← GMAT Prep

GMAT Probability and Combinatorics: Rules, Worked Examples, and Practice

Probability and counting questions on the GMAT Focus Edition come down to a handful of rules: multiply for "and," add for "or," use the complement for "at least one," and decide whether order matters before you count. This guide covers each rule, the traps the answer choices are built around, six fully worked Problem Solving examples, and four practice questions.

What GMAT Focus Quant Tests in Probability and Counting

GMAC lists "probability rules and combinatorics" among the topics in Quantitative Reasoning, and the Official Guide's math review has a section on statistics, sets, counting, probability, estimation, and series. On the Focus Edition these topics appear only in the Quant section, which has 21 Problem Solving questions in 45 minutes, five answer choices each, and no calculator. Focus Quant has no geometry and no Data Sufficiency (Data Sufficiency now sits in Data Insights). The GMAT Quantitative Reasoning overview covers the rest of the section.

GMAC doesn't publish how many questions come from each topic, so treat probability and counting as one topic among many: worth learning well, not worth months on its own. The good news is that the tested material is narrow. Nearly every question uses the ideas in this table.

The ideas behind GMAT Focus probability and counting questions, and the tool each one calls for.
IdeaHow the question signals itCore tool
Basic probability"chosen at random," equally likely outcomesfavorable outcomes ÷ total outcomes
Independent events (AND)coins, dice, "with replacement," separate machinesmultiply the probabilities
Dependent events (AND)"without replacement," drawing people or items one after anothermultiply, updating each fraction
Either/or (OR)"A or B"add, then subtract the overlap if both can happen
At least one"at least one," "one or more"1 − P(none)
Counting principlecodes, outfits, routes, stages of a choicemultiply the options at each stage
Permutationsorder matters: rankings, seats, codes, officer rolesnPk = n! ÷ (n − k)!
Combinationsorder doesn't matter: committees, teams, handshakesnCk = n! ÷ (k!(n − k)!)
Restrictions"must sit together," "can't be first," "at least one analyst"glue items into a block, fill the restricted spot first, or count the complement

Probability Rules and When to Use Them

Probability is a number from 0 (impossible) to 1 (certain). When every outcome is equally likely, P(event) = favorable outcomes ÷ total outcomes. Everything else is a way of combining events.

Five probability rules cover nearly every GMAT probability question.
RuleFormulaUse it when
Complement (NOT)P(not A) = 1 − P(A)The question says "at least one," or "not" is easier to count
AND, independentP(A and B) = P(A) × P(B)One event doesn't change the other (coin flips, dice, with replacement)
AND, dependentP(A and B) = P(A) × P(B given A)The first draw changes what's left (without replacement)
OR, can't both happenP(A or B) = P(A) + P(B)A and B are mutually exclusive (one die roll can't be both 2 and 5)
OR, can both happenP(A or B) = P(A) + P(B) − P(A and B)Outcomes can belong to both groups (multiples of 3 or 5)

Two ways to solve "without replacement"

For a question such as "two marbles are drawn without replacement; what is the probability both are red?" you can multiply step by step (4/10 × 3/9) or count groups (red pairs ÷ all pairs, using combinations). Both give the same answer. Step-by-step multiplication is usually faster when the question names an order or a small number of draws; counting with combinations is cleaner when the question asks for "exactly 2 of 3" or mixes groups.

"Exactly" means count the arrangements

If you multiply probabilities for "exactly two women out of three," the product 4/10 × 3/9 × 6/8 covers only one order (woman, woman, man). There are three orders, so multiply by 3. Forgetting that factor is one of the most common wrong answers. Worked Example 6 shows both methods.

Rule of thumb: "And" multiplies, "or" adds, "at least one" flips to 1 − P(none). If an "or" answer comes out above 1, or an "and" answer comes out larger than either single probability, you've used the wrong rule.

Counting: Permutations and Combinations

Counting questions ask "how many ways?" Start with the counting principle: if a choice happens in stages, multiply the number of options at each stage. A code with 3 letters followed by 2 digits has 26 × 26 × 26 × 10 × 10 possibilities. Permutations and combinations are shortcuts for the special case of choosing k items from n distinct items without repeats.

Counting formulas, with the example each one fits. A combination count is the matching permutation count divided by k!.
SituationFormulaExample
Arrange all n distinct itemsn!5 books on a shelf: 5! = 120
Choose k of n, order mattersnPk = n! ÷ (n − k)!President and VP from 8 people: 8 × 7 = 56
Choose k of n, order doesn't matternCk = n! ÷ (k!(n − k)!)3-person team from 7 people: 7C3 = 35
Arrange items with repeatsn! ÷ (a! × b! × …)Letters of LEVEL: 5! ÷ (2! × 2!) = 30
Choose from two groupsmultiply the separate counts2 of 4 analysts and 1 of 5 managers: 6 × 5 = 30

The swap test: does order matter?

Pick two items in a possible selection and swap them. If the swap produces a different outcome (Ana as president and Ben as VP is different from the reverse), order matters, so use a permutation. If the swap produces the same outcome (a committee of Ana and Ben is the same committee either way), use a combination. Notice that 8P3 = 336 and 8C3 = 56: the permutation count is 3! = 6 times larger because each group of three can be ordered six ways.

Values worth knowing without a calculator

Quant has no calculator, so cancel before you multiply (8C3 = (8 × 7 × 6) ÷ (3 × 2 × 1) = 56) and know the small values: 4! = 24, 5! = 120, 6! = 720, 7! = 5,040; 5C2 = 10, 6C2 = 15, 6C3 = 20, 7C3 = 35, 8C3 = 56, 9C3 = 84, 10C3 = 120. Also remember nCk = nC(n − k), so 10C7 = 10C3. The GMAT math formulas cheat sheet has these alongside the other Quant formulas.

Handling restrictions

  • "Must be together": glue the items into one block, arrange the blocks, then multiply by the arrangements inside the block.
  • "Can't be together" or "at least one": count everything, then subtract the cases you don't want.
  • A position with a rule (first digit can't be 0, the chair must be a senior): fill the restricted position first, then the rest.

Common Traps in GMAT Probability and Counting

Wrong answer choices are usually the result of a specific, predictable mistake. These are the ones to check for before you click.

Six traps that wrong answer choices are designed around. Overlapping-groups questions have their own guide.
TrapWhat it looks likeFix
Treating a draw as independentUsing 4/10 × 4/10 when marbles aren't replacedReduce the numerator and denominator after each draw
Ordering a groupCounting a 3-person committee as 9 × 8 × 7Divide by 3! or use 9C3
Adding instead of using the complement"At least one fails" = 0.1 + 0.1 + 0.11 − P(none fail)
Double-counting an overlapMultiples of 3 or 5: 10 + 6 = 16 out of 30Subtract the numbers counted twice (multiples of 15)
Missing the orders in "exactly"Exactly 2 of 3 = one product with no multiplierMultiply by the number of orders, or count with combinations
Answering a different questionFinding "exactly one" when asked "at least one"Reread the question stem before choosing

The overlap trap is the same idea that drives two-group Venn questions; the GMAT overlapping sets guide covers that format in depth.

Six Worked Examples

Each example is in the Focus Quant format: a Problem Solving question with five answer choices. Try it before you read the solution.

Worked Example 1 — Without replacement

A jar contains 4 red marbles and 6 blue marbles. Two marbles are drawn at random without replacement. What is the probability that both marbles are red?

  1. 4/25
  2. 2/15
  3. 1/5
  4. 4/15
  5. 2/5

Solution

  1. First draw: 4 of the 10 marbles are red, so P(red) = 4/10.
  2. Second draw: one red marble is gone, leaving 3 red out of 9, so P(red again) = 3/9.
  3. Multiply: 4/10 × 3/9 = 12/90 = 2/15.
  4. Check with counting: red pairs ÷ all pairs = 4C2 ÷ 10C2 = 6 ÷ 45 = 2/15.
  5. Choice A (4/25 = 4/10 × 4/10) is the with-replacement trap.
Answer: (B) 2/15

Worked Example 2 — At least one (complement)

Each part made by a machine passes inspection with probability 0.9, independently of every other part. If 3 parts are inspected, what is the probability that at least one of them fails?

  1. 0.100
  2. 0.243
  3. 0.271
  4. 0.300
  5. 0.729

Solution

  1. "At least one fails" is the complement of "none fail."
  2. P(all 3 pass) = 0.9 × 0.9 × 0.9 = 0.729.
  3. P(at least one fails) = 1 − 0.729 = 0.271.
  4. Traps: 0.300 adds 0.1 three times; 0.243 is the probability that exactly one fails (3 × 0.1 × 0.81); 0.729 is the probability that none fail.
Answer: (C) 0.271

Worked Example 3 — Or, with overlap

An integer is chosen at random from the integers 1 through 30, inclusive. What is the probability that it is a multiple of 3 or a multiple of 5?

  1. 1/3
  2. 7/15
  3. 8/15
  4. 1/2
  5. 3/5

Solution

  1. Multiples of 3 from 1 to 30: 30 ÷ 3 = 10.
  2. Multiples of 5: 30 ÷ 5 = 6.
  3. Multiples of both (multiples of 15): 15 and 30, so 2. These were counted twice.
  4. Favorable outcomes: 10 + 6 − 2 = 14, so P = 14/30 = 7/15.
  5. Choice C (16/30 = 8/15) forgets to subtract the overlap.
Answer: (B) 7/15

Worked Example 4 — Combinations with "at least one"

A 3-person committee is to be chosen from 5 managers and 4 analysts. How many different committees include at least 1 analyst?

  1. 40
  2. 64
  3. 74
  4. 80
  5. 84

Solution

  1. Order doesn't matter (a committee), so use combinations.
  2. All committees: 9C3 = (9 × 8 × 7) ÷ (3 × 2 × 1) = 84.
  3. Committees with no analyst (all managers): 5C3 = 10.
  4. At least 1 analyst: 84 − 10 = 74.
  5. Check directly: exactly 1 analyst = 4C1 × 5C2 = 4 × 10 = 40; exactly 2 = 4C2 × 5C1 = 6 × 5 = 30; exactly 3 = 4C3 = 4. Total 40 + 30 + 4 = 74. Choice A (40) counts only "exactly 1."
Answer: (C) 74

Worked Example 5 — Arrangement with a "together" restriction

Five people, including Ana and Ben, are to be seated in a row of 5 chairs. In how many different arrangements are Ana and Ben seated next to each other?

  1. 24
  2. 48
  3. 60
  4. 72
  5. 120

Solution

  1. Glue Ana and Ben into one block. Now there are 4 units to arrange: the block plus 3 other people.
  2. Arrange the 4 units: 4! = 24.
  3. Inside the block, Ana and Ben can sit in 2 orders (AB or BA).
  4. Total: 24 × 2 = 48.
  5. Traps: 24 forgets the 2 orders inside the block; 120 is every arrangement; 72 (120 − 48) is the number of arrangements where they are not together.
Answer: (B) 48

Worked Example 6 — Probability by counting ("exactly")

A group consists of 6 men and 4 women. If 3 people are selected at random from the group, what is the probability that exactly 2 of the people selected are women?

  1. 1/10
  2. 1/5
  3. 3/10
  4. 2/5
  5. 1/2

Solution

  1. Total groups of 3: 10C3 = 120.
  2. Groups with exactly 2 women and 1 man: 4C2 × 6C1 = 6 × 6 = 36.
  3. P = 36/120 = 3/10.
  4. Check by multiplying: one order (W, W, M) has probability 4/10 × 3/9 × 6/8 = 1/10. There are 3 orders (WWM, WMW, MWW), so P = 3 × 1/10 = 3/10.
  5. Choice A (1/10) is the single-order trap.
Answer: (C) 3/10

Practice Questions

Choose an answer, then check it. Each explanation names the trap behind the most tempting wrong choice. For full timed sets across every Quant topic, use the GMAT quantitative practice questions.

Practice 1 — Two dice
Two fair six-sided dice are rolled. What is the probability that the sum of the two numbers rolled is 8?
Practice 2 — Repeated letters
How many different arrangements of the letters in the word LEVEL are possible?
Practice 3 — Code with a restriction
A 4-digit code is formed from the digits 0 through 9. No digit may be used more than once, and the first digit cannot be 0. How many different codes are possible?
Practice 4 — At least one, without replacement
A box contains 10 light bulbs, 3 of which are defective. If 2 bulbs are selected at random without replacement, what is the probability that at least 1 of the selected bulbs is defective?

Frequently Asked Questions

Is probability tested on the GMAT Focus Edition?

Yes. GMAC lists probability rules and combinatorics among the Quantitative Reasoning topics, and the Official Guide's math review covers counting and probability. On the Focus Edition these appear only as Problem Solving questions in the Quant section (21 questions, 45 minutes, no calculator). GMAC does not publish how many questions come from each topic.

How do I know whether to use a permutation or a combination?

Ask whether order matters. Swap two of the chosen items: if the swap creates a different outcome (a different ranking, seating, or code), use a permutation, nPk = n!/(n − k)!. If it is the same outcome (the same committee or group), use a combination, nCk = n!/(k!(n − k)!).

What is the fastest way to solve "at least one" probability questions?

Use the complement: P(at least one) = 1 − P(none). Working out "exactly one," "exactly two," and so on separately is slower and easier to get wrong.

Which counting formulas should I memorize for the GMAT?

The counting principle (multiply the choices at each stage), nPk = n!/(n − k)!, nCk = n!/(k!(n − k)!), n!/(a!b!...) for arrangements with repeated items, and the probability rules: multiply for AND (adjusting for dependence), add for OR (subtracting any overlap), and 1 − P(A) for NOT. Because Quant has no calculator, also know small factorials such as 5! = 120 and 6! = 720.