Thoughts on the TMUA

The TMUA (Test of Mathematics for University Admission) is the aptitude test used by nearly all of the top UK universities for admission into courses like Mathematics, Computer Science, and Economics. It's like ESAT, which is also run by UAT, but specifically focused on mathematics. There is a lot of information about the test and its history already available online, so in this article I'll primarily be giving my personal thoughts, experience, and advice.

My Own Experience

I took the test on the first of the two days in the October sitting. It took me a while to book, because UAT took three times longer than they said it would to deliver my bursary payment. This meant I had to sit the test at a centre in Manchester, so it was a fun day out. I noticed a queue of South Asians waiting for something at Uber, but then I found someone my own age, and understood I was in the right place. After waiting for too long, I sat the test, and found it quite easy. The test was a bit harder than the average paper I'd done, but it wasn't as hard as people made the 2024 paper out to be. Basically, there had been a change in test structure, so relatively difficult questions were placed at the start of the paper in 2024 and 2025, which confused lots of students who expected a linear difficulty gradient as had been the case in the earlier papers.

My revision amounted to doing lots of past papers, 75 minutes each, and not much else. The spreadsheet has all the information you could reasonably ask for. I ended up with an 8.3, so I'd say my strategy worked quite well, and I would advise anyone to spend the bulk of their time grinding through past papers, and community papers. Never pay for TMUA papers, or the 2024 and 2025 papers, because you can definitely find them for free. I might be able to help with getting these hidden papers, but I would advise starting the hunt on Discord. There are lots of scammers about who will charge you £800 for resources, so be careful. At the start of your preparation, I would advise you to review the specification, ensuring that you are confident with all the content to the A-Level standard. I would try to understand the logic content of paper 2, as questions which amount to something like "find is the contrapositive of..." are actually trivial if you understand what the key word means.

For a few hours on the 31st of October, while logged in, you could inspect element the TMUA website, type 'result' into the console, and it would give you your result. Results were only officially released weeks later, but the inspect element results were correct, contrary to the beliefs of some. Really, since the results were obviously there and finalised, UAT should've simply released them earlier. Multiple-choice tests are marked and reviewed almost instantly, so we should've got the results back only about a week after sitting the test.

The Somewhat Overblown Criticism

Of course UAT is incompetent for allowing the leak to happen. Did nobody anticipate that a bunch of Cambridge Maths and CS applicants would try inspect element? The test itself isn't perfect either. The test would be better at differentiating between students if it wasn't multiple-choice, and instead featured longer, more challenging questions, with more time. Unfortunately, this isn't possible, because it would take too long to mark all the papers, including ESAT papers, due to how many people sit these tests. STEP is the exception, featuring long challenging questions, because relatively few candidates sit it to begin with, because it's only used by universities as part of candidates' offers, like A-Levels.

The grading for TMUA is a bit odd, and the curve is harsh, but most people complaining about results in my year were so obviously only doing so because they didn't score as high as they wanted. Nearly everyone sitting the test is predicted some number of A* grades, and half of them have to score less than 4.0, which is considered a poor result. UAT is transparent with the proportion of students getting each grade, and you can see that getting even a 7.0 puts you in the top 3-5% of home (British) students. When people are led to believe anything less than 7.0 as mediocre, of course people are going to be disappointed and angry.

The most misleading and inaccurate criticism came from parents, who obviously didn't understand the test, and refused to read UAT's reports, who complained on behalf of their children. It doesn't even seem like the students themselves thought the results were incorrect, but many entitled parents just couldn't cope with the fact that their child just didn't perform in the top 1% or so that is necessary to get the result they wanted.

All that said, fuck UAT for being so slow with their bursary payments, and charging I believe £75 to ordinary students sitting the test to begin with. At least we all have something to complain about.

Oh, also, UAT effectively doxxed some students, causing them to get phone calls from people who tried to log into their own account, but were randomly logged into the other student's. This is a much worse problem that UAT escaped accountability from because people were too preoccupied complaining about not getting the grade they wanted.

Association with Oxbridge Admission and STEP Performance

Even in my cohort, for whom the TMUA wasn't required for Mathematics at either Oxford or Cambridge, there was strong positive association between TMUA performance and getting an offer. If you have an offer, or you perform well on STEP, you probably did well on TMUA, but this doesn't necessarily hold the other way round, because TMUA is relatively easy and simple to crack. It's agreed that a grade 1 on either STEP paper is more impressive than a 9.0 on TMUA, and perhaps a grade 2 on a STEP paper is more impressive than a result of 7.5 or so on TMUA. This is because the difference between a 7.0 and an 8.0 on TMUA is ultimately just the difference between 80% and 90% (for example) on a relatively undemanding but fast-paced multiple-choice test. Past a certain point, most marks lost on TMUA are due to silly errors or time pressure, rather than a lack of skill and understanding. That said, I do know someone who is capable of getting high 1 grades on STEP papers, but only got a middling (4.0 to 5.0) score on TMUA, so I'm not saying you shouldn't prepare for TMUA.