GMAT Probability and Combinations: Complete Strategy Guide
Quick Takeaways
- The only question that matters: Does order matter? Yes → Permutation (or slot method). No → Combination.
- 'At least one' reflex: Always use 1 − P(none). Direct calculation is slower and more error-prone.
- AND rule: Multiply probabilities of independent events.
- OR rule (mutually exclusive): Add probabilities.
- Glue method: Treat items that must stay together as a single unit.
Why These Questions Feel Hard
Most students find probability and combinatorics questions disproportionately difficult relative to their underlying mathematical complexity. The formulas involved — nCr, nPr, P = Favourable/Total — are not advanced. The difficulty is in knowing which formula to apply, recognising what type of counting problem you are facing, and avoiding the small number of specific conceptual errors that the GMAT repeatedly exploits.
This guide is structured to address those specific difficulties directly. It starts with the decision framework that governs which method to use, covers the four core probability rules, introduces three advanced techniques for harder questions, and then demonstrates all of it through five worked examples. The goal is not just that you can follow the solutions — it is that you can recognise which technique applies within the first 10 seconds of reading the question.
Formula Reference: Counting and Probability at a Glance
| Method | When to Use | Formula | Example |
|---|---|---|---|
| Permutation (nPr) | Arrangements where order matters — ranking, assigning specific roles, sequences | nPr = n! / (n − r)! | Top-3 finishers from 8 runners: 8P3 = 8 × 7 × 6 = 336 |
| Slot method | Arrangements with specific constraints — often faster than nPr for restricted problems | Multiply choices available for each slot | 4-digit code, no repeated digits: 10 × 9 × 8 × 7 = 5,040 |
| Combination (nCr) | Selections where order does not matter — committees, teams, subsets | nCr = n! / [r!(n − r)!] | Committee of 3 from 8: 8C3 = 56 |
| Basic probability | Any probability question | P(Event) = Favourable outcomes / Total outcomes | P(red card from standard deck) = 26/52 = 1/2 |
| Complement rule | 'At least one' questions, or when direct calculation requires many cases | P(Event) = 1 − P(Not Event) | P(at least one head in 3 flips) = 1 − (1/2)³ = 7/8 |
| AND rule | Both of two independent events occur | P(A and B) = P(A) × P(B) | P(two heads in a row) = 1/2 × 1/2 = 1/4 |
| OR rule (mutually exclusive) | Either of two events occurs, and they cannot both occur simultaneously | P(A or B) = P(A) + P(B) | P(roll a 2 or a 5 on a die) = 1/6 + 1/6 = 1/3 |
Combinations vs Permutations: The Order Question
Before applying any counting formula, answer one question: does the order of selection matter? This single decision determines your entire approach.
- Order matters → Permutation. The arrangement or sequence of items is significant. Assigning people to specific roles (President, Vice-President, Treasurer), creating a code where position matters, or ranking competitors. Arrangement A-B-C is different from C-B-A.
- Order does not matter → Combination. You are choosing a group where the membership is what counts, not the sequence of selection. Picking a committee, forming a team, selecting toppings. Team of {Alice, Bob, Carol} is identical regardless of the order they were chosen.
- Shortcut — look at the question's framing: 'How many ways can X be arranged' or 'how many different sequences' signals permutation. 'How many groups', 'how many committees', or 'how many subsets' signals combination.
The slot method alternative: For many permutation problems — especially those with restrictions — the slot method is faster than the nPr formula. Draw a blank for each position, fill in the number of available choices for each slot (applying restrictions as you go), and multiply. It is more intuitive than factorials and adapts naturally to constrained problems.
The Four Rules of Probability
Every GMAT probability question uses one or more of four foundational rules. Recognising which rule applies is faster than re-deriving logic from scratch on each question.
- Rule 1 — Basic probability: P(Event) = Favourable outcomes ÷ Total outcomes. All outcomes must be equally likely for this formula to apply directly.
- Rule 2 — AND rule (independent events): P(A and B) = P(A) × P(B). Independent means the outcome of A does not affect the probability of B. Two coin flips are independent; drawing two cards without replacement is not.
- Rule 3 — OR rule (mutually exclusive events): P(A or B) = P(A) + P(B). Mutually exclusive means A and B cannot both occur in the same trial. For non-mutually exclusive events (where both can occur), use: P(A or B) = P(A) + P(B) − P(A and B).
- Rule 4 — Complement rule: P(A) = 1 − P(not A). This is the most underused rule and the fastest path for 'at least one' questions. Instead of computing P(exactly 1) + P(exactly 2) + ... + P(exactly n), compute 1 − P(zero occurrences).
The 'at least one' reflex: Every time you see 'at least one' in a probability question, immediately set up the complement: P(at least one) = 1 − P(none). This reduces a multi-case calculation to a single step. On the GMAT, where the direct approach might require 3–4 separate calculations, the complement almost always saves 30–45 seconds.
Three Advanced Techniques for Hard Questions
Three specific techniques account for the majority of 650–750 level combinatorics questions. They are not complicated, but they require recognition before application — you need to see the pattern in the question before you can choose the right tool.
- The glue method — for 'items must stay together' constraints. Treat the constrained items as a single unit. If Alice and Bob must always sit together in a row of 6 people, glue them: you now have 5 units to arrange (5! = 120 ways), and within the glued unit, Alice and Bob can swap (2! = 2 ways). Total = 120 × 2 = 240 arrangements.
- The complement method — for 'items must NOT be together.' Find the total arrangements without the constraint, subtract the arrangements where the forbidden items ARE together (using the glue method). This is faster than directly counting all valid arrangements.
- The matrix/grid method — for two-variable probability problems. When a problem involves two independent variables each with multiple outcomes (e.g., rolling two dice), draw a grid of all possible outcomes and count the favourable cells. This is slower than algebra but eliminates the counting errors that algebraic shortcuts sometimes produce on complex problems.
Five Fully Worked Examples
Work through each question before reading the solution. Identifying the technique before calculating is the skill to practise — not the arithmetic.
Example 1 — Combination: selecting a committee
Question: A company wants to select a committee of 3 from a team of 8 engineers. How many different committees are possible?
Technique: Order does not matter (Alice-Bob-Carol = Carol-Bob-Alice). Use combination. ⁸C₃ = 8! ÷ (3! × 5!) = (8 × 7 × 6) ÷ (3 × 2 × 1) = 336 ÷ 6 = 56 committees.
Example 2 — Slot method: constrained arrangement
Question: A 4-digit PIN code is formed from the digits 1–9, with no repeated digits. How many valid PINs are possible?
Technique: Order matters (position is significant) and there is a restriction (no repeats). Use the slot method. Position 1: 9 choices. Position 2: 8 remaining. Position 3: 7. Position 4: 6. Total = 9 × 8 × 7 × 6 = 3,024 PINs.
Example 3 — Complement rule: 'at least one red'
Question: A bag contains 3 red balls and 7 blue balls. Two balls are drawn at random without replacement. What is the probability that at least one ball is red?
Technique: 'At least one' → use the complement. P(at least one red) = 1 − P(zero red) = 1 − P(both blue). P(both blue) = (7/10) × (6/9) = 42/90 = 7/15. P(at least one red) = 1 − 7/15 = 8/15. Why this is faster: the direct method requires computing P(exactly 1 red) + P(exactly 2 red) — two separate multi-step calculations. The complement reduces it to one subtraction.
Example 4 — Glue method: items that must stay together
Question: Alice, Bob, and 4 other colleagues are seated in a row of 6 chairs. Alice and Bob must sit next to each other. How many valid seating arrangements are there?
Technique: Glue Alice and Bob together as one unit. Arrange 5 units (glued pair + 4 others): 5! = 120 arrangements. Within the glued pair, Alice and Bob can swap: 2! = 2 ways. Total = 120 × 2 = 240 arrangements.
Example 5 — Combined counting and probability
Question: A committee of 3 is randomly selected from a group of 4 men and 3 women. What is the probability that the committee contains at least 1 woman?
Technique: 'At least 1' → complement. Total committees = ⁷C₃ = 35. Committees with zero women (all men) = ⁴C₃ = 4. P(at least 1 woman) = 1 − P(zero women) = 1 − 4/35 = 31/35. The direct method would require computing P(exactly 1 woman) + P(exactly 2 women) + P(exactly 3 women) — three separate calculations. The complement requires two.
Building Combinatorics Instinct Through Practice
Combinatorics and probability questions require pattern recognition more than formula memorisation. You do not need to derive the complement rule from first principles on exam day — you need to see 'at least one' in the question stem and immediately know to set up 1 minus P(none). That recognition builds through targeted repetition, not through understanding the theory more deeply.
The typical path to comfortable combinatorics performance on the GMAT is 30–40 questions practised with careful error review — not 100 questions done quickly. After each session, classify your errors: did you choose the wrong method (combination vs permutation), make a counting error within the right method, or miss a constraint? These are different weaknesses requiring different practice focus.
Targeted sub-topic drilling: OpenPrep's Combinatorics and Probability sub-topic drill allows you to practise the complement method, AND/OR rules, and the glue method in separate focused sets before mixing them. Drilling each method in isolation first builds the pattern recognition that makes mixed questions manageable — students who jump straight to mixed practice often never develop the automatic method-selection instinct.