If you want to know how to study Matrices and Determinants for JEE — Main or Advanced — this guide is built entirely from real previous-year papers. We pulled every question this chapter produced in the most recent cycles — 107 questions across JEE Main 2024, 2025 and 2026, plus 19 from JEE Advanced 2020–2026 — and sorted every one into a sub-topic. What follows is the actual pattern, not textbook guesswork.
Matrices and Determinants carries a 6.95% weightage in JEE Main 2026, down 2.93% from 2025 — essentially flat, making it one of the most stable, predictable chapters in the whole Algebra syllabus. On JEE Advanced it sits at 5.88%, down 5.92%. Neither exam shows the dramatic swings we've seen in other chapters this series — instead, this is a chapter that quietly, reliably produces marks every single year, which makes it a low-risk, high-return investment of revision time.
| Year | Questions (MCQ + Numerical) | Note |
|---|---|---|
| JEE Main 2024 | 43 | Highest of the three years |
| JEE Main 2025 | 46 | Slightly higher again |
| JEE Main 2026 | 34 | 6.95% weightage, down 2.93% year-on-year |
ExamSIDE's archive shows 375 total JEE Main questions from this chapter since 2002 — 187 papers deep, split 296 single-correct MCQ and 79 Numerical. That 375 total is among the largest of any chapter in our series so far, and even at its "lowest" recent year (2026, 34 questions), it's still a heavyweight contributor to your Algebra score. Numericals hold at roughly 27% of the chapter's questions across our window — a healthy but not dominant share, meaning you need both calculation fluency (numericals) and conceptual recognition (MCQ) in equal measure.
| Year | MCQ (Single) | MCQ (Multi) | Numerical | Total |
|---|---|---|---|---|
| 2020 | 0 | 1 | 1 | 2 |
| 2021 | 0 | 2 | 2 | 4 |
| 2022 | 1 | 0 | 1 | 2 |
| 2023 | 1 | 1 | 1 | 3 |
| 2024 | 1 | 1 | 1 | 3 |
| 2025 | 1 | 1 | 0 | 2 |
| 2026 | 1 | 1 | 0 | 2 |
ExamSIDE's full archive (1978–2026) shows 62 total JEE Advanced questions across 32 papers. Unlike some of the other chapters in this series, this one shows real consistency — at least 2 questions every single year since 2020, without a single skipped year. That reliability is itself useful information: you can count on Matrices and Determinants showing up on Advanced, which means the "should I even prepare this for Advanced" question that came up for Sequences and Series simply doesn't apply here.
| Sub-topic | JEE Main (2024–26) | JEE Advanced (2020–26) | Combined |
|---|---|---|---|
| Determinants, Adjoint & Inverse | 35 | 7 | 42 |
| Systems of Linear Equations | 33 | 4 | 37 |
| Matrix Types & Operations | 5 | 7 | 12 |
| Matrix Powers & Characteristic Equations | 5 | 1 | 6 |
28 of the 107 JEE Main questions (26%) couldn't be confidently sub-topic-classified — matrix questions frequently render their actual matrix entries as image/notation blocks in the source rather than plain text, and where the surrounding words alone didn't give us enough to classify confidently, we excluded rather than guessed. We're stating this plainly rather than smoothing the numbers.
The story here is a genuine two-way tie at the top: Determinants, Adjoint & Inverse (42) and Systems of Linear Equations (37) together account for 79 of the 97 categorised combined questions — 81%. If you master these two sub-topics and nothing else, you've covered the overwhelming majority of this chapter's marks. Everything else — matrix types like symmetric/skew-symmetric, and matrix-power/characteristic-equation problems — is real but secondary.
Our Maths HOD, AKS Yadav Sir, structures every 1-on-1 Matrices session around this exact priority order, so you spend your hours where the marks actually are.
Free 30-min mentorship · 100% refund guarantee · 7-day WhatsApp access
"This is one of the most 'trainable' chapters in the whole syllabus, because it isn't testing creativity — it's testing whether you can execute a known procedure (find the adjoint, check for infinitely many solutions, multiply matrices) without an arithmetic slip. Students lose marks here from rushing, not from not knowing the theory. I spend more time on slow, careful practice in this chapter than almost any other." — AKS Yadav Sir, HOD Mathematics, JEE Prep Master
Questions like the 2026 8th April Evening Shift ("det(adj(3adj(A²·adj(2A))))") stack three or four adjoint/determinant operations in a single expression. You can't shortcut these — you need the determinant-of-adjoint scaling rules (|adj(A)| = |A|ⁿ⁻¹ for an n×n matrix, and how scalar multiples inside adj() scale the result) applied correctly at each nested layer.
Across 2024–2026, a large share of Numerical questions give a 3-variable linear system with unknown parameters and ask for a value consistent with infinitely many solutions — this format alone (not unique-solution or no-solution variants) is disproportionately common as a numerical-answer format, likely because "infinitely many solutions" reduces cleanly to a determinant-equals-zero condition that produces one clean numeric answer.
Several 2024–2026 questions define a matrix A only through statements like "A[1,0,1]ᵀ=[3,4,4]ᵀ" rather than giving you A directly — you must reconstruct A's action from these vector equations before you can answer anything about its determinant or powers. This format tests linear-transformation understanding, not just mechanical determinant calculation.
The 2024 4th April Morning Shift question turns a matrix-power equation into a hyperbola eccentricity problem. This cross-topic linking means "just study matrices in isolation" leaves gaps — you need to recognise when a matrix condition is secretly defining a curve or another algebraic object.
Questions asking "how many n×n matrices with entries from set S satisfy condition X" combine matrix property-checking with combinatorial counting — a distinct skill from pure determinant/system-solving, and one that's easy to under-practice because it doesn't fit neatly into either "algebra" or "combinatorics" study blocks.
The core relationship is A·adj(A) = |A|·I, from which the inverse formula A⁻¹ = adj(A)/|A| follows directly. Know the scaling rules cold: for an n×n matrix, |kA| = kⁿ|A|, |adj(A)| = |A|ⁿ⁻¹, and |A⁻¹| = 1/|A|. These rules are what let you collapse a nested expression like det(adj(3adj(2A))) into a single number without ever writing out the actual matrix. Also know that adj(adj(A)) = |A|ⁿ⁻²·A for an n×n matrix — this specific identity shows up repeatedly in the "nested adjoint" question style described above.
Worked example: If |A|=6 for a 3×3 matrix, find |adj(3·adj(A²·adj(2A)))|. Work inside-out: |2A| = 2³|A| = 8·6 = 48, so |adj(2A)| = |2A|² = 48² = 2304 (using |adj(M)|=|M|ⁿ⁻¹ with n=3, so exponent is 2). Then |A²·adj(2A)| = |A²|·|adj(2A)| = |A|²·2304 = 36·2304. Then |adj(that)| = (that value)². Finally |3·(that adjoint)| = 3³ × (that value)². Track each step's exponent separately and this always resolves cleanly — the entire question is exponent bookkeeping, not matrix computation.
For a 3-variable system, form the coefficient determinant Δ. If Δ≠0, the system has a unique solution (Cramer's rule applies directly). If Δ=0, you must check the individual numerator determinants (Δₓ, Δᵧ, Δᵤ) — if all three are also zero, the system has infinitely many solutions; if at least one is non-zero, it has no solution. The overwhelming majority of "infinitely many solutions" questions reduce to setting the coefficient determinant to zero and solving for the unknown parameter — that single step is usually 80% of the work.
Worked example: For what value of λ+μ does x+y+z=6, x+2y+5z=10, 2x+3y+λz=μ have infinitely many solutions? First set the coefficient determinant to zero: |1 1 1; 1 2 5; 2 3 λ| = 0, solve for λ. Then, separately, substitute that λ back into the system and require the augmented determinant (replacing one column with the constants) to also vanish — this gives the condition on μ. Only after finding both λ and μ do you compute λ+μ. Skipping the second check (assuming any μ works once Δ=0) is the single most common error in this sub-topic.
This same Δ=0-then-check-numerators logic extends naturally to the "non-trivial solution" homogeneous-system questions that appear regularly in the Advanced List-matching format — a homogeneous system (all constants equal to zero) always has the trivial all-zero solution, so the interesting question is always whether a non-trivial one also exists, which happens exactly when Δ=0. There's no separate numerator check needed for homogeneous systems specifically because the trivial solution already satisfies every numerator condition automatically — one fewer step than the non-homogeneous case above, and a useful shortcut to recognise immediately from the question's phrasing.
Know your definitions cold: symmetric (A=Aᵀ), skew-symmetric (A=-Aᵀ, with all diagonal entries forced to zero), orthogonal (AAᵀ=I), and idempotent (A²=A). A useful standing fact: any square matrix can be written as the sum of a symmetric and a skew-symmetric matrix — A = ½(A+Aᵀ) + ½(A-Aᵀ) — and this decomposition itself is occasionally the entire question. Trace (the sum of diagonal entries) is additive and scales linearly, which makes trace-based questions usually faster to solve than full determinant-based ones when both are options.
Worked example: Is a matrix M with M(AAᵀ)=I an orthogonal matrix, and what does that tell you about |M|? If AAᵀ=I, then A is orthogonal by definition. Taking the determinant of both sides of AAᵀ=I gives |A|·|Aᵀ|=|I|=1. Since |Aᵀ|=|A| for any square matrix, this becomes |A|²=1, so |A|=±1. This is a fast, reusable fact: any orthogonal matrix has determinant exactly +1 or -1, never anything else — and recognising this immediately settles several JEE Advanced questions that otherwise look like they need a full explicit determinant calculation.
Every square matrix satisfies its own characteristic equation (Cayley-Hamilton theorem) — for a 2×2 matrix A with trace t and determinant d, A² - tA + dI = O. This single identity lets you reduce any high power of A (like A¹³ or A²⁰) down to a linear combination of A and I, avoiding repeated matrix multiplication entirely. When a question gives you a matrix equation like A³=4A²-A-21I and asks you to find unknown entries, recognise this as Cayley-Hamilton in disguise — match it against the actual characteristic equation of the given matrix to solve for the unknowns directly.
Worked example: If A=[[2,-1],[1,1]] and the sum of the diagonal elements of A¹³ is 3ⁿ, find n. First find A's characteristic equation: trace=3, det=2(1)-(-1)(1)=3, so A²=3A-3I by Cayley-Hamilton. This means every power of A can be written as a linear combination pA+qI. Set up a recurrence: if Aᵏ = pₖA + qₖI, then Aᵏ⁺¹ = pₖA² + qₖA = pₖ(3A-3I) + qₖA = (3pₖ+qₖ)A - 3pₖI, giving you pₖ₊₁=3pₖ+qₖ and qₖ₊₁=-3pₖ. Iterating this recurrence from A¹=1·A+0·I up to k=13 (by hand or by spotting the pattern) gets you A¹³'s coefficients directly — and the trace of A¹³ is then 2p₁₃+q₁₃ (since trace(A)=2, trace(I)=2 for a 2×2 identity... more precisely trace(pA+qI)=p·trace(A)+q·trace(I)=3p+2q). This recurrence approach is dramatically faster than multiplying A by itself twelve times, and it's the exact technique the 2024 8th April Morning Shift numerical is built to reward.
Once you see the pattern, these become the fastest marks in the chapter. AKS Yadav Sir teaches the exact step-by-step breakdown used above until it's automatic.
Free 30-min mentorship · 100% refund guarantee · 7-day WhatsApp access
| Concept | Formula / Rule |
|---|---|
| Inverse via adjoint | A⁻¹ = adj(A) / |A| |
| Determinant of scalar multiple | |kA| = kⁿ |A| (n×n matrix) |
| Determinant of adjoint | |adj(A)| = |A|ⁿ⁻¹ |
| Adjoint of adjoint | adj(adj(A)) = |A|ⁿ⁻² · A |
| Cramer's rule condition | Δ≠0 → unique solution; Δ=0 with all numerator Δ's=0 → infinite solutions |
| Symmetric/skew decomposition | A = ½(A+Aᵀ) + ½(A-Aᵀ) |
| Cayley-Hamilton (2×2) | A² - (trace)A + (det)I = O |
Advanced leans much harder on List-matching formats than Main does — the 2024 Paper 1 and 2023 Paper 1 questions both use List-I/List-II matching to test multiple matrix properties (symmetric, skew-symmetric, trace conditions) simultaneously against multiple answer scenarios, rather than asking one direct question. The 2022 Paper 1 question builds an entire scenario around harmonic-progression terms feeding into a linear system's consistency — another example of the cross-topic linking we're also seeing more of in Main. If you're targeting Advanced specifically, practice List-matching questions that require verifying 4 separate matrix conditions in one sitting, since that's now a defined recurring format.
Unlike Sequences and Series (this series' most recent post, where JEE Advanced weightage collapsed to 0%), Matrices and Determinants shows real cross-exam consistency — present every single year on Advanced since 2020, with only mild single-digit percentage swings on Main. Against the Chemistry chapters we've covered (Coordination Compounds, Aldehydes/Ketones/Carboxylic Acids), this chapter rewards procedural precision over reaction-breadth recall — two sub-topics carry 81% of the marks, versus Chemistry's more evenly-spread three-pillar structure. If you're choosing where to invest revision hours for guaranteed, predictable return, this chapter is currently one of the safest bets in the whole syllabus.
Week 1 — Determinants & Adjoint foundations: Days 1–3: Determinant calculation, the A·adj(A)=|A|I relationship, and inverse via adjoint. Days 4–7: Scaling rules (|kA|, |adj(A)|, adj(adj(A))) drilled until automatic, with nested-expression practice from real 2024–2026 PYQs.
Week 2 — Systems of linear equations: Days 1–3: Cramer's rule, unique/no-solution/infinite-solution conditions via the coefficient and numerator determinants. Days 4–7: "Infinitely many solutions" parameter-finding problems specifically, since these dominate the Numerical format.
Week 3 — Matrix types and Cayley-Hamilton: Days 1–3: Symmetric/skew-symmetric/orthogonal/idempotent definitions and the standard decomposition identity. Days 4–7: Cayley-Hamilton theorem and reducing high matrix powers using it.
Week 4 — Cross-topic links and full drilling: Days 1–3: Practice questions that link matrices to conics, complex numbers, or combinatorics — don't study this chapter in total isolation. Days 4–5: Full timed sets from JEE Main 2024–2026. Days 6–7: JEE Advanced 2020–2026, with focus on List-matching formats.
Nested adjoint expressions and multi-step system-of-equations problems punish small slips. AKS Yadav Sir's sessions are built around catching exactly these slips before exam day.
Free 30-min mentorship · 100% refund guarantee · 7-day WhatsApp access
Every mistake above is one AKS Yadav Sir sees repeated every session, and corrects before it costs marks on an actual paper.
Free 30-min mentorship · 100% refund guarantee · 7-day WhatsApp access
For anyone figuring out how to study Matrices and Determinants for JEE — yes. At 6.95% weightage in 2026, it's one of the largest, most consistent chapters in JEE Main Algebra, with 107 questions across just the last three years.
Based on our 2024–2026 analysis: 43 questions in 2024, 46 in 2025, and 34 in 2026 — among the highest per-chapter counts in our entire series.
Yes, and reliably so — it's appeared with at least 2 questions every single year since 2020, unlike some chapters that show large year-to-year swings.
Determinants, Adjoint & Inverse — it's the largest sub-topic (42 of 97 categorised combined PYQs) and the scaling rules you learn here (|kA|, |adj(A)|, adj(adj(A))) are prerequisites for nearly every other question type in the chapter.
Sequences and Series rewards translating word conditions into equations; Matrices and Determinants rewards precise, error-free execution of known procedures (adjoint rules, Cramer's rule, Cayley-Hamilton) under time pressure — it's a chapter where careful practice matters more than creative problem-solving.
It's effectively required — any question asking you to find a high matrix power (A¹³, A²⁰, A⁹⁹) is testing this theorem specifically, and attempting repeated matrix multiplication instead will cost you significant time under exam conditions.
Given its consistent high weightage on both exams, treat it as a top-priority Algebra chapter — a full 4 weeks if starting fresh, or 5–6 focused days for pure revision given how procedural (rather than conceptually novel) most of the content is.
They're much more common on JEE Advanced than JEE Main — Advanced questions frequently bundle 4 matrix-property checks into one List-I/List-II format, while Main tends toward direct single-answer questions.
Matrices and Determinants is the most procedurally consistent chapter we've covered in this series — steady weightage on both exams, and a clear two-sub-topic concentration (Determinants/Adjoint/Inverse and Systems of Linear Equations together carrying 81% of the marks) that makes prioritisation straightforward. This guide was built by pulling every JEE Main question from 2024–2026 and every JEE Advanced question from 2020–2026 directly from ExamSIDE's chapter-wise archives, classifying what we could confidently sub-topic-tag, and being explicit about the 28 JEE Main questions (26%) we couldn't classify due to matrix entries rendering as unextractable notation in the source — we said so rather than forcing a number. If you're still working out how to study Matrices and Determinants for JEE efficiently, the short version is: master the two big sub-topics first, drill the scaling rules until they're reflexive, and treat everything else as a secondary layer on top.
