TMUA SCORING: THE 1 TO 9 SCALE

Each paper has 20 multiple choice questions and no negative marking, so the raw mark is a whole number from 0 to 20. That raw mark is then converted onto the 1.0 to 9.0 scale.

The conversion is not fixed across sittings. It exists precisely so that a score means the same thing from one year to the next even when one year's paper was harder. A raw 16 might convert to 7.0 in one sitting and something a little different in another.

The practical consequence is that chasing a raw mark target from a past paper is unreliable, and comparing your raw marks with someone else's from a different year is meaningless. Track raw marks to see whether your own work is improving. Do not translate them into a predicted score.

The overall score is the plain average of Paper 1 and Paper 2. Two candidates with the same total look identical on paper, which is exactly why the imbalanced candidate is the one who tends to be disappointed.

8.5 on Paper 1 and 6.0 on Paper 2 gives 7.25. Two flat 7.5s give 7.5, and the second candidate did less well on their best paper. Every hour spent lifting a weak Paper 2 from 6.0 to 7.0 moves the overall score by 0.5, and the same hour spent moving an already strong Paper 1 from 8.5 to 8.7 moves it by 0.1.

This is the single most reliable piece of strategy available before the test: work on the weaker paper, even when the stronger paper is more enjoyable to practise. In practice the weaker paper is usually Paper 2, because Paper 2 asks for something that sixth form mathematics rarely trains.

There is no pass mark and no official grade boundary. Universities describe the region they find persuasive rather than publishing a cut.

Broadly, the published guidance across users of the test puts a strong application in the upper part of the scale, with the most competitive mathematics and computer science courses looking well above the midpoint, and courses that use the test as supporting evidence satisfied by less. Warwick has historically set an explicit threshold and confirmed it after the sitting rather than before.

Because the wording changes year to year, the only safe rule is to read the requirement on the specific course page for your year of entry, and to treat any figure quoted on a forum as a rumour until you have found it on the university's own page.

No preparation, here or anywhere, can promise a particular number. The written commitment on this site is about the work and the standard, never about a score.

On both papers the marks lost divide into four kinds, and the four have nothing in common as fixes.

Knowledge: the fact or identity was not known. This is the cheapest gap to close and the least common one among strong candidates.

Method: the content was known but the route chosen was longer than the intended one. This is where most of the missing score sits, and it is closed by rules of the form when I see X, do Y.

Execution: a sign slip, a dropped term, a misread bound. Closed by a checking habit, not by more theory.

Time: the question was solvable and was never reached. Closed by a triage rule about when to abandon a question, not by working faster.

A paper reviewed without that classification produces the feeling of progress and very little of it, because the correction applied is almost always the wrong one for the error made.

Jamie Baek scored 9.0 on the TMUA in October 2024, on first sitting, with both papers completed in a third of the time. Candidate ID UATUK004019, registration ID 485240275.

That result is stated here because it is checkable, not because it transfers. What transfers is the method behind it, which is written out in full and given away: the Blueprint, the annotation system, the live sittings and the paper analyses are all free in the Galleria.