Free beta · AMC 8
Top of the class is a fact about a classroom. Competition mathematics asks different questions, and a child at full coverage of the school curriculum can still have never met counting and probability, number theory, or the geometry AMC 8 quietly assumes.
A fifteen-question placement check is where it starts. Practice then shows you which, cluster by cluster, as counts you can check.
Free during the beta. A parent signs up with an email address; no card, no password.
| Skill cluster | Solved |
|---|---|
| Algebra & Equations | 9 of 9 |
| Number & Arithmetic | 8 of 9 |
| Ratio, Percent & Rate | 7 of 9 |
| Area | 6 of 9 |
| Sequences & Patterns | 5 of 9 |
| Data & Probability | 5 of 9 |
| Length, Angle & Similarity | 4 of 9 |
| Divisibility & Factors | 3 of 9 |
| Spatial & Transformation | 2 of 9 |
| Counting & Arrangements | 1 of 9 |
| Modular Arithmetic & Place Value | 0 of 9 |
Two clusters here are barely being solved at all. That is the finding worth having, and it is the finding a single overall score would have averaged away.
A single number cannot separate a child who has never seen a topic from one who has seen it and finds it hard. Those two children need opposite things, so we track them apart.
The eleven clusters we report against
Derived from every problem across nine years of AMC 8 papers, then grouped so that each row is something a parent can act on and large enough that one verdict about it means something.
The same rows carry upward. AMC 10 and AMC 12 bring material of their own, but they test many of these skills at greater difficulty rather than replacing them — so the map your child starts on is the one they keep, and it extends instead of resetting when they outgrow AMC 8.
Four parts, none of them complicated. The discipline is in what each one refuses to do.
Fifteen questions, about twenty-five minutes, sat once. Your child writes answers rather than picking from options, so a lucky guess cannot look like understanding. The questions climb until they are past your child, which is how it finds where practice should begin. Fifteen questions cannot reach every part of AMC 8, so the results say which clusters it touched and which it did not, rather than guessing about the rest.
You see the results as soon as the sitting ends, as counts per cluster, with the clusters it did not reach listed as not measured.
Practice comes in sittings of eight questions, pitched at the level your child currently solves, and that level moves with their first-try results. A wrong answer gets a second try. After that they see the working and why the option they chose does not work. If they cannot see how to begin, “I’m stuck” shows the first step and nothing more.
There is no chat tutor beside practice. That is a design decision, not a gap. See the FAQ.
Twenty-five questions in forty minutes, the shape of the real AMC 8 paper, with no marking until it is handed in. Afterwards your child goes back through it question by question, so a paper becomes something to learn from rather than just a number.
A practice paper is practice. We do not turn its result into a predicted contest score.
Your child chooses how many sessions a week to aim for and what they are working towards, and the plan shows how the week is going. You see the same plan from your account. Consistency matters more than volume, so the plan asks for a steady week rather than a heroic one.
A weekly email to you about your child’s practice is built but not yet switched on.
Every one of them is a number we could generate and a claim that would sell better than the truth. They are listed here because a product that tells you where the evidence stops is the only kind whose other numbers are worth anything.
Each is counted on its own terms, because each is different evidence. None of them is folded into a single grade.
After the placement check
What the fifteen questions reached, as counts per cluster, and the clusters they did not reach, listed as not measured rather than guessed at. Your child sees the same numbers, worded for them rather than as a mark.
As they practise
For each cluster: solved first time, solved on the second try, and not yet. Counts only, with no percentage, level or streak, and never added to the placement results.
After each paper
A timed practice paper, marked once it is handed in, with every question there to go back through. It is a record of that paper, not a predicted contest score.
The sentence we are actually aiming at
“On problems that need the cases split out, he solved 3 of 8. On problems of similar difficulty that don’t, 7 of 8.”
That is the strong child who cruises to question 15 and comes apart after 20 — the most useful thing anyone can tell you, and checkable rather than consoling. It is also why we report clusters instead of a single grade.
AMC 8 today. One level built properly rather than a shallow ladder across several. Champion works on the AMC pathway and only that pathway: AMC 8 is the entry point, and AMC 10 and then AMC 12 are where a strong student goes next. The design already assumes that move — there is one skill map across the pathway rather than one per contest, so counting with restrictions is the same skill at AMC 10 as at AMC 8, sat at a different difficulty. Nothing resets when the target changes. What we will not do is imply the higher bands are ready when they are not. AMC 8 is what exists.
We do not cover either one. AIME asks for sustained work on problems well above the band this product teaches, and preparing a child for it is not a near-term extension of what we do — we would rather say that than let you assume otherwise. MATHCOUNTS we simply do not cover either, and we are not going to put it behind a “coming soon”, because we have committed to no date and a date is what those words imply. If one of them is what you need this year, we are not the right product for you, and it is better that you know now than after joining a list.
No, and it would be a worse product if it were. Champion works on competition scope, which overlaps school maths far less than most parents expect — a child at the top of their year group can be at full coverage of the school curriculum and have never met counting and probability, number theory, or the geometry AMC 8 assumes. If what you want is help with tonight’s homework, a general tutor serves you better than we will.
No. If your child cannot see how to begin, “I’m stuck” shows the first step of the working and nothing more. A wrong answer gets a second try before any explanation appears, and the explanation then shows why the option they chose does not work. The habit being built is tolerance for being stuck, and that is exactly the habit an always-available answer destroys.
Two reasons, and both are deliberate. It would corrupt the measurement: everything we tell you rests on the record being a clean account of what your child did unaided, and a chat window beside the practice loop means we can no longer tell whose reasoning we recorded. And for a child who is already ahead, the specific weakness is usually avoiding productive struggle — an always-available tutor is the shortest route around the thing that would have made them better. Later we may allow discussion of a single problem after it is closed. Not during.
As observed counts, never a percentage. The placement check shows, for each cluster it reached, how many questions your child got right out of how many they were asked, and lists the clusters it did not reach as not measured. Practice is counted three ways per cluster: solved first time, solved on the second try, and not yet. The two are kept apart, because they are different evidence.
Their placement results, worded for them rather than as a mark, their practice sittings and practice papers with the review afterwards, their own counts of what they have solved, and the weekly plan they set. Nothing is turned into a score, a level or a streak. They practise on your device, or sign in on their own with a sign-in name and password you set. Either way it is your account, and you can see what they do.
Every problem in the library is human-reviewed after an independent AI pass that solves it cold without access to the answer key. If a problem still turns out to be wrong once it is in use, we retire it. Corrections apply going forward: we do not quietly rewrite your child’s past results, because a record that changes after the fact is harder to trust than the original imprecision.
As little as the product can run on: the name you choose for them (a nickname is fine), an optional school year, the region where they would sit the contest, and their answers and how long each took. We do not ask for their email, a school, a photograph or a date of birth. Your child is never publicly identified: there is no leaderboard and no ranking a stranger can look them up in. You hold the account, and your child’s own sign-in, if you set one up, cannot change it. Where multiple children are on one account they are shown on separate cards, never side by side, because a sibling comparison surface is not something we are willing to build. The details are in our privacy policy.
Nothing during the beta. If that changes, we will say so clearly before it does, and you will hear it from us rather than discover it at a checkout.
That is up to your child and you. Practice comes in sittings of eight questions, and your child sets a weekly plan for how many sessions to aim for. A practice paper takes forty minutes and is there whenever they want one. It is designed to sit alongside school rather than compete with it, and consistency matters considerably more than volume.
Champion is in beta on AMC 8, and the beta is free. If that changes, we will say so clearly before it does.
Sign up, add your child, and they can do the placement check today.
For parents of children in grade 8 or below who are preparing for AMC 8.