Skip to content
Paper Comparison

December 2025 against December 2023

What one sitting asked that the other did not — as a share of each paper, because December 2025 carries 98 subject questions and December 2023 carries 96. Comparing raw counts across papers of different sizes invents a trend in every row.

Asked more in December 2025

  • Singularities and Residues+3.0 pp

    4.1% (4) in December 2025 · 1.0% (1) in December 2023

  • The Real Line+3.0 pp

    7.1% (7) in December 2025 · 4.2% (4) in December 2023

  • Topology+2.0 pp

    3.1% (3) in December 2025 · 1.0% (1) in December 2023

  • Groups+2.0 pp

    5.1% (5) in December 2025 · 3.1% (3) in December 2023

  • Eigenvalues and Canonical Forms+1.9 pp

    7.1% (7) in December 2025 · 5.2% (5) in December 2023

Asked more in December 2023

  • Continuity and Differentiation-4.2 pp

    2.0% (2) in December 2025 · 6.3% (6) in December 2023

  • Rings and Fields-3.2 pp

    3.1% (3) in December 2025 · 6.3% (6) in December 2023

  • Hypothesis Testing-2.1 pp

    3.1% (3) in December 2025 · 5.2% (5) in December 2023

  • Vector Spaces and Linear Maps-2.1 pp

    3.1% (3) in December 2025 · 5.2% (5) in December 2023

Present on one paper and absent from the other

Absent from one sitting is not absent from the syllabus. A topic carrying two questions one session and none the next has told you almost nothing.

Only in December 2025

  • Linear Programming · 1
  • Metric Spaces · 1

Only in December 2023

  • — none

Every topic, both papers

TopicDecember 2025December 2023Change
ProbabilityProbability, Statistics & Linear Programming7.1% (7)7.3% (7)-0.1 pp
Eigenvalues and Canonical FormsAnalysis & Linear Algebra7.1% (7)5.2% (5)+1.9 pp
Linear Models and MultivariateProbability, Statistics & Linear Programming5.1% (5)6.3% (6)-1.1 pp
The Real LineAnalysis & Linear Algebra7.1% (7)4.2% (4)+3.0 pp
Rings and FieldsComplex Analysis, Algebra & Topology3.1% (3)6.3% (6)-3.2 pp
Partial Differential EquationsODE, PDE & Applied Mathematics5.1% (5)4.2% (4)+0.9 pp
EstimationProbability, Statistics & Linear Programming5.1% (5)4.2% (4)+0.9 pp
Continuity and DifferentiationAnalysis & Linear Algebra2.0% (2)6.3% (6)-4.2 pp
Hypothesis TestingProbability, Statistics & Linear Programming3.1% (3)5.2% (5)-2.1 pp
Vector Spaces and Linear MapsAnalysis & Linear Algebra3.1% (3)5.2% (5)-2.1 pp
Ordinary Differential EquationsODE, PDE & Applied Mathematics4.1% (4)4.2% (4)-0.1 pp
GroupsComplex Analysis, Algebra & Topology5.1% (5)3.1% (3)+2.0 pp
Limit Theorems and Markov ChainsProbability, Statistics & Linear Programming4.1% (4)3.1% (3)+1.0 pp
Inner Product Spaces and FormsAnalysis & Linear Algebra3.1% (3)3.1% (3)-0.1 pp
Numerical AnalysisODE, PDE & Applied Mathematics3.1% (3)3.1% (3)-0.1 pp
Calculus of VariationsODE, PDE & Applied Mathematics3.1% (3)3.1% (3)-0.1 pp
Linear Integral EquationsODE, PDE & Applied Mathematics3.1% (3)3.1% (3)-0.1 pp
Sampling and Design of ExperimentsProbability, Statistics & Linear Programming3.1% (3)3.1% (3)-0.1 pp
Analytic FunctionsComplex Analysis, Algebra & Topology2.0% (2)3.1% (3)-1.1 pp
Functions of Several VariablesAnalysis & Linear Algebra2.0% (2)3.1% (3)-1.1 pp
IntegrationAnalysis & Linear Algebra3.1% (3)2.1% (2)+1.0 pp
Singularities and ResiduesComplex Analysis, Algebra & Topology4.1% (4)1.0% (1)+3.0 pp
Cauchy TheoryComplex Analysis, Algebra & Topology2.0% (2)2.1% (2)-0.0 pp
TopologyComplex Analysis, Algebra & Topology3.1% (3)1.0% (1)+2.0 pp
Lebesgue Measure and IntegrationAnalysis & Linear Algebra1.0% (1)2.1% (2)-1.1 pp
Classical MechanicsODE, PDE & Applied Mathematics1.0% (1)2.1% (2)-1.1 pp
Zeros and MappingsComplex Analysis, Algebra & Topology1.0% (1)2.1% (2)-1.1 pp
Sequences and Series of FunctionsAnalysis & Linear Algebra2.0% (2)1.0% (1)+1.0 pp
Linear ProgrammingProbability, Statistics & Linear Programming1.0% (1)0.0% (0)+1.0 pp
Metric SpacesAnalysis & Linear Algebra1.0% (1)0.0% (0)+1.0 pp

21 of 30 topics moved by less than 1.5 percentage points and are greyed out — at four to thirteen questions a topic, a swing of one or two questions is noise, not a signal.

Two sittings are two sittings. Everything above describes what these two papers contained. None of it predicts the next one, and a topic that fell here is not a topic to drop — the examiners are not running a trend. The aggregate across every complete paper is the more reliable number, and it is still only four papers.