TMUA Grade Boundaries: Why There Aren't Any Fixed Ones

If you have come from A-levels you are used to grade boundaries: a published table, released after each exam, that tells you exactly how many marks earned an A or a B. It is natural to go looking for the same thing for the TMUA, type "TMUA grade boundaries" into a search bar, and expect a tidy chart. The honest answer is that the table you are looking for does not exist, and this guide explains why, what UAT-UK actually does instead, and what we genuinely do know about how raw performance maps onto a score.

What people mean by "grade boundaries"

At A-level, a grade boundary is a hard line: score this many raw marks and you are awarded that grade, full stop. The boundaries are set after the exam and published, so you can check exactly where you landed. That model trains everyone to expect a public, fixed mark-to-grade table for any serious exam.

The TMUA is not built that way. It does not award letter grades and it does not pass or fail you. Instead it places you on a continuous scale from 1.0 to 9.0, reported in steps of 0.1, and your raw mark is converted onto that scale by a statistical process rather than read off against a fixed line. So the very thing the phrase "grade boundaries" assumes, a published lookup table, is not part of how the TMUA reports results.

So are there TMUA grade boundaries at all?

Not in the form you are picturing. There is no released document that says "14 out of 20 is a 6.0 this year". What exists is a calibration: UAT-UK takes your raw count of correct answers on a paper and maps it onto the 1.0 to 9.0 scale using statistics that account for how hard that particular paper was. The principle behind it is public; the exact numbers are not.

This matters because plenty of forum posts and revision sites confidently quote "boundaries" for the TMUA. Treat those with real caution. They are at best reverse-engineered guesses from a handful of reported results, and at worst simply made up. UAT-UK does not hand out the conversion, so nobody outside the exam board can give you an authoritative one.

Why the TMUA uses a scale, not boundaries

The scale exists to solve one problem: fairness across sittings. No two versions of the test are ever exactly equally hard. One year's Paper 2 might contain a couple of genuinely nasty reasoning questions; the next year's might be a shade gentler. If raw marks were reported directly, a 15 out of 20 in a hard year would look identical to a 15 out of 20 in an easy year, even though the first was the stronger performance.

The calibration fixes this by adjusting raw marks statistically so that a given scaled score means the same standard of performance whatever paper produced it. In a harder year it takes fewer raw marks to reach a 6.5; in an easier year it takes more. The scaled score, not the raw mark, is the thing that stays comparable over time. That is precisely why a fixed boundary table would defeat the purpose: the whole point is that the mark-to-score relationship is allowed to move. We walk through the mechanics in full in our scoring guide.

What this means for you in practice

The most useful consequence is also the most freeing: you should stop trying to reverse-engineer "I need exactly 16 out of 20 to hit my target". You cannot, because the raw mark behind any given score shifts with the paper. Aim instead at fluency and accuracy across the whole syllabus and the Paper 2 reasoning toolkit, which is what actually moves a scaled score in any sitting.

A few practical points fall out of this:

• There is no pass mark. The TMUA does not have a threshold you clear or miss. Your score is a position on a scale, not a verdict.
• There is no public mark scheme to chase. You will not find an official "marks needed" figure, so do not build your preparation around one.
• Universities set the bar, not the exam board. A score is good or bad only relative to your target course. The course-by-course targets live in TMUA score requirements 2026.
• Old past-paper "boundaries" do not transfer. If you marked an old paper against an old conversion, the result will not match today's scale. More on that below.

The anchors we genuinely do know

While there is no boundary table, the scale does have a known shape, and a few solid anchors are worth keeping in mind. These describe where you sit among all candidates, which is the closest honest substitute for a boundary:

• The median score is around 4.5. That is roughly the average candidate, and it sits in the lower-middle of the 1.0 to 9.0 range, lower than most applicants expect.
• Only about 25% of candidates clear 6.0.
• Only about 15% clear 6.5.
• Only about 10% clear 7.0.
• A score of 7.5 and above is exceptional, reached by a small fraction of the field.

Notice these are percentile anchors, not raw-mark boundaries. They tell you what a score is worth relative to the cohort, which is what you actually want to know, without pretending to a precision the exam board never published. For how to read these against a "good" result, see what is a good TMUA score.

Try a real one

It helps to see the kind of question that separates the bands rather than just reading numbers. Here is a genuine past-paper Paper 2 question pitched around the competitive level. Give it a proper attempt before you reveal the solution, and you will get a far better feel for where you stand than any boundary table could give you:

The 2024 rescale: do not compare across it

There is one more reason a single boundary table could never be right: the scale itself was reset. Until 2023 the test was run by Cambridge Assessment; from 2024, UAT-UK and Pearson VUE took over, moved it fully on-screen, and rescaled the grades. The practical effect is that the same raw performance now comes out roughly 1 to 1.5 points lower than it would have on the older papers.

So if you find an old "boundary" or score from a pre-2024 paper, it is not comparable to today's scale: a given score today reflects a tougher scale than the same number on a pre-2024 paper, and universities adjusted their published targets after the reset. Comparing your mock score on an old paper against a modern target is one of the most common ways applicants needlessly panic. The scale moved; your ability did not. There is more on this in our is the TMUA hard guide.

What we know vs what we don't

To pull it together honestly, here is the line between solid fact and the things no public source can give you:

Use the left column to plan, and ignore any source pretending to fill in the right one with precision.

The closest thing to a personal "boundary"

If what you really want is a sense of where your current ability would land on the scale, the answer is not a public table; it is your own performance under realistic conditions. The most reliable read comes from sitting timed questions and full mocks and seeing where your accuracy puts you against the percentile anchors above.

That is exactly what our predicted band tool is for. It estimates the scaled band your recent performance would map to, with topic-by-topic analytics so you can see what is dragging you down. It is an estimate, not an official boundary, and it is a Premium feature, but it is the closest honest substitute for the lookup table the TMUA deliberately does not provide.

The honest takeaway: stop hunting for a boundary table that does not exist, and aim instead at genuine fluency across the syllabus and Paper 2 reasoning. CrackTMUA is built to get you there: a free tier gives you 10 questions a day, and Premium opens 400+ questions including 100+ original trap-based ones, 18+ full timed mocks in a replica of the Pearson VUE screen, spaced repetition, and the predicted band, all for £37 one-time (12 months access). Start practising free and see where your scores actually land.

Frequently asked questions