How to Study for a Math Exam: Practice, Not Reading
Math exams test whether you can execute a method under time pressure, not whether you recognize it when you see it solved. That means the highest-value use of your study time is solving problems yourself, cold, and reviewing exactly where you went wrong โ not re-reading worked examples or lecture notes. Here's how to structure that.
Step 1: Separate "I understand this" from "I can execute this"
A common trap in math study is confusing recognition with ability. You read a worked example, nod along, and feel like you understand it โ but understanding someone else's solution is a different skill from producing your own solution from a blank page under time pressure. Test yourself honestly:
- Cover the worked example
- Try the problem (or a similar one) from scratch
- Only then compare to the solution
If you can't do step 2 without peeking, you don't actually have the skill yet, no matter how clear the example felt.
Step 2: Build a list of problem types, not topics
Rather than studying "derivatives" or "systems of equations" as broad topics, break each topic into the specific problem types your exam is likely to include (e.g., "related rates with a ladder sliding down a wall," "solving a 3-variable system by elimination"). For each type:
- Do at least one problem cold, without notes
- Note the specific method or trigger that tells you which approach to use
- Repeat with a slightly different version of the same problem type later in your study period
This structure matters because math exams often test whether you can recognize which method a problem calls for, and that recognition skill only builds through repeated exposure to varied versions of the same problem type.
Step 3: Interleave problem types instead of blocking them
Studying 20 related-rates problems in a row (blocked practice) feels productive, but it lets you solve each one using the method you already know you just used on the last problem: you're not practicing the harder skill of figuring out which method applies. Interleaving different problem types within the same study session: mixing related rates, optimization, and limits problems together, for example: is closer to what the actual exam will ask of you, since exam problems don't come pre-labeled by chapter.
A simple way to interleave: shuffle your practice problems from different chapters into one mixed set instead of working straight through a textbook chapter by chapter.
Step 4: Keep an error log, and make it specific
Every time you miss a problem, write down:
- The specific step where you went wrong (not "I got the wrong answer")
- Whether it was a conceptual error, an arithmetic slip, or a wrong choice of method
- A corrected version of that step
Over a few weeks, patterns show up: maybe you consistently mess up sign errors in integration by parts, or you default to the wrong method when a problem combines two topics. This error log becomes your highest-value review material before the exam, far more useful than redoing problems you already know you can solve.
Step 5: Sample two-week study schedule
| Day | Focus |
|---|---|
| 1-2 | List all problem types covered, do one cold attempt per type |
| 3-5 | Targeted practice on weakest problem types, build error log |
| 6-7 | Mixed/interleaved practice sets across all problem types |
| 8-9 | Timed practice: full problem sets under exam-length time limits |
| 10 | Review error log, redo only the problems you got wrong |
| 11-12 | Second round of mixed interleaved practice, new problem instances |
| 13 | One full timed practice exam under real conditions |
| 14 | Light review of error log only, rest before the exam |
Compress or extend this based on how much time you actually have before the exam.
Step 6: Practice under timed conditions at least once
Math exams usually have a real time constraint, and pacing is its own skill separate from knowing the material. Do at least one full practice set or past exam under strict timed conditions before the real exam, so you find out whether you're too slow on a particular problem type while you still have time to fix it. If you consistently run out of time, figure out which problem types are eating the most minutes and drill those specifically for speed.
Turn your own material into a study course
Upload a PDF, lecture recording or YouTube link. gigabrainz builds chapters, quizzes and spaced-repetition flashcards from it in a couple of minutes. Your first 3 courses are free, no card needed.
Try gigabrainz free โIf you have a textbook chapter, problem set, or lecture notes for your math course, gigabrainz can turn that material into short review chapters with quiz questions, which is a quick way to generate extra practice questions and a spaced-repetition review layer on top of material you already have.
Step 7: Don't skip the "why," even for procedural topics
It's tempting to memorize a procedure (steps 1-2-3 for this problem type) without understanding why it works, and this gets you through similar problems but falls apart on an exam problem that's phrased differently than anything you practiced. When you learn a new method, spend a few minutes explaining to yourself (or out loud) why each step works, not just what to do. the Feynman technique applies as well to math as it does to any other subject, and it's a fast way to find out if you actually understand a method or just memorized its shape.
What to do the day before the exam
Review your error log and do a short, mixed set of problems you've already seen before: not new, unfamiliar problem types. Keep it to active practice rather than passive re-reading of notes, and stop with enough time to get a full night's sleep. A tired brain makes more careless arithmetic errors than a rested one, and those errors cost just as many points as not knowing the method at all.
Dealing with word problems and multi-step questions
Word problems and multi-step questions are often where students lose the most points, not because they don't know the underlying math but because they struggle to translate the problem into the right setup. Practice this translation step explicitly:
- Read the problem once for general understanding, without trying to solve anything yet
- Identify what's being asked and what variables or quantities are given
- Write out the setup (the equation or system you need) before solving
- Only then work through the actual computation
If you consistently get the setup wrong but can compute correctly once it's set up, that's a specific, fixable issue: drill setup-only practice (writing the equation without solving it) on a batch of word problems until that step becomes reliable.
Using a calculator (or not) strategically
If your exam allows a calculator, practice with the same calculator you'll use on exam day well before the exam, not for the first time during a practice test. Know which functions you'll need and how to access them quickly, since fumbling with calculator buttons under time pressure wastes minutes you don't have. If your exam is calculator-free, practice all your problems without one during your prep too: switching to no-calculator mode for the first time on exam day will slow you down on basic arithmetic you've been letting the calculator handle.
Checking your work without redoing everything
Full-length practice exams are a good place to build a quick, reliable work-checking habit, since redoing an entire problem from scratch under time pressure usually isn't realistic. A faster check: plug your answer back into the original equation or problem, estimate whether the magnitude of your answer makes sense (a wrong sign or an answer ten times too large is often obvious this way), and glance back at your steps for an obvious copying error. Build this into your timed practice sessions so it becomes a habit rather than something you only think to do when you have time to spare.
Group study for math exams
Studying math with others can help, specifically when you take turns explaining problems to each other rather than just comparing answers. Explaining your reasoning out loud exposes gaps that silent solo practice won't catch, and hearing a classmate's different approach to the same problem type can add a second method to your toolkit. Keep group sessions focused on working problems together, not just checking who got which answer: a group session that turns into copying finished solutions skips the actual practice that builds the skill.
Frequently asked questions
Is it enough to just re-read my math notes before an exam?
No. Math is a performance skill, and re-reading notes doesn't build the ability to solve problems under time pressure. Most of your study time should go to solving problems, not reading solutions.
Should I study one topic at a time or mix topics when studying math?
Mix them. Interleaving different problem types forces you to first identify which method a problem calls for, which is often the actual skill being tested on an exam that covers multiple topics.
What's the best way to review mistakes on practice problems?
Keep an error log: write down the specific step where you went wrong and why, not just that you missed the problem. Revisit that log instead of re-doing every old problem from scratch.
How many practice problems should I do before a math exam?
Enough to see every problem type at least a few times and to do so under timed conditions at least once. A specific number matters less than whether you're doing problems cold, without looking at a solved example first.