize a binomial setting and work inside it. The topic page for 2.10 The Binomial Distribution covers what that looks like, and how to tell if a situation is binomial is the skill the exam actually tests.

Telling the two apart, if you still want to

The distinction has not stopped being useful. It has stopped being examined.

Ask one question: is the number of trials fixed, or is it the thing you are solving for?

Fixed number of trials, count the successes. Binomial. Phrases like "in 10 attempts" or "out of 20 patients" hand you $n$ .
Keep going until the first success. Geometric. Phrases like "until the first" or "the first defective is the 4th one" mean the trial number is the unknown.

Binomial vs geometric is the full comparison, with the conditions the two share, a side-by-side table, and worked examples of each. This page is the shorter answer to the exam question.

Why it is still worth knowing

Removing a topic from an exam does not remove it from statistics. The geometric distribution is standard content in a one-semester college introductory statistics course, which is the course AP Statistics is meant to stand in for. If you take that course after your AP credit, you are likely to meet it there.

It is also the natural model for a class of real questions the binomial cannot answer: how many calls before one converts, how many components before one fails, how many attempts before a first success. Those are waiting-time questions, and the binomial has no way to express them.

So the honest advice is about priority, not value. If you are revising for the exam with limited time, put it into topic 2.10 and leave the geometric distribution alone. If you are learning statistics because you want to use it, learn both.

Why your review book probably still covers it

Most AP Statistics material online and in print describes the old nine-unit course, where the geometric distribution was topic 4.12 and inference for regression slopes was a whole unit. Those resources are not wrong about statistics. They are aimed at an exam that no longer exists.

Before you trust a review book, a video, or a unit guide, check which version of the course it was built for. A quick test: if it lists nine units, or if it teaches a chi-square goodness-of-fit test as examinable, it predates the redesign.

The new exam format in full sets out the current structure, and what is on the AP Statistics exam maps the content you are actually responsible for.

The binomial version, which the exam does ask

A spinner lands on red on 20% of spins, and spins are independent. You spin it 5 times. What is the probability of exactly 1 red?

Classify the setting. The number of trials is fixed at $n = 5$ and you count successes, so this is binomial with $n = 5$ and $p = 0.20$ .
Write the formula for exactly $x = 1$ success: $P(X = 1) = \binom{5}{1}(0.20)^1(0.80)^4$ .
Count the combinations: $\binom{5}{1} = 5$ .
Raise to the power: $(0.80)^2 = 0.64$ , so $(0.80)^4 = 0.64^2 = 0.4096$ .
Multiply: $5 \times 0.20 = 1$ , and $1 \times 0.4096 = 0.4096$ .

$P(X = 1) = 0.4096$ , about a 41.0% chance of exactly one red in 5 spins. The mean number of reds is $\mu_X = np = 5 \times 0.20 = 1$ .

The geometric version, which it does not

The same spinner lands on red on 20% of spins. You keep spinning until the first red. What is the probability that the first red is the 3rd spin, and how many spins does it take on average? This calculation is not assessed on the redesigned AP exam; it is here so you can see what was cut.

Classify the setting. The number of trials is not fixed and you are waiting for the first success, so this is geometric with $p = 0.20$ .
You want the first success on trial $x = 3$ , so $P(X = 3) = (1 - 0.20)^{3-1}(0.20) = (0.80)^2(0.20)$ .
Raise to the power: $(0.80)^2 = 0.64$ .
Multiply by $p$ : $0.64 \times 0.20 = 0.128$ .
Find the mean with $\mu_X = 1/p = 1/0.20 = 5$ .

$P(X = 3) = 0.128$ , about a 12.8% chance the first red is the 3rd spin, and the first red arrives on spin 5 on average. Same spinner, same $p$ , a question the redesigned exam will not ask.

Frequently asked questions

Was the geometric distribution ever on the AP Statistics exam?

Yes. It was topic 4.12 in the course the Fall 2026 redesign replaced. College Board's future revisions page lists it among the topics removed, and that removal applies to the May 2027 exam and every exam after it.

Do I need the geometric probability formula for the May 2027 exam?

No. You need the binomial formula, $P(X = x) = \binom{n}{x}p^x(1-p)^{n-x}$ , which is topic 2.10 and is printed on the formula sheet. The geometric formula is not on the formula sheet and is not examined.

What replaced the geometric distribution in Unit 2?

Nothing replaced it directly. Unit 2 runs from two-way tables and probability rules through random variables (2.8 and 2.9), the binomial distribution (2.10), the normal distribution (2.11), and sampling distributions with the central limit theorem (2.12). The waiting-time idea has no topic of its own.

Is the geometric distribution still worth learning?

For the AP exam, no. For statistics, yes. It is standard in college introductory statistics and it is the natural model for waiting-time questions, such as how many attempts happen before a first success, which a binomial cannot express.

Why does my review book still teach it?

Because it was written for the nine-unit course. Most AP Statistics material still describes that version. If a resource lists nine units or teaches chi-square goodness of fit as examinable, it predates the Fall 2026 redesign.