Stop Solving for x When the Question Doesn't ASK for x
Say you're fifteen questions into the Quant section. The question hands you two equations with x and y in them. You do exactly what your algebra teacher trained you to do. You solve for y in terms of x. You plug it in. You clear a fraction. You catch a sign error and fix it. Two and a half minutes later you have x = 1, and you feel great.
Then you read the last line again. It asked for x + y.
Now you need y too. That's another 40 seconds. And here's the part that should hurt: the question was BUILT so you could get x + y in about 15 seconds without ever finding x at all.
I watch Test Takers do this every single week. Smart people. Strong math people. In fact, the stronger the math background, the worse the habit tends to be. That's because this isn't a math problem. It's a reading problem wearing a math costume.
The GMAT doesn't care about x
Right now the Quant section gives you 31 questions in 62 minutes, split between Problem Solving and Data Sufficiency (you can check the current format on GMAC's page on the exam sections and timing). That's an average of 2 minutes per question. The test is adaptive, so you can't go back and change an answer later. Every minute you spend on a letter nobody asked for comes straight out of a question you could have gotten right.
The question writers know how you were taught. You were taught that two unknowns need two equations, and that 'solving' means finding every letter. So they write questions where the thing they want is a COMBINATION: x + y, x − y, 2a + b, xy, x/y. The combination is often easy to reach directly. The single letters are the long way around.
Stop reading for a second and think about the last algebra question you did in practice. Did you find every variable? Did the question actually ASK for every variable? I'll bet that, at least some of the time, it didn't.
See it on a real question
First up is Problem Solving: a regular multiple choice question with five answers. Do it on paper before you open the solution.
Did you add the equations, or did you substitute? Be honest. If you substituted, you didn't do anything 'wrong'. You just did something SLOW, and slow costs points on a timed test.
Why the long way feels 'safe'
When I point this out, people push back. "Solving for both letters is more thorough." No. It's more steps. More steps means more arithmetic, and more arithmetic means more chances for a silly mistake. The careful move is the one with the fewest places to slip.
There's also some pride mixed in. Finding every variable feels like 'real' math. Building the target directly feels like a trick. But the GMAT isn't a math test. It's a critical thinking test that uses math as the measuring stick. Spotting the shortcut IS the skill being measured.
Now the Data Sufficiency version
Data Sufficiency (DS for short) gives you a question and two facts, called statements. You don't find the answer. You decide whether the facts are ENOUGH to find exactly one answer. This is where the 'solve for every letter' habit does real damage.
5 mistakes I see every week (and what to do INSTEAD)
Reading the question's last line once, quickly, then diving in. The thing being asked is often in the final few words, and that's the part people skim.
INSTEAD: Before any math, write the target on your pad and box it. Every time. No exceptions.
Starting with substitution on autopilot. It's the method you learned first, so it's the one your hand reaches for.
INSTEAD: Spend five seconds LOOKING. Are the coefficients swapped? Do they share a factor? Does the target show up if you add, subtract or factor? Then pick a method.
Clicking the last number you wrote down. If you solved for x, x is staring at you, and x is almost always one of the wrong answers.
INSTEAD: Before you click, look at your box. Does the number you're about to pick match what's in it?
Counting equations and letters in DS. "Two unknowns, one equation, not enough" is a reflex, not an analysis.
INSTEAD: Ask one question of each statement: can I shape this into the target? If you're unsure, TEST VALUES and see whether the target changes.
Doing the work in your head because it 'looks simple'. That's how 23 + 17 becomes 30 and a gettable question becomes a wrong one.
INSTEAD: Write the stacked equations down, line them up and do the arithmetic on paper. It takes seconds and it saves points.
A quick list of signals
Train your eye to spot these. When you see one, the question probably wants you to build the combination directly.
| What you notice | What to try first |
|---|---|
| Two equations with swapped coefficients | Add them, or subtract them |
| Every term in an equation shares a factor | Factor it out and look for the target |
| The question asks for a ratio like x/y | Divide through by y instead of finding each letter |
| The question asks for x + y and gives (x + y) squared | Work with the whole group, not the pieces |
Your challenge for today
Pull 10 algebra questions you've already done, ideally from GMAC's official practice materials. For each one, do three things on paper:
- Write the target and box it.
- Ask whether you could have built that target WITHOUT finding every letter.
- If yes, solve it that way and time yourself. Then compare that time to what you spent the first time.
Count how many of the 10 asked for a combination. Count how many of those you solved the long way. I'll bet that second number is bigger than you'd like. Good. Now you know exactly what to train, and this is a habit you can absolutely fix before Test Day.