Quick answer
Spaced repetition brings a question back just before you would forget it, so a fix you learn once actually sticks. The scheduling algorithm matters: FSRS is the modern successor to SM-2 (the 1980s SuperMemo method), it models memory with three variables instead of one, and in published benchmarks it reaches the same retention with fewer reviews. CrackTMUA runs FSRS on every question you get wrong.
You solve a hard TMUA question, read the worked solution, and feel the click of understanding. Two weeks later the same idea comes up in a mock and it is gone. That is not a discipline problem; it is how memory works. Spaced repetition is the technique that fixes it, and the algorithm doing the scheduling matters more than most people realise. This guide explains what spaced repetition is, why it is a near-perfect fit for TMUA prep, and how the modern FSRS algorithm improves on the classic SM-2 method that most study tools still use.
Key fact
The one-line version: reviewing a question right before you are about to forget it is far more efficient than re-reading it constantly or cramming. FSRS predicts that moment more accurately than SM-2, so you retain more of what you fix while spending less time reviewing.
What is spaced repetition?
In the 1880s the psychologist Hermann Ebbinghaus measured his own memory and found it decays along a predictable curve: recall drops fast at first, then levels off. Crucially, each time you successfully recall something, the curve gets flatter, so the memory lasts longer before the next review is needed.
Spaced repetition exploits this. Instead of reviewing everything every day (wasteful) or only once (useless), it schedules each item for review at expanding intervals, timed to catch you just before you would have forgotten. A question you keep getting right is pushed weeks, then months, into the future. A question you keep missing comes back tomorrow. Your effort is spent exactly where memory is weakest.
Why spaced repetition suits the TMUA specifically
Three features of the TMUA make it an unusually good match.
First, the test rewards retained technique, not one-off effort. The TMUA is built on a small set of recurring ideas and traps: quantifier logic, "must be true" reasoning, disguised quadratics, log and surd manipulation. Once you have seen how a trap works, the value is in still remembering it on exam day, possibly months later. Spaced repetition is the tool built for exactly that.
Second, your prep window is long and finite. Most candidates start months before an October or January sitting. Over that window, unmanaged, early learning quietly decays. Spaced repetition keeps every fix you make alive to test day without you having to manually track what to revisit.
Third, Paper 2 reasoning patterns are learnable and forgettable in equal measure. The mathematical reasoning paper leans on a handful of proof and logic structures. They are trainable, but they slip away fast if not revisited. Scheduling them for review is how you keep them sharp.
SM-2: the classic algorithm
Most spaced-repetition tools, and some TMUA platforms, run SM-2. It comes from SuperMemo, the software that popularised spaced repetition in the late 1980s. SM-2 tracks a single number per card, an "ease factor", and adjusts it up or down each time you review based on how well you recalled the item. The interval to the next review is the old interval multiplied by that ease factor.
SM-2 was a genuine breakthrough and it still works. But it has real limits. It describes memory with one variable, so it cannot separate how hard an item is for you from how durable the memory currently is, two things that behave differently. Its ease-factor rule is a hand-tuned heuristic from decades ago, not fitted to real data. And it has no concept of a target retention rate, so you cannot tell it "keep me at 90% recall" and have it schedule accordingly.
FSRS: the modern successor
FSRS (the Free Spaced Repetition Scheduler) is the current state of the art, and it is what CrackTMUA uses. Instead of one number, it models your memory of each item with three:
- Stability: how many days the memory will last before recall probability drops to your target. High stability means you can wait a long time before reviewing.
- Difficulty: how inherently hard this particular item is for you, tracked separately from stability.
- Retrievability: the probability you can recall the item right now, which decays over time along a power-law forgetting curve.
The key difference is that FSRS is fitted to data, not guessed. Its default parameters were trained on hundreds of millions of real reviews from tens of thousands of learners, and it can even personalise to an individual once it has enough of their history. You also set a target retention (90% is the sensible default), and FSRS works out the interval that keeps you there. In the open-source project's published benchmarks, FSRS reaches the same retention as SM-2 with meaningfully fewer reviews, which for a time-pressed applicant means less revision for the same result.
SM-2 vs FSRS: what actually differs
| SM-2 (1980s SuperMemo) | FSRS (modern) | |
|---|---|---|
| Memory model | One number (ease factor) | Three (stability, difficulty, retrievability) |
| Where the rules come from | Hand-tuned heuristic | Fitted to hundreds of millions of reviews |
| Forgetting curve | Not modelled explicitly | Explicit power-law curve |
| Target retention | None | You set it (e.g. 90%) |
| Personalisation | No | Yes, from your own review history |
| Typical result | Works | Same retention, fewer reviews |
Neither is magic, and both are far better than re-reading notes. But if a tool is going to schedule your revision, there is no reason to use the 1980s method when the modern one is free, better-tested, and demonstrably more efficient.
How CrackTMUA uses FSRS
On CrackTMUA, spaced repetition is not a separate deck you have to build by hand. Every question you attempt becomes a card automatically, and the FSRS engine schedules it from your result. Get one wrong and it comes back soon; get it right and it drifts into the future. When cards are due, your Review page serves them one at a time: you answer, see the full worked solution, then rate how well you knew it (Again / Hard / Good / Easy), and FSRS uses that to schedule the next appearance, showing you the predicted interval under each button.
Free accounts get 5 reviews a day, which is enough to keep a light deck healthy; Premium removes the cap and adds a retention forecast so you can clear a backlog and see how much you are truly holding on to. Either way the scheduling is FSRS, so the questions you have already fixed keep coming back at the right moment instead of quietly fading before the exam. Because CrackTMUA knows your exam sitting, it also caps intervals at your test date, so a card is never scheduled for a day after you sit the paper.
How to use spaced repetition well for the TMUA
A few habits are the difference between spaced repetition working and it feeling like a chore.
- Little and often. A short daily review of your due cards beats a long weekly binge. The whole point is to catch each item just before you forget it, which only works if you show up regularly.
- Rate honestly. When you rate a card Again, Hard, Good or Easy, be truthful about how it felt. FSRS schedules from your ratings, so flattering yourself only means you see the card too late next time and forget it for real.
- Trust the schedule. If a card is not due, you do not need to touch it. Resist the urge to re-drill things you already know; that time is better spent on new practice questions.
- Keep practising. Reviews only exist because you practised. Your deck grows and stays useful as long as you keep working new questions, so pair a daily review with fresh practice from the question bank.
Used this way, spaced repetition turns the questions you get wrong into permanent gains. Over a months-long TMUA prep window, that retained technique is where a large part of your final score is quietly won. For the bigger picture, see our TMUA revision guide and how to prepare for the TMUA.
The fastest way to feel the difference is to build a small deck: do a handful of questions in the question bank, get some wrong on purpose, and watch them resurface on the schedule FSRS chooses.
Practise the real TMUA, free
Drill 400+ questions, every official past paper plus 100+ original, trap-based ones, each with a full worked solution, then sit full mocks in a replica of the real exam screen. Spaced repetition and a predicted band included. No PDFs.