Course-level comparison¶
This page compares a shortlist of mathematics programmes — actual courses, structure, flexibility, subject coverage and the professors who teach the priority areas — rather than general university standing — though a subject-ranking snapshot is included below for calibration (fuller standing lives on Comparison and the Math-axis deep dive). Imperial is the benchmark; every other course is judged against it.
Status: deep dives complete for all 7 (2026-05-28), with overseas fees, entry requirements and a Tier-1/2 credit-emphasis estimate added. Still being refined: exact per-module credit tallies and deeper forum sentiment. A few items remain marked (confirm).
What this comparison optimizes for¶
Programmes are scored against one priority-ordered interest set:
- Tier 1 — Statistics; Probability.
- Tier 2 — Stochastic processes & stochastic calculus (the maths under algorithmic trading); optimization & machine-learning theory; linear algebra.
- Tier 3 — Real analysis & measure theory (the rigorous foundation under probability/statistics); logic, set theory, algorithms.
- Out of scope: number theory, algebraic geometry and other blue-sky pure topics — a department being world-class in these does not count in its favour here.
The point: a "second-tier" university that is genuinely first-tier in these areas — with specific strong professors — can beat a higher-ranked generalist programme.
Measurement¶
UK degrees are measured in credits (CATS: 360 = full BSc, 120/year; an integrated MMath/MSci is 480 over 4 years). Imperial reports in ECTS (~60/year). Credits are used as the comparable backbone, with contact hours added where published. 1 credit ≈ 10 notional learning hours.
Table 1 — Programmes offered (multiple courses per university)¶
| University | Best-fit route (statistics-leaning) | Other maths programmes | Degree types |
|---|---|---|---|
| Imperial (benchmark) | Mathematics with Statistics (G1G3); with Statistics for Finance (G1GH) | BSc Mathematics (G100), Pure Maths (G103), with Math Computation (G102), with Applied Maths/Math Physics (G125); joint Maths & Computing | 6 BSc + 2 MSci |
| Warwick | BSc Mathematics & Statistics (GG13); MMathStat | BSc/MMath Mathematics; MORSE / MMORSE; Data Science BSc/MSci; Discrete Maths BSc | BSc, MMath, MMathStat |
| Manchester | BSc Mathematics → Probability & Statistics theme | Actuarial Science & Mathematics; Mathematics with Finance; Mathematics & Statistics (confirm) | BSc, MMath |
| Durham | BSc Mathematics & Statistics (G111) | BSc/MMath Mathematics (G100/G103); Mathematics & Physics (G427) | BSc, MMath |
| Bristol | BSc Maths with Statistics; Maths with Statistics for Finance | BSc/MSci Mathematics; with Study in Europe/Abroad; Data Science; Economics & Mathematics; Maths & CS; Maths & Philosophy/Physics | BSc, MSci, MEng |
| Bath | BSc Maths & Statistics; BSc Maths, Statistics & Data Science | BSc Mathematics; + professional-placement / study-abroad variants | BSc (+ placement) |
| UCL | BSc Mathematics & Statistical Science (GG13) | Statistics (G300); Statistics, Economics & Finance (GLN0); BSc/MSci Mathematics | BSc, MSci |
Table 2 — Structure & flexibility¶
How much is compulsory vs optional, and how early you can steer toward statistics/probability. (M&S = the Maths & Statistics route.)
| University | Credits/yr | Year 1 | Year 2 | Year 3 | Specialize from | Flexibility |
|---|---|---|---|---|---|---|
| Imperial | ~60 ECTS | general core | core + electives | wide options (50+ modules) | Y3 (partial Y2) | Common core Y1–2; high in Y3/4 |
| Warwick (M&S) | 120 CATS | core | core | no compulsory modules; ¾ Maths/Stats + ¼ options (must incl. ≥4 Statistics); ≤25% other depts | Y3 | Very high final year |
| Manchester | 120 | all mandatory (incl. Probability I, Statistics I) | 20 mandatory + 100 optional; 2 of 3 themes (Pure / Applied / Prob & Stats) | all optional (54+) | Y2 | Locked Y1, very open Y2–3 |
| Durham (M&S) | 120 | broad core (incl. probability, statistics) | flexible | ~20 option areas + project | Y2 | High Y2–3; ≤40 cr other depts |
| Bristol | 120 | shared core | 2 core (Calculus + Math Programming) + options (stats route → more stats/probability) | wide options | Y2 | High from Y2 |
| Bath (M&S) | ~60 (≈120 ECTS) | all compulsory (pure + applied) | compulsory Statistics + options | statistics-focused core + options | Y2 | Structured stats spine + options |
| UCL (M&StatSci) | 120 | core (analysis, algebra, methods, prob & stats) | core (Probability & Inference, Linear Models & ANOVA, Applied Probability) + options | 1 compulsory (Statistical Inference) + wide options | Y3 | Heaviest early core; open Y3 |
Flexibility spectrum: Warwick & Manchester open up most by the final year; UCL front-loads the heaviest guaranteed statistics core; Bath builds a structured statistics spine; Durham & Bristol open from Year 2.
Table 3 — Coverage in the Tier-1/2 areas (named modules / groups)¶
| University | Probability | Statistics | Stochastic / finance | Optimization / ML | Analysis / foundations |
|---|---|---|---|---|---|
| Imperial | applied probability; stochastic simulation | mathematical statistics (with-Statistics stream) | largest UK math-finance group; stochastic calculus | optimization options | analysis core |
| Warwick | ST115 Probability; ST202 Stochastic Processes | mathematical statistics; Bayesian | ST333 / ST406 Applied Stochastic Processes; SF@W finance group | options | analysis core |
| Manchester | Probability I; Markov Processes; Foundations of Modern Probability | Statistics I; Linear Regression; Generalised Linear Models | Stochastic Processes; Martingales w/ Applications to Finance; Time Series | Multivariate Statistics & Machine Learning | Real Analysis; Foundations & Analysis |
| Durham | probability core; PiNE group | statistics + ML options | MATH3251 Stochastic Processes III; MATH3301 Mathematical Finance III (stochastic calculus, Black-Scholes) | machine-learning options | analysis; complex analysis |
| Bristol | Probability; Complex Networks | Monte Carlo methods; time series; big-data | Maths with Statistics for Finance route; Financial Risk Management | statistical ML | analysis |
| Bath | applied probability (Prob-L@B) | Bayesian; Statistical Modelling & Data Analytics | MA30089 Stochastic Processes & Finance | data science / SAMBa | Pure Maths → analysis |
| UCL | Probability & Inference; Applied Probability | Linear Models & ANOVA; Statistical Inference | stochastic systems; stochastic methods in finance | Optimisation Algorithms; CS&ML group | Analysis 1–4 |
Table 4 — Standout professors in the Tier-1/2 areas¶
Where you could get strong supervision on this axis (verified affiliations, 2026-05-28). See Research groups & mentors for the wider professor layer and the "choose a supervisor first" method.
| University | Standout faculty (area) |
|---|---|
| Imperial | Axel Gandy (applied prob/stats; HoD) · Guy Nason (wavelets & time series) · Johannes Muhle-Karbe (math finance) · Eyal Neumann (stochastic analysis/finance) |
| Warwick | Gareth Roberts FRS (MCMC) · Wilfrid Kendall (stochastic calculus/geometry) · Jon Warren (probability) — elite Stats dept + CRiSM + Stochastic Finance @ Warwick |
| Manchester | Peter Neal (probability, MCMC, epidemics) · Tusheng Zhang (stochastic analysis / financial maths) |
| Durham | Andrew Wade (random walks / applied probability) · Sunil Chhita (probability & stat-mechanics) · Ostap Hryniv (probability) |
| Bristol | Christophe Andrieu (MCMC) · Nick Whiteley (data science / statistical ML) · Haeran Cho (time series) · Márton Balázs (probability) |
| Bath | Tim Rogers (random processes/networks; HoD) · Cécile Mailler (probability) · Matt Nunes (time series) — Prob-L@B + SAMBa |
| UCL | Petros Dellaportas (Bayesian / ML / financial modelling) · Codina Cotar (probability / optimal transport) — RSS-accredited Statistical Science |
Fees & entry (2026/27, overseas)¶
| University | Overseas fee/yr | A-level offer | Admissions test | Foundation route accepted |
|---|---|---|---|---|
| Imperial | £42,700 | A*A*A (A* Maths, A* Further Maths) | TMUA required | UCL UPCSE / Warwick IFP (≥80% incl. Maths) |
| UCL | £42,700 | A*A*A (A*A* Maths+FM) + STEP 2 / AEA | STEP (in offer) | UCL UPC |
| Manchester | £36,300 | A*AA (A* in Maths or FM) | none required | INTO Manchester / NCUK / integrated FY |
| Warwick | £35,530 | A*A*A (A*A* Maths+FM), or A*AA + STEP 2 / TMUA 5.0 | TMUA/STEP (for lower offer) | Warwick IFP (selected courses) |
| Bristol | £31,300 | A*A*A, or A*AA (A*A Maths+FM) | optional | foundation pathways |
| Durham | ~£30,000 | A*A*A; A*AA with TMUA 5.0 / STEP 1 / MAT | TMUA/STEP (reduced offer) | Durham ISC pathway |
| Bath | ~£30,000 | A*A*A (A*A Maths+FM); Further Maths mandatory | none if you have Further Maths | Bath International Foundation Year |
Fees are 2026/27 overseas (2027/28 not yet set — expect a small uplift). FM = Further Maths.
Observations: Imperial and UCL are the priciest (£42,700); Durham and Bath the cheapest (~£30,000) — so Bath pairs strong area-fit with the lowest cost. Further Maths is required or strongly preferred almost everywhere (mandatory at Bath; A*A* Maths+FM at Warwick/UCL/Durham). Least test-heavy: Manchester (no test) and Bath (no test with Further Maths); most test-heavy: Imperial (TMUA) and UCL (STEP).
Tier-1/2 credit emphasis (approximate)¶
How much of each degree can be devoted to these areas. Exact per-module credit tallies need each programme's full module catalogue — several are PDF/gated and didn't auto-extract, so this is a best-effort estimate. An exact tally can be done on any one or two priority routes (e.g. the Warwick or Bath best-fit).
| University | Compulsory Tier-1/2 (Y1–2) | Max loadable Tier-1/2 (degree) |
|---|---|---|
| UCL | highest — heavy statistics core Y1–2 | high (wide Y3 options) |
| Bath | high — compulsory statistics spine Y2–3 | high |
| Warwick | moderate-high (probability, mathematical statistics, ST202) | very high — Y3 all optional, ≥4 statistics |
| Manchester | moderate (Probability I, Statistics I + Y2 theme) | very high — Y3 120 cr all optional |
| Bristol | moderate-high (statistics route from Y2) | high (Stats-for-Finance ≥100 cpts statistical) |
| Durham | moderate (Y1 probability + statistics) | high (Stochastic Processes III, Mathematical Finance III + stats/ML) |
| Imperial | moderate (common core probability & statistics) | high (Y3/4 50+ options; with-Statistics stream minimum) |
Rankings — global vs UK (they disagree)¶
Two systems rank these schools very differently, because they measure different things — worth seeing side by side.
| Imperial | Warwick | UCL | Manchester | Bristol | Durham | Bath | |
|---|---|---|---|---|---|---|---|
| QS World by Subject — Maths 2025 (global) | 10 | 31 | 38 | 49 | 51–100 | 51–100 | ~100–150 |
| Complete University Guide — Maths 2026 (UK) | 4 | 5 | 10 | 12 | 6 | 9 | 7 |
QS weights research reputation, citations and international metrics → rewards big research output and global brand. CUG (UK) weights entry standards, student satisfaction (NSS), graduate prospects and research quality → rewards student-focused, high-satisfaction departments. They invert on Bath and Manchester: Bath is 7th in the UK but ~100–150 globally; Manchester is 49th globally but 12th in the UK. Brand/prestige tracks the global (QS) axis, on which the order is Imperial > Warwick > UCL > Manchester > Bristol > Durham/Bath. The synthesis below is area-fit + student experience, a different lens — read it against this table, not instead of it.
Synthesis — on this axis¶
- Imperial (benchmark) — top area depth (dedicated Statistics + Stats-for-Finance streams; UK's largest math-finance group). Catch: hardest admission and highest intensity — only ~57–62% graduate with a 2:1+.
- Warwick — arguably the strongest pure statistics/probability fit (elite free-standing Stats dept, MCMC, stochastic-finance) with the most flexible final year. Rivals/exceeds Imperial on statistics, plausibly less brutal. Top contender.
- Bath — strong on the UK student-experience axis (7th in the UK table) with the best value (~£30k) and satisfaction, plus a genuine but smaller stats/probability group (Prob-L@B, SAMBa, time series). Modest globally (QS ~100–150): its real case is value, experience and admission attainability — not prestige or departmental scale.
- UCL — strongest quant-finance / ML statistical angle (SEF degree, Dellaportas, CS&ML group) and the most guaranteed early statistics; central London. Trade-off: large cohorts / mixed teaching satisfaction.
- Bristol — top statistics (Andrieu MCMC, Whiteley data science, Cho time series) + a dedicated Statistics-for-Finance route. Trade-off: large cohorts / self-study, stressful workload.
- Manchester — the best-signposted Tier-1/2 menu (fully-optional Y3) and the stronger research/brand of the two value picks (QS 49 vs Bath ~100–150). Trade-offs: very large / impersonal, and pricier (£36,300) than Bath/Bristol/Durham.
- Durham — best teaching support (small groups, strong pastoral care; lowest intensity) and strong probability + stochastic-finance; high prestige. Trade-offs: less statistics specialization (some strength is out-of-scope math-physics) and a noted isolation risk for non-British students.
Recurring theme on workload/well-being (priority #4): large-cohort / self-study complaints recur at UCL, Bristol, Manchester; Durham offers the most personal teaching; Bath the highest satisfaction.
Method & what's next¶
- Sources: official programme pages, module catalogues/handbooks and programme-spec PDFs; Discover Uni for contact hours & satisfaction; The Student Room / Reddit / StudentCrowd for what students actually take; department staff pages, Google Scholar and learned-society prizes for professor credentials.
- Done in this pass: overseas fees + entry requirements + foundation routes; the Tier-1/2 credit-emphasis estimate; resolved verify flags (Tusheng Zhang confirmed at Manchester; Bath = 60 credits/yr; Imperial G1F3 = Mathematics, Optimisation & Statistics).
- Still open: exact per-module credit tallies for the best-fit routes; confirm the Warwick IFP covers Maths & Statistics; Bath's exact fee band; deeper forum sentiment.
Deep dives 2026-05-28. Structure figures are first-pass from official pages; professor affiliations verified on that date — professors move, so re-check before relying on any single name.