← Back to all articles

Competition Roundups

Year in Review 2026: What the AMC 10 and AMC 12 A/B Papers Really Tested

A four-paper review of topic balance, difficulty structure and what genuinely changed from recent AMC exams.

18 Nov 202614 min read
Article cover image

Year in Review 2026: What the AMC 10 and AMC 12 A/B Papers Really Tested

Analysis reviewed: 18 November 2026. This article examines the released 2026 papers. MAA had not published a full score distribution when this review was completed, so difficulty judgements are based on paper structure rather than candidate averages.

The first reactions to a major maths competition are often too neat. One paper is called “easy”, another is labelled “number-theory heavy”, and a handful of memorable questions are quickly turned into a prediction about the future of AIME.

The four 2026 AMC 10 and AMC 12 papers do not support such a simple story.

After reviewing all 100 questions, the clearest distinction is not that A was easy and B was hard, or that one topic suddenly took over the series. The more useful conclusion is about where each paper placed its friction. AMC 10A generally offered clearer entry points and distributed its late difficulty across several subjects. AMC 10B demanded more switching between geometry, counting and probability, particularly in the middle and final thirds. AMC 12A leaned strongly on algebraic representation, while AMC 12B combined a geometry-rich paper with a more visible layer of number theory and advanced high-school content.

The 2026 papers did not rewrite the AMC syllabus. They made method selection, representation and movement between topics more important than any single fashionable technique.

Students can check format, eligibility and official links on the AMC 10 competition page and AMC 12 competition page. This article is a post-exam analysis rather than a registration guide.


Editorial overview

The 2026 cycle is best understood as four related but deliberately different papers. A and B are alternative administrations, not a main paper followed by a harder sequel. AMC 10 and AMC 12 also share a substantial core on each date, but the shared questions appear in different positions and are surrounded by different material. That makes the transition from AMC 10 to AMC 12 visible without reducing it to a fixed claim such as “AMC 12 Question 12 always equals AMC 10 Question 16”. The gap changes from question to question.

Across the four papers, there are fresh presentations, but most of the mathematics remains recognisably AMC: turn a verbal condition into structure, choose an efficient representation, and control the cases. The more meaningful changes concern where those demands appear and how quickly students must move between them.

How we classified the papers

Each question was assigned one primary area so that every paper totals 25 questions. This is a CompeteMap editorial classification, not an official MAA taxonomy. Mixed questions inevitably lose some detail when forced into one column: a geometric-probability problem may use both a diagram and an area calculation, while a divisibility-counting problem may reasonably be taught in either number theory or combinatorics.

We therefore use the table to compare balance, not to claim that subjects live in separate boxes.

PaperAlgebra and functionsSequences, process and statisticsGeometry and measurementNumber theoryCombinatorics and probability
AMC 10A75733
AMC 10B42838
AMC 12A92635
AMC 12B52954

Two cautions follow immediately. First, AMC 10B was not a “number theory explosion”: by this primary classification it contains the same number of number-theory questions as AMC 10A. Its real shift is toward counting, probability and geometry. Second, AMC 12B does increase the number-theory presence relative to 12A, but geometry remains the largest single category.


AMC 10A: accessible entry points, then broad structural reasoning

AMC 10A begins in a familiar way: ratios, percentages, elementary area, polygon angles and a sequence defined by averages. The opening does not give marks away automatically—Question 8, for example, turns rounded distance and time into a bounds problem—but most early questions signal a plausible first move.

The middle of the paper is where breadth becomes decisive. LCM, stopping probability, handshake counting, triangle inequalities, a rational-number tree, an urn problem, regular-hexagon geometry, recurrence, three-dimensional measurement and divisor enumeration arrive in quick succession. A student can know each topic individually and still lose time if they cannot decide which representation to use.

The final five are unusually balanced. They include:

  • a partition-and-scoring problem on a 3-by-3 grid;
  • an ellipse configuration that is ultimately controlled by geometry and distance;
  • counting possible pairs arising from multiples of 2 and 3 in consecutive integer lists;
  • a cubic polynomial evaluated at conjugate radical inputs;
  • a redistribution process in which an invariant is more useful than day-by-day simulation.

This supports the earlier CompeteMap judgement that the paper is more navigable than radically easier. Compared with the geometry-heavy upper end of the 2025 AMC 10A, the 2026 closing section offers more routes for students with different strengths. It remains selective because each route requires a method to be created rather than recalled.

For a closer A-paper comparison, see our 2026 AMC 10A versus 2025 analysis.

AMC 10B: the pressure came from switching modes

AMC 10B feels different very early. A scaling problem is followed by three-dimensional volume, a circle-region count and divisibility. By Question 10, students have already moved through exponents, similarity, spatial measurement, equations, geometry, number theory, sequences, constrained seating and coordinate curves.

The middle section is even more revealing. It includes a coin-path probability, random chords, exponential equations, probability over a parameter region, constrained colourings, a self-referential truth-count problem, prime valuations in a factorial, an infinite-turn game and modular counting. This is not a single-topic wall. It is a decision-cost wall: every few minutes the student must abandon one language and enter another.

The final questions reinforce that pattern. Tangent semicircles are followed by restricted lattice paths, expected value on cube faces, a three-term averaging recurrence and a rotating fan of congruent rectangles. The paper therefore rewards students who can recognise the type of object in front of them and select a compact model quickly.

Was 10B harder than 10A? Without an official score distribution, the responsible answer is narrower: 10B appears less forgiving under time pressure. Its combination of longer case structures and rapid topic changes can make the same level of mathematical knowledge produce a lower completed-question count. That is a structural observation, not a claim about the eventual mean score.


AMC 12A: algebra was the language connecting the paper

AMC 12A shares several early questions with AMC 10A, then introduces the wider AMC 12 toolkit: complex numbers, congruences, trigonometry and more advanced functional structure.

The algebra count is high, but “algebra” here does not mean repetitive equation solving. It includes a complex-number power, a nonlinear system, trigonometric products, a trigonometric equation, polynomial symmetry and an infinite product reduced using an algebraic relation. The common skill is not expansion. It is choosing a representation that exposes structure.

Question 19 is a useful example. The equation involving sin x, sin 2x and sin 3x rewards identities and factorisation. It is not the rational-function sign problem described in some early online notes. Question 20 then uses the symmetry of 3 plus or minus the square root of 5 inside a cubic polynomial. Together they show why a student who manipulates expressions mechanically may struggle even when the underlying syllabus is familiar.

The paper also keeps geometry substantial. The box-fitting problem, elliptical packing, mutually perpendicular sphere sections and the tangent-circle trapezoid all require the student to translate a spatial or geometric configuration into equations. Geometry has not disappeared; it has become closely interwoven with algebraic control.

AMC 12B: geometry led the count, advanced content raised the ceiling

AMC 12B does contain more number theory than 12A. Divisibility of a rational expression, a linear Diophantine equation, modular counting, a long remainder chain and the structure of exponent towers all appear. That is enough to make number theory more visible, especially at the top end, but not enough to describe the whole paper as number-theory dominated.

Geometry is more pervasive. The paper moves through scaled solids, unreachable volume in a rectangular box, a region count on a circle, an octahedron cross-section, circle geometry, repeated rotations, a ribbon around a cylinder, a fan of rectangles and an incenter configuration. Several of these problems are solved most cleanly with algebra or trigonometry, but the geometric model is still essential.

AMC 12B also makes fuller use of the AMC 12 syllabus. Logarithms, trigonometric products, complex or rotational reasoning, infinite series, graph loci and advanced exponent structure all appear. This is the clearest difference from AMC 10: the extra content is not confined to one late block. It changes the set of tools a student may need throughout the paper.

The closing run is especially demanding because the questions are unlike one another. A trigonometric locus is followed by a generating-function-style series, rotational geometry, modular residues, incenter geometry and exponent-tower counting. The challenge is not merely depth. It is recovering quickly after each change of mathematical language.


What genuinely changed from recent AMC papers?

1. The 2026 papers favour connected representation over isolated tricks

Recent AMC papers have already been moving toward problems that can be entered in more than one way. The 2026 set continues that direction. A diagram may become coordinates; a recurrence may become an invariant; a probability question may become an area; a polynomial condition may become symmetry around a shifted variable.

This is an evolution of emphasis, not a new syllabus. Students still need standard results, but they gain more from understanding when to convert a problem than from collecting increasingly specialised formulas.

2. Geometry did not collapse

The claim that geometry is “declining” does not survive all four papers. AMC 10A has a less geometry-concentrated final section than 2025, but AMC 10B and especially AMC 12B contain substantial geometry throughout. What has changed is that many geometry questions are algebraically mediated. A student may need coordinates, a discriminant, trigonometry or a distance equation before the diagram yields.

Calling those questions “not really geometry” hides the actual preparation demand. Modern AMC geometry often tests whether the student can move between synthetic and algebraic descriptions.

3. Process and recurrence questions remain strategically important

The raw number of recurrence questions is not enormous, but they appear in consequential positions. Average-defined sequences, redistribution processes, long-term behaviour and infinite sums all ask students to reason about what persists as a process evolves.

This continues a recent AMC pattern: making a short table is useful, but the decisive step is identifying the invariant, limit, weighting or reverse relation that makes a very large index manageable.

4. B papers were alternatives, not uniformly harder sequels

The B papers place difficulty differently. AMC 10B has a heavier combinatorics-and-probability profile and a more expensive middle section. AMC 12B contains more visible number theory and geometry than 12A. Neither fact proves that every B question is harder than its A counterpart.

Students who sat both versions should compare where they slowed down. A stronger score on A may reflect better alignment with algebra and recurrence; a stronger score on B may reflect comfort with geometry, counting and modular structure. The comparison is diagnostic only when tied to actual questions.

5. The qualification rule changed more sharply than the exam format

MAA states that the 2026 AMC 10 and AMC 12 were administered as in 2025. The major policy shift is the fixed AIME qualification line: eligible students qualify with at least 100 on AMC 10 or 85 on AMC 12.

That does not show that the papers were intentionally made easier. It changes strategy because the target is known before the exam. Students can make more deliberate decisions about leaving a question blank, protecting an achieved score and choosing which late questions deserve time.

The next stage has also changed. MAA says 2027 USAJMO and USAMO selection will be based primarily on AIME performance, with AMC scores no longer used in the qualification calculation. A high AMC score remains meaningful, but for olympiad-focused students the competitive separation now moves more clearly to AIME.


Claims from early commentary that should not be repeated

ClaimWhat the papers support
2026 AMC 10A Question 21 was a new bowl-shaped solidIncorrect. That description belongs to a 2022 problem; the 2026 question is the Triples Game on a 3-by-3 grid.
2026 AMC 12A Question 19 was a rational-function sign problemIncorrect. It is a trigonometric equation.
AMC 10B had a dramatic increase in number theoryNot supported by a one-primary-topic count; the more visible increase is in combinatorics, probability and geometry.
Geometry declined across the 2026 seriesNot supported across all four papers; geometry is especially prominent in 12B.
The 2026 papers predict that 2027 AIME will emphasise number theorySpeculation. AMC topic balance is not evidence for the future AIME distribution.
The fixed cutoff proves MAA made the papers easierUnsupported. Qualification policy and paper difficulty are separate questions.

This does not mean informal post-exam discussion is useless. It is often good at identifying where students lost time. It becomes unreliable when remembered question numbers are treated as a dataset or when an impression is promoted into a trend without checking all four papers.

How students should review the 2026 papers

Do not begin by counting how many questions came from each topic. Begin with the point at which your solution process failed.

Use four labels:

  1. Recognition failure — you knew the topic but did not identify the method.
  2. Representation failure — you understood the statement but chose an inefficient diagram, variable or sample space.
  3. Execution failure — the method was sound, but algebra, arithmetic or case control broke down.
  4. Time-allocation failure — the question was possible, but not worth the minutes spent during the exam.

For AMC 10 students, the most useful next step is usually to strengthen the middle-paper transition: move fluently among algebra, divisibility, probability and geometry before chasing every Question 25. For AMC 12 students, review whether complex numbers, trigonometry, logarithms and modular arithmetic are genuinely usable tools rather than familiar chapter names.

Students who met the fixed qualification score should not use this paper to predict the next AIME. Instead, they should move toward longer, answer-without-options problems where the route is not suggested by five choices. CompeteMap's five-year AMC final-review article can help identify recurring knowledge areas, while the 2027 AIME changes guide explains the new testing and progression rules.

CompeteMap judgement

The fairest summary of 2026 is not that AMC became dramatically harder, less geometric or dominated by number theory. It is that the four papers reward flexible mathematical translation more consistently than a narrow topic label suggests.

AMC 10A is the cleanest route through the cycle, but its middle and final sections still distinguish students who can organise unfamiliar information. AMC 10B places more strain on pacing and model selection. AMC 12A makes algebra the connecting language, while AMC 12B expands the technical ceiling through geometry, number theory and the full high-school toolkit.

That is a useful direction for preparation. Students do not need to guess next year's fashionable topic. They need to become better at recognising the structure underneath a new presentation.

Key Takeaways

  • AMC 10A was comparatively navigable; AMC 10B created more time pressure through rapid changes of method and a heavier counting/probability profile.
  • AMC 12A was algebraically connected, while AMC 12B was more geometry-rich and contained a stronger—but not dominant—number-theory layer.
  • Geometry did not disappear. Across the four papers, it increasingly works together with coordinates, algebra, probability and trigonometry.
  • Several widely circulated “2026” examples were actually older questions, so they should not be used as evidence of a new trend.
  • The fixed AIME thresholds of 100 for AMC 10 and 85 for AMC 12 change score strategy, but do not prove the exams were designed to be easier.
  • No responsible conclusion about average difficulty should be made until official score data are available.

Sources checked

Original CompeteMap editorial content. Brief quotations must clearly credit CompeteMap and link to this article. For complete reuse terms, see Copyright & Reuse.

EXPLORE NEXT

Not sure where to start?

Find the right competition
View all articles →

Loading comments…