A Six Week Cambridge 9709 A2 Maths Revision Plan That Actually Finishes the Syllabus

Cambridge A-Level Mathematics 9709 A2 revision goes wrong in a predictable way. Students drill one Pure 3 chapter at a time, get strong on each individually, then sit a mixed paper and discover that real A2 questions do not arrive labelled. The technique that worked yesterday is no longer signposted. Two techniques combine in one question. Pure 1 algebra suddenly has to be reflex, not a thinking step.

This plan is built to fix that, not just to cover the 9709 syllabus. It is six weeks, leads with Paper 3 (Pure Mathematics 3), layers in Paper 6 (Probability & Statistics 2), and spends a deliberate week on the one thing topic-by-topic practice never trains: choosing the right method when no one tells you which one to use.

It works whether you are doing Paper 3 only or the full Paper 3 plus Paper 6 A2 pathway.

Before week 1: do this one diagnostic

Sit one full Cambridge 9709 Paper 3 past paper from 2025 onwards, untimed, with no help. Mark it against the official mark scheme, slowly. Then sort your lost marks into four columns: Pure 1 drag (algebra was slow or wrong), method selection (right topic, wrong tool), layering (one topic was fine, the second one inside the same question collapsed), and new content (the topic was actually unfamiliar). The size of the first column tells you whether you need to add Pure 1 maintenance throughout the six weeks. For most students, the answer is yes. Note that the Paper 3 syllabus covers sections 3.1 to 3.9 (Algebra, Logarithmic and exponential functions, Trigonometry, Differentiation, Integration, Numerical solution of equations, Vectors, Differential equations, Complex numbers); the plan covers all nine.

Week 1: Pure 3 algebra and logarithmic foundations (syllabus 3.1 and 3.2)

A2 silently assumes the Pure 1 and Pure 2 toolkit is automatic. Spend Week 1 making sure of that, then layer in the Pure 3 algebra extensions (syllabus section 3.1) and the Pure 3 logarithmic and exponential functions (syllabus section 3.2).

• Pure 1 and Pure 2 maintenance, twice this week: factorising, surds, completing the square, quadratic formula, polynomial division, the factor theorem, modulus equations, $a^x = b$ via logarithms. If any of these still costs you thought, it is the priority.
• The modulus function in Pure 3 form (3.1): solving $|ax + b| = |cx + d|$, $|f(x)| < g(x)$, sketching $y = |ax + b|$, and using $|a| = |b| \iff a^2 = b^2$ and $|x - a| < b \iff a - b < x < a + b$.
• Polynomial division (3.1): dividing a polynomial of degree up to 4 by a linear or quadratic divisor, the factor theorem and the remainder theorem.
• Partial fractions (3.1): the three denominator forms in scope, $(ax + b)(cx + d)(ex + f)$, $(ax + b)(cx + d)^2$ (repeated linear factor), and $(ax + b)(cx^2 + d)$ (quadratic factor that does not factorise). The technique is not the hard part, recognising when a question secretly needs it is.
• Binomial expansion for rational $n$ (3.1): $(1 + x)^n$ valid for $|x| < 1$, including the trick where you factor out a constant first to expand $(a + bx)^n$ as $a^n(1 + \tfrac{bx}{a})^n$, and determining the set of values of $x$ for which the expansion is valid.
• Logarithmic and exponential functions (3.2): laws of logarithms (excluding change of base), $e^x$ and $\ln x$ as inverse functions, solving equations and inequalities like $2^{3x - 1} < 5$, and using logarithms to transform $y = kx^n$ or $y = ka^x$ to linear form for determining unknown constants from a gradient and intercept.

End each session by redoing, from blank paper, one question you got wrong. If you cannot, you have not learned it yet.

Week 2: Pure 3 differentiation, integration techniques, and differential equations (3.4, 3.5, 3.8)

This is the heaviest week. The 9709 Paper 3 syllabus extends differentiation and integration well beyond Paper 2, and then layers them inside differential equations.

• Differentiation (3.4): derivatives of $e^x$, $\ln x$, $\sin x$, $\cos x$, $\tan x$, and $\tan^{-1} x$, plus constant multiples, sums, differences, and composites. The product rule, the quotient rule, and the first derivative of functions defined parametrically ($x = t - e^{2t}$, $y = t + e^{2t}$) or implicitly ($x^2 + y^2 = xy + 7$), including tangents and normals. Derivatives of $\sin^{-1} x$ and $\cos^{-1} x$ are not required.
• Integration by substitution (3.5): standard substitutions, including those signposted by the question (substitution will always be given for indefinite integrals, e.g. integrate $\sin^2 2x \cos x$ using $u = \sin x$).
• Integration by parts (3.5): the formula $\int u,\frac{dv}{dx},dx = uv - \int v,\frac{du}{dx},dx$, when to apply it twice, and the standard targets like $\int x\sin 2x,dx$, $\int x^2 e^{-x},dx$, $\int \ln x,dx$, $\int x\tan^{-1} x,dx$.
• Standard integrals (3.5): $\int e^{ax + b},dx$, $\int \frac{1}{ax + b},dx$, $\int \sin(ax + b),dx$, $\int \cos(ax + b),dx$, $\int \sec^2(ax + b),dx$, and the key Paper 3 addition $\int \frac{1}{x^2 + a^2},dx = \frac{1}{a}\tan^{-1}\frac{x}{a} + c$.
• The $\frac{f'(x)}{f(x)}$ recognition pattern: $\int \frac{f'(x)}{f(x)},dx = \ln|f(x)| + c$, including $\int \tan x,dx$ and $\int \frac{x}{x^2 + 1},dx$.
• Combining partial fractions with integration: the most common Paper 3 integral type. $\int \frac{3x + 1}{(x + 1)(x - 2)},dx$ is solvable only after the decomposition. Also use double-angle formulae to integrate $\sin^2 x$ or $\cos^2(2x)$.
• First-order differential equations (3.8): formulate a rate-of-change statement as a differential equation, find by integration a general form of solution where the variables are separable, apply an initial condition for the particular solution, and interpret the solution in context (the modelling questions: cooling, population, mixing).

By the end of this week, you should be able to look at an unfamiliar integral and identify which technique it needs before writing anything down. If you reach for parts when substitution would have been faster, log it, that is exactly the method-selection error A2 punishes.

Week 3: Pure 3 trigonometry and numerical methods (3.3 and 3.6)

Both topics are smaller than the integration week, but both lose marks for predictable reasons.

• The six trigonometric functions (3.3): secant, cosecant, cotangent and their relationships to cosine, sine, tangent, with properties and graphs for angles of any magnitude.
• Trigonometric identities (3.3): $\sin^2\theta + \cos^2\theta = 1$, $1 + \tan^2\theta = \sec^2\theta$, $1 + \cot^2\theta = \csc^2\theta$, the expansions of $\sin(A \pm B)$, $\cos(A \pm B)$ and $\tan(A \pm B)$, and the double-angle formulae for $\sin 2A$, $\cos 2A$, $\tan 2A$.
• The $R\sin(\theta \pm \alpha)$ and $R\cos(\theta \pm \alpha)$ forms (3.3): how to find $R$ and $\alpha$, when to use them for maximum/minimum values, and when to use them for equation solving.
• Equation solving that requires choosing the right identity: typical Paper 3 examples include $\tan\theta + \cot\theta = 4$, $\sec^2\theta - 5\tan\theta - 2 = 0$, and $3\cos\theta + 2\sin\theta = 1$. The questions never label which identity to reach for.
• Numerical solution of equations (3.6): locating a root by graphical means or sign change (e.g. finding consecutive integers a root lies between), the iteration notation $x_{n+1} = F(x_n)$, the staircase or cobweb diagram interpretation, and understanding that an iteration may fail to converge (knowledge of the formal convergence condition is not required).

Trigonometry is where method selection lives. Spend half this week practising "name the identity" on a random page of past-paper questions, before solving them. Two minutes of recognition practice prevents most lost marks.

Week 4: Pure 3 complex numbers and vectors (3.7 and 3.9)

The two newest A2 topics. They feel intimidating until you separate the algebra from the geometry, then they become some of the most predictable marks on the paper.

• Complex numbers (3.9): real and imaginary parts, modulus, argument (usually in $-\pi < \theta \leq \pi$), conjugate $z^*$, arithmetic in Cartesian form $x + iy$ (multiplication and division with full working), conjugate pairs as roots of real-coefficient polynomials (solving a cubic or quartic given one complex root), modulus-argument form $r(\cos\theta + i\sin\theta)$ or $r\mathrm{e}^{i\theta}$, multiplication and division in polar form using $|z_1 z_2| = |z_1||z_2|$ and $\arg(z_1 z_2) = \arg z_1 + \arg z_2$, finding the two square roots of a complex number in exact Cartesian form (e.g. of $5 + 12i$), and loci on an Argand diagram ($|z - a| < k$, $|z - a| = |z - b|$, $\arg(z - a) = \alpha$).
• Vectors (3.7): standard notations including $\binom{x}{y}$, $x\mathbf{i} + y\mathbf{j}$, $\overrightarrow{AB}$, addition and subtraction, scalar multiplication, magnitude, unit vectors, displacement and position vectors in 2D and 3D. The vector form of a line $\mathbf{r} = \mathbf{a} + t\mathbf{b}$, determining whether two lines are parallel, intersect, or are skew (finding the point of intersection where it exists, but shortest distance between skew lines is not required), and using the scalar product to find the angle between two lines and the foot of the perpendicular from a point to a line. Questions may involve 3D solids like cuboids and tetrahedra. The vector product is not required.

Practise the loci questions on paper, not in your head. The diagram is the question, the algebra is the answer.

Week 5: Mixed unsignposted Paper 3 past papers, under exam conditions

Stop revising by topic. This is the week that separates students who got an A in their mocks from students who get an A in the real paper.

Sit whole Cambridge 9709 Paper 3 papers from 2025 onwards, to time, in one sitting, with the calculator, formula booklet, and pen you will use on the day. Mark them honestly. Keep an error log with four columns: the question, what you did, what the mark scheme wanted, and the type of error (Pure 1 drag, method selection, layering, arithmetic). Re-attempt every logged question 48 hours later, from blank paper.

Target: at least three full Paper 3 past papers under timed conditions this week. If you cannot fit three, prioritise the most recent series first. Cambridge tends to repeat question styles closely from one year to the next, so the most recent papers are the highest-signal practice you can do.

Week 6: Probability & Statistics 2 (Paper 6), or a second mixed Paper 3 week if you do not sit Paper 6

Paper 6 rewards a small number of well-drilled habits more than any other 9709 component. The syllabus splits it into sections 6.1 to 6.5.

• The Poisson distribution (6.1): $P(X = r) = \frac{e^{-\lambda}\lambda^r}{r!}$ with mean and variance both equal to $\lambda$, the Poisson as a model for random events, the Poisson approximation to the binomial when $n$ is large and $p$ is small (rough guide: $n > 50$, $np < 5$), and the normal approximation to the Poisson with continuity correction when $\lambda > 15$ approximately.
• Linear combinations of random variables (6.2): $E(aX + b) = aE(X) + b$, $\text{Var}(aX + b) = a^2\text{Var}(X)$, $E(aX + bY) = aE(X) + bE(Y)$, and $\text{Var}(aX + bY) = a^2\text{Var}(X) + b^2\text{Var}(Y)$ for independent $X$ and $Y$. If $X$ is normal then so is $aX + b$; if $X$ and $Y$ are independent and normal then $aX + bY$ is normal; if $X$ and $Y$ are independent Poisson then $X + Y$ is Poisson.
• Continuous random variables (6.3): the concept of a probability density function over a single interval, using a pdf to calculate probabilities and the mean and variance of a distribution, and locating the median or other percentiles by direct consideration of area under the density function. Explicit knowledge of the cumulative distribution function is not required.
• Sampling and estimation (6.4): the distinction between a sample and a population, the sample mean as a random variable with $E(\bar{X}) = \mu$ and $\text{Var}(\bar{X}) = \frac{\sigma^2}{n}$, the Central Limit Theorem (informal: for large $n$, $\bar{X}$ is approximately normal), unbiased estimators of the population mean and variance from raw or summarised data (the $\frac{n}{n-1}$ factor on the sample variance), the confidence interval for a population mean (normal with known variance, or large sample), and an approximate confidence interval for a population proportion from a large sample.
• Hypothesis tests (6.5): state $H_0$, $H_1$, the significance level, the test statistic, the rejection (critical) region or p-value, and the conclusion in context. The tests in scope are: a test on a single observation from a binomial or Poisson distribution (direct evaluation or normal approximation), and a test on the population mean when the population is normal with known variance or the sample is large. Understand the terms Type I error (rejecting $H_0$ when it is true) and Type II error (failing to reject when $H_0$ is false), and calculate these probabilities for normal-based tests or direct binomial and Poisson evaluations.

Paper 6 questions are wordy. Underline the values, write the distribution, then translate every English sentence into a probability expression before reaching for a formula.

How to know you are actually ready

You are ready when you can open any Cambridge 9709 A2 paper at a random question and know, within fifteen seconds, which two techniques the question is testing and which to start with. That recognition, layered on automatic Pure 1 and Pure 2 algebra, is what A2 grades reward. If you can still only score well on chapter-by-chapter practice but flop on mixed papers, Week 5 is the missing piece, do it again.

Inside The Practice Book, every Cambridge 9709 A2 subtopic has chapter-mapped questions, unsignposted combined-style questions, and full Paper 3 and Paper 6 timed papers from 2025 onwards, so the fix-every-mistake loop takes minutes a day instead of an evening of hunting through mark schemes.