What TMUA Score Do You Need for Cambridge, Imperial and LSE?

No university publishes a hard TMUA cutoff. That is the first honest thing to say, and most articles on this topic avoid saying it. What we do know: scores run from 1.0 to 9.0, the most common score in October 2025 was 4.5, and roughly 6.0-6.5 or above is broadly competitive for Cambridge and Imperial. The more useful question is not "what is the cutoff" but "what does each university do with the number", because the same 5.8 plays very differently at different universities.

Table of contents

• The scale, and the only honest calibration
• Why nobody publishes a cutoff
• Cambridge: one input in a contextual, college-based read
• Imperial: the score as a screen
• LSE: holistic, with the score as evidence
• UCL, Warwick, Durham and the rest
• What to do with a target band
• Frequently asked questions

Intro

The TMUA is scored on a scale from 1.0 to 9.0. In October 2025, the most common score was 4.5. Those two facts are public. Almost everything else you will read about "required" scores is inference, and some of it is invention. This post sticks to what is verifiable: the published scale, the modal score, the broadly competitive band, and the structural differences in how Cambridge, Imperial and LSE actually use the number. The differences matter more than the number itself.

The scale, and the only honest calibration

Three anchor points:

• 4.5 was the most common score in October 2025. If you score there, you are at the centre of the pack.
• 6.0-6.5+ is broadly competitive for the most selective users of the test, Cambridge and Imperial in particular. It is a band, not a line, and being inside it is evidence rather than a guarantee.
• 9.0 is the ceiling and is rare. You do not need it, and chasing it past 7.0 has poor marginal returns compared with strengthening the rest of your application.

The gap between 4.5 and 6.5 is the whole game. Raw marks do not map linearly onto the scale, so the number of additional correct answers needed to move from one to the other is smaller than the two-point gap suggests. You can see how raw marks convert with our free score converter, and TMUA Score Conversion explained covers why the mapping flexes with paper difficulty and what that does to a raw-mark target.

Why nobody publishes a cutoff

None of the universities using the TMUA publishes a hard cutoff, and this is not coyness. Cohorts shift year to year, papers differ in difficulty, and every one of these universities reserves the right to weigh the score against grades, school context and, at Cambridge, interview performance. A published cutoff would remove flexibility they actively use. So when a forum post or a tutoring company quotes "Cambridge needs 6.8", treat it as marketing. The verifiable statement is the band: 6.0-6.5+ is broadly competitive, lower scores are carried by stronger files more often than people admit, and higher scores do not guarantee anything on their own.

Cambridge: one input in a contextual, college-based read

Cambridge uses the TMUA for relevant courses as one input among several: A-level (or equivalent) performance and predictions, the interview, school and contextual data, all read by individual colleges. There is no central machine that rejects below a number. A college can and does interview applicants whose scores sit below the competitive band when the rest of the file argues for it, particularly with contextual flags, and a high score with a weak interview is not safe.

Practical implication: for Cambridge, the score buys you credibility, not a decision. A 6.5 says the interview is worth taking seriously. A 5.0 does not end the application, but it transfers pressure onto the interview and your grades. You can sanity-check how the whole funnel multiplies out for your profile with the free offer calculator, which decomposes the stages rather than guessing a single odds number.

Imperial: the score as a screen

Imperial sits at the other end of the structural spectrum. With no interviews for most relevant courses and very large application volumes, the score does its work early: it helps determine which applications progress at all. That makes the score relatively more decisive at Imperial than at Cambridge, even though neither publishes a threshold. There is no later interview at which a borderline score gets rescued by a brilliant conversation.

Practical implication: if Imperial is your first choice, the score deserves a larger share of your preparation hours than any other part of the application, because it is the part most likely to be load-bearing. The 6.0-6.5+ band is the sensible target; at the October 2025 mode of 4.5 you are asking the rest of a screened file to do work it will rarely get the chance to do.

LSE: holistic, with the score as evidence

LSE reads applications holistically: grades and predictions, the personal statement, contextual information, and the TMUA score where the course uses it, considered together rather than gated in sequence. The score functions as independent quantitative evidence in that read. A strong score corroborates strong predicted grades; a weaker score invites the reader to look harder at everything else.

Practical implication: at LSE a moderate score is genuinely survivable in a way a screening process makes difficult, but it is never neutral. And because LSE economics-side courses attract enormous fields of applicants with near-identical A-level profiles, the test is one of the few places a file can actually differentiate itself. Treat "holistic" as "everything counts", not "the test barely counts".

UCL, Warwick, Durham and the rest

UCL, Warwick and Durham, among others, also use the TMUA across various maths, computer science and economics courses, in roles ranging from required to optional-but-considered. The honest guidance is unglamorous: read the current course page for each one, because usage changes and a blog post is not a substitute for the source. As a rule of thumb, the competitive band is lower than at Cambridge and Imperial, and a good score submitted optionally is far more likely to help than hurt. If you are sitting the test for one course, it is usually worth flagging your score to every university on your form that will look at it.

If you are still deciding between courses that want the TMUA and Oxford courses that want the MAT, we cover the decision in TMUA vs MAT: Which Test, and Can You Sit Both?.

What to do with a target band

A target of 6.5 is only useful if it changes behaviour. Three ways it should:

1. It sets your hours. Moving from the 4.5 mode to the competitive band is a multi-week project, not a fortnight of past papers. The realistic version is laid out in our 16-week plan.
2. It defines done. When your timed mocks consistently convert to 6.5+, additional grinding has worse returns than fixing your personal statement or interview prep.
3. It tells you what to measure. Mock scores only mean anything if marking and score conversion match the real thing. TMUA Pro includes six auto-marked mocks with analytics, and its score guarantee is set at 6.5 for exactly the reason this post describes: that is where "competitive" starts at the sharpest end.

Related reading

• How to Prepare for the TMUA in 16 Weeks (a Realistic Plan)
• TMUA vs MAT: Which Test, and Can You Sit Both?
• TMUA 2026: Key Dates, Registration and the 28 September Deadline
• Is the TMUA Hard? An Honest Difficulty Breakdown

Closing

There is no cutoff, and anyone selling you one is selling. What exists instead: a 1.0-9.0 scale, a 4.5 mode from October 2025, a broadly competitive band at 6.0-6.5+, and three structurally different uses of the number, Cambridge reading it in context, Imperial screening with it, LSE weighing it holistically. Pick your target band from the university that uses the score most harshly among your choices, then build backwards from there. The score converter and offer calculator are free and exist to make that calibration concrete.

Frequently asked questions

What TMUA score do I need for Cambridge? Cambridge publishes no hard cutoff, and the score is read alongside interviews, grades and contextual data by individual colleges. As a calibration point, 6.0-6.5 or above is broadly competitive; the most common score in October 2025 was 4.5.

What TMUA score do I need for Imperial? Imperial publishes no fixed threshold either, but it leans on the score earlier in the process, using it to help decide which applications progress. A score around the 6.0-6.5+ band is broadly competitive; below the modal 4.5, the score is working against the rest of your file.

What TMUA score does LSE want? LSE does not publish a cutoff and reads the score as one part of a holistic assessment alongside grades, the personal statement and contextual information. A strong score helps; a moderate one is not automatically fatal there in the way a screening-based process can make it.

Is 4.5 a good TMUA score? 4.5 was the most common score in October 2025, so it is the definition of average. For the most competitive courses that is below the broadly competitive 6.0-6.5+ band, but how much it matters depends on how each university uses the score.

Do universities publish TMUA cutoffs? No. None of the universities using the TMUA publishes a hard cutoff, and anyone quoting you an exact number is guessing. The honest version is a competitive band, 6.0-6.5+, plus an understanding of how each university weighs the score.

Can a high TMUA score make up for weaker grades? Partially, and more at some universities than others. A strong score is independent evidence of mathematical ability, which matters most where the score is read holistically. It does not override published grade requirements anywhere.

---

Sources: UAT-UK published score scale; October 2025 results distribution (modal score); university course pages for current test usage. Last updated 2026-06-10.