The Reframe — A Different Kind of Math
A math olympiad is not a faster version of the school test. The problems do not test the syllabus; they test whether your child can think with the syllabus. A Grade 6 olympiad question rarely needs anything beyond Grade 6 content — but it asks the student to combine three Grade 6 ideas in a way the school book never showed.
That distinction changes everything about preparation. Drilling chapter problems does not help. Teaching your child to sit with a single problem for twenty minutes — that is what helps.
If your child is curious, asks "why," and does not give up easily, olympiads are worth a look. If your child treats math as a sprint to the answer, the olympiad will frustrate them — and that is fixable with the right preparation, not the wrong one.
What Olympiads Actually Test
Most school tests measure recall plus speed. Olympiads measure patience plus creativity. The four skills that show up on every major contest (AMC 8, MOEMS, Math Kangaroo, AMC 10/12, AIME, USAMO, IMO):
Pattern recognition. Seeing that a problem about handshakes is really a problem about triangles.
Working backwards. Starting from the answer choices, or from "what would have to be true."
Case analysis. Splitting a problem into 2–4 possibilities and checking each.
Estimation and sanity checks. Knowing whether 17 or 1700 is a plausible answer before doing the arithmetic.
None of those four are taught in a typical school week. They are taught by spending time with hard problems — under guidance, not alone.
The Olympiad Levels — School to IMO
Parents often hear "math olympiad" and imagine one event. In reality, math olympiads form a tiered pipeline. Most children stay in the bottom three tiers and that is perfectly fine. Only a handful per country reach the top.
Level | Examples | Who takes it | Format | What it leads to |
|---|---|---|---|---|
School / Foundation | MOEMS, Noetic Math, school-level SOF IMO | Grades 1–8 at the child's own school | 5 problems × 30 min (MOEMS); 24 MCQ × 75 min (Kangaroo) | Exposure; a positive first contest |
Regional / National qualifier | AMC 8, Math Kangaroo, SOF IMO Level 1 | Grades 3–8 (Kangaroo 1–12) | 25 MCQ × 40 min (AMC 8); MCQ tests | Distinction certificates; entry to harder tracks |
National invitational | AMC 10/12, AIME, MATHCOUNTS State + National | Grades 6–12 (MATHCOUNTS 6–8; AMC 10 ≤10; AMC 12 ≤12) | 25 MCQ × 75 min (AMC); 15 integer answers × 3 hr (AIME) | Qualification for the national olympiad |
National olympiad | USAMO / USAJMO (US); INMO (India); BMO (UK) | Top ~250–500 AIME scorers per country | 6 proof problems across 2 × 4.5-hour days | Selection for the IMO training camp |
International | IMO (International Mathematical Olympiad) | Six students per country, ~110 countries | 6 problems × 2 × 4.5-hour days; medals at ~top half | The most prestigious pre-university math competition globally |
Two practical notes for parents reading this for the first time. Out of roughly 300,000 AMC test-takers each year in the US, only six students reach the IMO. The point of the pipeline is not the top — it is that every tier teaches something. Most children stop at AMC 8 or AMC 10, and that is a complete olympiad experience.
The other note: school-level and foundation contests (MOEMS, Math Kangaroo) are not "fake" olympiads. They use the same four skills as the IMO, scaled to grade-level mathematics. A Grade 4 Math Kangaroo question is genuine olympiad thinking.
Signs Your Child Is Olympiad-Ready
You will recognise olympiad readiness from these specific behaviours, not from school grades:
They argue with the wording of a homework problem — "the question is ambiguous" — because they have read it carefully.
They re-do a problem they got right, "to see if there is another way."
They notice patterns in things that are not math — license plates, train numbers, the layout of a parking lot.
They are bored by routine arithmetic but light up when you show them a puzzle.
They lose at chess or strategy games and want to play again immediately.
A child with three of those signals will enjoy an olympiad. A child with none of them will not — and pushing them into one is the fastest way to make math feel like a punishment.
Age-by-Age — What to Do at Each Stage
Preparation is shaped by where your child is in school. The same parent move that works for Grade 4 backfires in Grade 10.
Grade 3–4 (age 8–9): Exposure, not preparation
Math Kangaroo Level 3–4 at home. Twenty-four questions across 75 minutes, casual. Read riddle books at bedtime — The Number Devil by Hans Magnus Enzensberger is a good one. The goal is to find out whether your child is curious about puzzles, not to win anything.
Grade 5–6 (age 10–11): The first real attempt
MOEMS through school if available. Math Kangaroo annually. Try the AMC 8 in Grade 6 only if your child has done two contests already and asked to try a harder one. One 20-minute problem session per week at home, on a published past paper.
Grade 7–8 (age 12–13): Serious AMC 8 preparation
This is the prime AMC 8 window. Three sessions a week of 20–25 minutes each. Past AMC 8 papers — the MAA archive is free. Aim for a 12+ at first attempt; 15+ on the second sitting. MATHCOUNTS is the strong supplementary contest at this level — chapter round in February, state in March.
Grade 9–10 (age 14–15): AMC 10 territory
The contest is 25 questions in 75 minutes. Problems are harder; the four skills now matter more than school-syllabus fluency. Aim for an AMC 10 score that qualifies for AIME (typically top ~2.5% — around 100/150 in recent years). At this point, structured coaching (Art of Problem Solving's online courses, or a Bhanzu-style track) starts paying back the cost.
Grade 11–12 (age 16–18): AMC 12, AIME, USAMO path
This is the year the pipeline narrows. AMC 12 → AIME → USAJMO/USAMO → MOP → IMO. The selection index used by MAA is AMC 12 Score + 20 × AIME Score (USAMO) or AMC 10 Score + 20 × AIME Score (USAJMO). At this level, preparation is full-time problem solving — Art of Problem Solving Volume II, Putnam past papers, and proof writing become necessary.
A pattern worth knowing: students who reach USAMO almost always started before Grade 7. Late-bloomer USAMO qualifiers exist but are rare. If your child is in Grade 10 and has not done any contest, the realistic target is AMC 10 Distinction, not IMO.
A 12-Month Preparation Arc
If your child is ready and the contest is a year away, here is the shape we have seen work. None of it requires a coaching center.
Quick — Months 1 to 3 (Build the Habit)
Twenty minutes a day, three days a week, on a single problem. Not a set of ten. One. If your child solves it in five minutes, the problem was too easy; pick a harder one the next session.
A good source is the AMC 8 problem archive (free, with solutions) or Art of Problem Solving's Introduction series. The Math Olympiad Contest Problems (MOEMS) book set is excellent for Grades 4 to 6.
The goal of the first three months is stamina — the ability to sit with a problem for twenty minutes without checking the answer. Most children cannot do this for the first two weeks. By Week 8 they can.
Standard — Months 4 to 8 (Add Technique)
Now introduce the four olympiad skills explicitly. One week on case analysis (problems where the child has to split into 2–4 cases). One week on working backwards (problems where the answer is given and the question is "find the path"). One week on pattern recognition. One week on estimation.
Cycle through. Each technique gets ten to fifteen problems over the four weeks, with the child writing out the solution — not just announcing the answer.
This is also the right time to enter a small practice contest if available — MOEMS divisions, Math Kangaroo, or a regional MATHCOUNTS chapter round. The point is exposure to the format, not the result.
Stretch — Months 9 to 12 (Full Contest Simulation)
In the final three months, simulate the contest under timed conditions. Print last year's paper (every major contest publishes past papers free), sit at the kitchen table, set the timer. Your child writes the full paper alone.
After the timer ends, do not mark the paper that night. Mark it together the next morning, slowly, focusing on the problems they did not solve. Each one becomes a study session.
Two full simulations a month is enough. More than that and the child burns out before contest day.
Three Worked Examples Across the Skill Levels
Olympiad prep is concrete only when you have seen actual problems. Three examples below — Quick, Standard, Stretch — show the shape of the questions and how the four skills land.
Example 1 — Quick (MOEMS / early AMC 8 difficulty)
Problem. A bakery sells donuts in boxes of 6 and 8. What is the largest number of donuts that cannot be bought using any combination of these boxes?
Solution. List the achievable totals starting from 6: 6, 8, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, …
What's missing? 7, 9, 10, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33. From 34 onwards every number is achievable (since 6, 8, 14, 16 cover all parity-and-modular cases).
Final answer: 33.
This is case analysis with a small sanity check. The technique is just patient listing. A Grade 5 student can solve it in eight minutes with no prior contest training.
Example 2 — Standard (AMC 8 mid-difficulty) — A Common Slip Worth Walking Through
Problem. In a class of 30 students, every student plays either chess or soccer (or both). 20 play chess. 16 play soccer. How many play both?
The wrong path most students take first. 20 chess + 16 soccer = 36 total. The class has 30 students. So 6 students must play "extra." The answer is 6.
That feels right. It is not — and it took us three sessions to get our Grade 7 cohort to stop doing this without prompting.
Where the slip is. "36 total" double-counts every student who plays both sports. The 6 is not the number of overlap students directly — it is the size of the double-count, which equals the overlap. So actually, in this problem, the wrong-path-first answer coincidentally lands on the right number. The reasoning was wrong; the answer was right.
The clean way (set-overlap formula). |Chess ∪ Soccer| = |Chess| + |Soccer| − |Chess ∩ Soccer| 30 = 20 + 16 − |Both| |Both| = 36 − 30 = 6.
Final answer: 6.
The reason this example matters: olympiads punish lucky right answers on later problems. The student who learned the inclusion-exclusion principle properly here can extend it to three sets (Problem 22 of AMC 8) where the lucky-arithmetic move fails.
Example 3 — Stretch (AMC 8 hard / AMC 10 entry)
Problem. How many positive integers less than 1000 have digit sum equal to 9? (e.g., 27 has digit sum 9; 405 has digit sum 9.)
Solution. Reframe: count the non-negative integer solutions to $a + b + c = 9$ where $a, b, c$ are digits (each 0–9) and not all zero, with leading-digit constraint.
For three-digit-or-fewer numbers, the problem becomes "stars and bars with bounded variables." Without upper bounds, $\binom{9+2}{2} = 55$. But each digit must be ≤ 9. The only violation would be a digit ≥ 10 — but since the sum is exactly 9, no digit can exceed 9. So no subtraction needed.
Subtract the all-zero case: 55 − 1 = 54. But wait — "less than 1000" includes 0 only if we count it; we are counting positive integers. The all-zero case (number 0) is already excluded. The answer is 54.
Final answer: 54.
This is pattern recognition (recognising stars-and-bars) layered on case analysis (checking the bound). It is the prototype of an AMC 10 problem and the kind of question a child cannot solve without the four skills already trained.
Where Most Parents Lose the Thread
The instinct is to push more problems — a workbook a week, double sessions on weekends, a coaching center on top of school. It almost always backfires.
Olympiad math is the opposite of cramming. The child who does one hard problem a day for a year beats the child who does ten problems a day for two months. The hard problem teaches stamina; the ten problems teach speed, which the olympiad does not test the way school tests do.
The other common move is solving the problem for your child when they are stuck. Resist. A child who watches you solve a problem learns to wait for someone to solve it for them. A child who sits with the problem for thirty minutes and then asks one clarifying question — that child is doing olympiad math.
Tripping Points to Avoid
Four habits derail olympiad prep more than weak math:
Treating practice as a race. Speed wins school tests; depth wins olympiads. A child who races through ten problems learns less than one who sits with two.
Skipping the write-up. Solving in the head and announcing the answer is half the work. USAMO and IMO require written proofs. Practise writing — even on AMC 8 problems where the official answer is just a letter, the student should produce a paragraph of reasoning.
Studying only what you are good at. If your child loves geometry, they will avoid number theory. The actual contest covers all four areas — algebra, geometry, number theory, combinatorics. Force a rotation.
Coaching center as a substitute for thinking. A center can teach techniques. It cannot teach stamina. The twenty-minute solo sessions at home are non-negotiable.
A pattern we have observed in Bhanzu's Saturday Grade 7 competition cohort over the past three years: the students who place in the top decile of regional contests almost always have one habit in common — they keep a notebook of problems they could not solve, and revisit them three months later. The students who skip that habit plateau by Month 6.
When to Bring in Outside Help
A coach or structured program is worth the call when:
Your child has consistently solved 80% of the problems at the level below the target contest, but the target contest itself is sitting at 30–40%.
Your child has plateaued for two months — same accuracy, no improvement, frustration rising.
You have run out of problems you can grade. Above AMC 8 level, most parents cannot mark a written solution accurately.
Below those thresholds, home preparation with a good problem source is enough. Above them, a tutor or coaching program adds genuine value.
Key Takeaways
How to prepare for a math olympiad is mostly a question of stamina — twenty minutes a day on a single hard problem builds more than ten problems an hour.
The four core skills (pattern recognition, working backwards, case analysis, estimation) need explicit weekly practice — none of them are taught at school.
The olympiad pipeline runs from school-level (MOEMS, Math Kangaroo) through national (AMC 8/10/12, AIME) to international (USAMO, IMO) — most children stop midway and that is fine.
Coaching adds value above AMC 10 distinction; below that, home preparation with past papers is enough.
The first contest is exposure, not a result. Children who reframe it that way continue. Those who treat it as a verdict do not.
Your Next Move This Week
Open the AMC 8 archive at the MAA. Pick one problem from three years ago. Hand it to your child, set a timer for twenty minutes, leave the room. When the timer ends, sit with them and ask one question: what did you try? That conversation is the beginning of olympiad preparation.
Was this article helpful?
Your feedback helps us write better content