TMUA 2017: The Format Change Year, and How Its Scores Were Read

TMUA 2017 was the year the test changed shape: total time went from the two hours of the first sitting to 2 hours 30 minutes, with Paper 1 and Paper 2 taken together, each 20 multiple-choice questions in 75 minutes, and that format has been kept ever since. It was also the first year under the Cambridge Assessment Admissions Testing brand, and the official Explanation of Results moved to a new scale anchor: the scale was designed so that approximately one third of candidates achieve overall scores higher than 6.5, with high scores capped at 9.0. This page gathers the 2017 paper structure, the scoring and reporting rules, what the official Explanation of Results says about that year's score distributions, and the topic emphasis and difficulty as catalogued by FrontierVUE, plus why a 2017 score lines up neither with 2016 nor with scores from 2024 onwards.
TMUA 2017 at a Glance
Papers sat
Paper 1 and Paper 2 taken together, 2 hours 30 minutes in total, each paper 20 multiple-choice questions in 75 minutes; this format was set in 2017 and is still in use
What changed from the year before
The 2016 first sitting ran to two hours with 60 minutes per paper, in that year only; from 2017 it became 2 hours 30 minutes with 75 minutes per paper
How it was scored
Marks for correct answers and no penalty for wrong ones; results on a 1 (low) to 9 (high) scale to one decimal place, with Paper 1 and Paper 2 each reported separately alongside the overall score
Official scale anchor
Official wording: the scale was designed so that approximately one third of candidates achieve overall scores higher than 6.5, and high scores are capped at 9.0; this sentence first appears in 2017 and stays word-for-word identical through 2022
Delivery body and universities
The brand from this year is Cambridge Assessment Admissions Testing; at this sitting Warwick joined Durham and Lancaster on the list of universities encouraging the test, with results also shareable with LSE
Difficulty (FrontierVUE catalogue)
Of the 20 Paper 1 questions held, 9 are rated easy and 11 medium, with none rated hard; of the 20 Paper 2 questions, 4 are easy, 12 medium and 4 hard

The paper structure, scoring rules and scale anchor come from the official Explanation of Results for TMUA 2017 and the official test description pages; the difficulty line is FrontierVUE's own catalogue rating, assigned by the platform rather than taken from an official document. FrontierVUE has no affiliation with, endorsement from, or partnership with UAT-UK, Cambridge Assessment or the University of Cambridge, and this page is an independent compilation.

01

What TMUA 2017 tested and how results were reported

TMUA 2017 is the year the format changed. The first sitting in 2016 ran to two hours with 60 minutes per paper, in that year alone; from 2017 the test was extended to 2 hours 30 minutes in total with 75 minutes per paper, while the question count stayed at 20 multiple-choice questions per paper, and that shape has held ever since. Scoring is based on the number of correct answers with no marks lost for wrong ones, so on the official guidance every question is worth attempting, and neither a calculator nor a dictionary may be used. In 2017 Paper 1 was officially titled Mathematical Thinking and Paper 2 Mathematical Reasoning; Paper 1 was renamed Applications of Mathematical Knowledge by 2021 at the latest, while the division of labour between the two papers stayed the same.
2016 (first sitting)2017 (after the change)
Total timeTwo hours2 hours 30 minutes
Timing per paper60 minutes75 minutes
Questions per paper20 multiple-choice20 multiple-choice
Official scale anchorApproximately 50% of candidates above 5.0, approximately 10% above 7.5Approximately one third of candidates above 6.5, high scores capped at 9.0
Brand on the paperworkAdmissions Testing ServiceCambridge Assessment Admissions Testing
Universities encouraging the testDurham and Lancaster, with results shareable with LSEDurham, Lancaster and Warwick, with results shareable with LSE

The differences come from the official test description pages and the two years' Explanation of Results. The scale anchor row compares only how the board worded the scale; it does not imply any conversion between the two years' scores.

Format

Two papers taken together, 2 hours 30 minutes in total

Established in 2017 and still in use today

Questions and timing per paper

20 multiple-choice questions in 75 minutes each

Marking

Marks for correct answers, none deducted for wrong ones

On the official guidance it is worth attempting every question

Score scale

1 (low) to 9 (high), reported to one decimal place

Candidates who do not respond to any items receive a score of 0; this wording appears in the Explanation of Results from 2016 to 2019 and drops out from 2020

What was reported

An overall score plus separate scores for Paper 1 and Paper 2

That year's Explanation of Results carries distribution charts for the overall score and for each of the two papers

Sittings

One autumn sitting a year, throughout the CAAT era

The October and January pair of sittings only arrived with UAT-UK

Calculator and dictionary

Not permitted

From the official current description of this two-paper format

Who saw the score

Candidates decided for themselves whether to share results with universities

A CAAT-era rule, quite unlike the automatic sending via UCAS in the UAT-UK era

On the uptake side, the 2017 sitting added Warwick to the list of universities encouraging the test alongside Durham and Lancaster from the first year, and candidates could also share results with LSE's mathematics and economics programmes. The list kept growing afterwards: by February 2018 at the latest, the official pages show Sheffield and Southampton on it too.
02

How to read a TMUA 2017 score

Reading a 2017 TMUA score comes down to three statements in that year's Explanation of Results: results are reported on a scale running from 1 (low) to 9 (high) to one decimal place; candidates who do not respond to any items receive a score of 0; and the scale was designed so that approximately one third of candidates achieve overall scores higher than 6.5, with high scores capped at 9.0. Alongside the overall figure, the document also shows the score distributions for Paper 1 and Paper 2. Note that the third statement is new wording for 2017 and is not the same anchor the board used in 2016.
The scale: 1 to 9, to one decimal place
The official wording is that results are reported on a scale running from 1 (low) to 9 (high), with scores reported to one decimal place. What appears on the result slip is a scaled score, not a raw count of correct answers and not a percentage.
What 6.5 meant that year
The quantitative statement about the scale in the 2017 Explanation of Results is that the scale was designed so that approximately one third of candidates achieve overall scores higher than 6.5. This wording is new in 2017, and the exam board repeats it word for word in each Explanation of Results through 2022. That document gives the anchor and the 9.0 cap; it does not set out band-by-band figures or a pass mark.
The cap at 9.0
High scores are capped at 9.0, so no score above 9.0 appears on a result slip. It also means the scale stops separating candidates at the very top.
When a 0 appears
The official wording is that candidates who do not respond to any items receive a score of 0. The sentence appears in the Explanation of Results from 2016 through 2019 and is absent from 2020 onwards.
Why there were three scores rather than one
In the CAAT era, Paper 1 and Paper 2 were each reported separately alongside the overall score, and the Explanation of Results carried three distributions every year: overall, Paper 1, Paper 2. That differs from the practice since 2024, where the two papers are equated and scaled together into a single overall score, and it is the detail most often misread on an older result slip.
What the distribution charts do and do not tell you
The 2017 Explanation of Results carries three bar charts, for the overall distribution and for Papers 1 and 2, introduced with the line that the distributions are shown below. The document presents them as charts and gives no accompanying numeric table, so this page makes no percentage estimates from them; to see their shape, go to the official PDF listed in the sources.
Even within the CAAT era, a 2017 score does not line up directly with 2016. For 2016 the board anchored the scale on approximately 50% of candidates above 5.0 and approximately 10% above 7.5; for 2017 it switched to approximately one third above 6.5 with high scores capped at 9.0. Those sentences describe different design targets for the scale, so before reading any older score, check which anchor the board used that year.
Still less can a 2017 score be set beside one from 2024 onwards. From 2024 the TMUA has been delivered by UAT-UK: Papers 1 and 2 are equated and scaled together into a single overall score, and the scale is anchored on the October cohort with the median fixed at 4.5, roughly 10% of candidates above 7.0, a floor at 1.0 and a cap at 9.0. The board is explicit about the change: this scaling process has been revised for 2024/25, and scores should not be compared directly with TMUA scores from earlier years. A 6.5 in 2017 and a 6.5 from 2024 onwards are measured with different rulers, so no single trend line can run across the two eras.
03

Topic emphasis and difficulty in 2017 (FrontierVUE catalogue)

Every figure below comes from the FrontierVUE catalogue: the platform's own difficulty ratings and topic tags for the 2017 papers. FrontierVUE assigns those ratings itself; the 2017 Explanation of Results reports the score scale and the distributions rather than commenting on how hard the two papers were.

20 questions

Held from Paper 1: 9 rated easy, 11 medium, none hard (FrontierVUE catalogue)

20 questions

Held from Paper 2: 4 rated easy, 12 medium, 4 hard (FrontierVUE catalogue)

9 vs 4

Questions rated easy: Paper 1 against Paper 2 (FrontierVUE catalogue)

In catalogue terms the gap between the two papers sits at both ends of the range. Paper 1 has 9 of its 20 questions rated easy and 11 medium, with none rated hard; on Paper 2 the easy count falls to 4, four questions come out as hard and the remaining 12 sit at medium. That fits the division of labour between them: Paper 1 asks you to apply mathematics you already know, while Paper 2 tests reasoning and argument itself, and reasoning questions can be made harder without moving outside the same syllabus. Splitting your time evenly across the two papers, as if they were one undifferentiated sitting, is how candidates get stuck on Paper 2.

Integrating powers of x: definite and indefinite integrals, including expressions that must be simplified first

5

Syllabus code MM7.2

Quadratic functions, the discriminant and completing the square

3

Syllabus code MM1.3

Sequences: nth-term formulae and simple recurrence relations

3

Syllabus code MM2.1

Solving simple trigonometric equations over a given interval

3

Syllabus code MM4.6

Differentiating powers of x, including expressions that must be simplified first

3

Syllabus code MM6.2

Applications of differentiation: tangents, normals, stationary points, increasing and decreasing functions

3

Syllabus code MM6.3

Graph transformations, their composition, and composite function notation

3

Syllabus code MM8.2

Necessary and sufficient conditions

2

Tag Arg2

Gradient and intercept of a straight line, parallel and perpendicular pairs, finding the equation of a line

2

Syllabus code M4.10

Setting up and solving equations and simultaneous equations, including linear with quadratic

2

Syllabus code M4.15

The counts above are topic-tag occurrences across both 2017 papers in the FrontierVUE catalogue. Calculus forms the spine of the year: integrating powers of x alone shows up 5 times, the highest count of any single tag, and together with differentiating powers of x and the applications of differentiation it is the block worth revising first. Next come quadratics, sequences and trigonometric equations, each appearing 3 times, plus graph transformations. The tag to drill on its own is necessary and sufficient conditions: those questions ask you to judge which of the two a condition is, rather than to compute an answer, and they train differently from calculation.
04

How to practise the 2017 papers on FrontierVUE

01Step 1

Open the TMUA past paper list

Go to /papers/tmua to see the TMUA past papers held on FrontierVUE.

02Step 2

Find the two 2017 papers

Locate Paper 1 and Paper 2 under 2017.

03Step 3

Sit each paper against the clock

Each paper is 20 questions in 75 minutes. Time yourself one paper at a time first to build the pacing, then run both back to back for the full 2 hours 30 minutes.

04Step 4

Review the solutions and close the gaps

After submitting you get a question-by-question breakdown with worked explanations; take the topic tags behind your mistakes back into topic practice, starting with the calculus tags that dominate this year.

16 papers

TMUA past papers held, 320 questions across the 16, plus 2 official specimen papers (FrontierVUE catalogue)

1102 questions

TMUA questions on the site in total, including mock papers and the topic bank, not the past-paper count (FrontierVUE catalogue)

Paper 1 and Paper 2

The two 2017 papers available to sit, 20 questions in 75 minutes each

2017 is the first paper on the modern format, with exactly the timing and question count still in use today, which makes it one of the better places to build a feel for the 75 minutes. Once you have sat both papers, fill the method gaps with the TMUA guides on the site: the pieces under /exams/tmua/guides/ are designed to pair with the past papers.

TMUA guides

/exams/tmua/guides/

paper-1What Paper 1 demands and how to train for it
paper-2The reasoning question types on Paper 2 and how to handle them
logic-and-proofLogic and proof, matching the necessary and sufficient condition questions
syllabusThe syllabus, for mapping your weak tags onto the specification
time-managementHow to spend the 75 minutes on a single paper
scoringHow scores are produced and read, including the difference between eras

Tap a branch to unfold

After a full run at the papers, the practice centre also offers topic-based and random practice, which is a faster way to drill the specific tags behind your mistakes than sitting the whole paper again.
FAQ

Frequently asked questions

Can I still practise the TMUA 2017 papers?
Yes. FrontierVUE holds both Paper 1 and Paper 2 from 2017, so you can sit each as 20 questions in 75 minutes and review worked explanations afterwards. Because 2017 is the first sitting on the modern format, its timing and question count match the test as it stands today, which makes it good practice. Bear in mind, though, that 2017 sits in the CAAT era, whose reporting model differs from the UAT-UK era that began in 2024: the questions are fine practice, but scores have to be read era by era.
How did TMUA 2017 differ from 2016?
Three things changed. First, the timing: the 2016 first sitting ran to two hours with 60 minutes per paper, in that year alone, while from 2017 the test became 2 hours 30 minutes with 75 minutes per paper, still 20 multiple-choice questions each. Second, how the board worded the scale anchor: for 2016 it was approximately 50% of candidates above 5.0 and approximately 10% above 7.5, and for 2017 it became approximately one third of candidates above 6.5 with high scores capped at 9.0. Third, the brand: the 2016 Explanation of Results carries the Admissions Testing Service header, and from 2017 it is Cambridge Assessment Admissions Testing. Warwick also joined the list of universities encouraging the test at this sitting.
What counted as a high TMUA score in 2017?
The 2017 Explanation of Results anchors the scale in one sentence: it was designed so that approximately one third of candidates achieve overall scores higher than 6.5, with high scores capped at 9.0. That document presents the distributions as bar charts and lists neither band-by-band figures nor a pass mark, so any claim pinned to a specific percentage comes from somewhere other than that document. This anchor also belongs to the CAAT era: it differs from the 2016 wording and cannot be carried over to scores from 2024 onwards.
Is a 6.5 from 2017 the same as a 6.5 today?
No. The official 2024/25 Explanation of Results states that this scaling process has been revised for 2024/25, and scores should not be compared directly with TMUA scores from earlier years. In 2017, under CAAT, the anchor was approximately one third of candidates above 6.5 with a cap at 9.0, and each paper was reported separately alongside the overall score. From 2024, the two papers are equated and scaled together into a single score anchored with the median at 4.5 and roughly 10% of candidates above 7.0, with a floor at 1.0 and a cap at 9.0. The two eras use different rulers and there is no conversion between them.
Sources

Official sources

By year

Other TMUA years

TMUA 2025TMUA 2024TMUA 2023TMUA 2022TMUA 2021TMUA 2020TMUA 2019TMUA 2018TMUA 2016

FrontierVUE is an independent practice platform. It is not affiliated with or endorsed by UAT-UK, Pearson, OCR, the University of Cambridge, Imperial College London, or any official admissions-test owner.

FrontierVUE 是独立的备考练习平台，与 UAT-UK、Pearson、OCR、剑桥大学、 帝国理工学院或任何官方入学考试主办方均无隶属或背书关系。

Practise TMUA 2017 online

Timed interface, worked solutions and per-question analytics. The two most recent past papers are free to try.

Practise TMUA papersAll TMUA past papers