What This Episode Covers
Episode 38 of Real GMAT® Problems tackles one of the most important strategic decisions in GMAT® quant: when to use number theory reasoning versus when to just test numbers. The episode works through three Official Guide problems that involve number properties — consecutive integers, even/odd rules, and divisibility — and uses them to make a broader case for why a simple, organized testing approach almost always beats a theoretical approach for most test-takers.
The core argument is straightforward. Number theory reasoning is fast and seductive — we can skip the pencil work and reason through properties in our head. But for the vast majority of us, it's also dangerous. The more we try to calculate and reason mentally, the more questions we get wrong that we actually know how to do. The analogy we use: imagine investing based on pure intuition without ever checking a spreadsheet. Fast? Absolutely. Reliable? Almost never.
The alternative is to develop a clear, organized approach to testing numbers. It requires almost no memorization. It takes almost no time to learn. And it takes just a little bit of conditioning to do well. For most of us, that's a much better trade than spending months developing deep number theory intuition that we'll use on the GMAT® and then rarely again.
Problems Covered
Problem 1 — Warm-Up: Consecutive Integer Properties If a, b, and c are consecutive positive integers and a < b < c, which of the following must be true? Three Roman numerals: (I) c − a = 2, (II) a × b × c is even, (III) (a + b + c) / 3 is an integer. We walk through writing all the constraints, testing small numbers (1, 2, 3), and seeing that all three statements hold. The testing approach also creates natural opportunities for theoretical insight — after a few examples, it starts to make sense WHY c − a always equals 2 and WHY three consecutive integers always include an even number.
Problem 2 — Mid-Level: Sum of N Consecutive Integers Equals Zero If the sum of N consecutive integers is zero, which must be true? (I) N is even, (II) N is odd, (III) the average of the N integers is zero. This problem has a 27% miss rate — almost double the warm-up. We use the same testing framework: start with small examples (−1, 0, 1 where N = 3), immediately eliminate Roman numeral I, then test a few more sets to build confidence that II and III both always hold. The key insight: the average equals the sum divided by N, and the sum is always 0, so the average must be 0.
Problem 3 — Harder: Positive Even Divisors of 4p If N = 4p, where p is a prime number greater than 2, how many different positive even divisors does N have, including N? This one has a 37% miss rate — the highest of the three — yet it's actually solvable faster than the others with a testing approach. We plug in p = 3 (small, easy, fits the constraints), get N = 12, list the factor pairs (1×12, 2×6, 3×4), cross off the odd factors (1 and 3), and count the remaining even divisors: 2, 4, 6, and 12. Answer: 4. Because this is not a Roman numeral question, one valid test case is enough.
Key Takeaways
- Testing numbers requires almost zero memorization and almost zero ramp-up time. For most of us, that's a far better trade than spending months building number theory intuition we'll only use on the GMAT®.
- The things that make number theory reasoning fast are the same things that make it dangerous. It skips writing things down and skips computation — which means more mental juggling, which means more careless errors.
- Write your constraints first, every time. "Consecutive," "positive," "integers," "a < b < c" — missing even one constraint can send you down a completely wrong path. These problems are designed to overload working memory.
- On "must be true" Roman numeral questions, testing numbers may not give 100% certainty. That's the one trade-off. But on non-Roman-numeral problems, a single valid test case is enough to confirm the answer.
- Testing numbers often leads to theoretical insight anyway. After plugging in a few examples, patterns become visible that pure abstraction might take much longer to reveal. Lead with the test, and let the theory follow.
- Problem difficulty does not always track with mathematical complexity. Problem 3 has the highest miss rate but is arguably the simplest to solve with a testing approach — the difficulty comes from constraint tracking, not from harder math.
Related Reading
- Real GMAT® Problems - Ep. 39 - Translating Percents — builds on the testing numbers framework with percent-specific strategies
- Real GMAT® Problems - Ep. 40 - Square Roots — more Official Guide problems with applied number testing