There is no universally best MRCPsych Paper B resource, and any guide that crowns one is usually selling a subscription. Paper B rewards two quite different skills — applied clinical knowledge across the whole psychiatric syllabus and a genuine command of critical appraisal and statistics — so the resource that suits a first-time sitter with twelve weeks is not the one that suits a retaking trainee with three. This article is a decision tree: segment yourself honestly, then follow the branch that fits your time, your budget and your learner profile.
What MRCPsych Paper B actually tests (the anchor for every decision)
Paper B is a written paper of approximately 150 questions in three hours, delivered by computer at Pearson VUE, mixing single-best-answer MCQs and extended-matching items in roughly a two-thirds to one-third ratio; confirm the current count and duration on rcpsych.ac.uk before you book. Two content blocks matter. The larger is clinical: psychopharmacology, organic and old-age psychiatry, substance misuse, the psychotherapies, forensic, and child and adolescent psychiatry, alongside the rest of the applied syllabus. The smaller but decisive block is critical review — on the order of a third of the paper — covering research methods, epidemiology, statistics and the appraisal of evidence. The clinical CASC is a separate, OSCE-style assessment; no written question bank, iatroX included, is a CASC simulator, so keep CASC work out of your Paper B plan entirely.
This distinction anchors every decision, because most candidates over-invest in the clinical block, where they already feel competent, and under-invest in critical review, where marks are quietly won and lost. Whatever resource you choose has to be judged against both blocks, not just the one you enjoy.
The minimum viable stack (and why more is usually worse)
You have four possible slots, and most candidates should fill only the first two.
- One primary question bank for volume and breadth across the clinical block.
- Official calibration material — the Royal College's syllabus, sample questions and marking scheme — to confirm your practice matches the real format and standard.
- One teaching or reference source, added only where your foundations are genuinely weak (for example, a structured critical-appraisal course if statistics unnerves you).
- One modality tool, added only where a specific component demands it — most often a dedicated statistics and critical-review resource.
Bolting on a second and third Q-bank feels productive but usually duplicates items, fragments your analytics and destroys your ability to read a clean unseen score. The two-Q-bank rule is simple: never add a second bank unless it does a job the first cannot, and keep one bank untouched as your unseen measurement instrument.
Segment yourself first: six candidate profiles
Before you spend anything, decide which profile describes you. The defining signal is objective, not a feeling.
| Profile | Defining signal | Branch to follow |
|---|---|---|
| First attempt, adequate runway | Never sat Paper B, eight or more weeks free | Build the two-slot stack; start with coverage, finish on unseen mocks |
| Retake | Failed a previous diet | Diagnose why (coverage vs pacing vs critical review) before buying anything new |
| Busy trainee revising around clinical work | Under six hours a week, on-calls and nights | Protect a small daily retrieval habit; prioritise high-yield domains and critical review |
| Weak foundations | Shaky on core science, psychopharmacology or statistics | Add one teaching source before high-volume questions |
| Strong knowledge, poor pacing | Good untimed, slow or panicked under the clock | Shift almost entirely to timed mixed blocks; stop learning content |
| Strong MCQ, weak quantitative | Good clinical MCQs, poor on calculations, data interpretation and critical review | Add a dedicated critical-review and statistics tool; drill the quantitative subset |
A word on that last profile. In Paper B, the "practical" difficulty is quantitative — reading a forest plot, computing a number needed to treat, judging a confidence interval under time pressure — not a clinical-skills problem. Clinical practical skills belong to the CASC and should not distort your Paper B resource choice.
Budget bands: free, one subscription, or a full stack
Verify every price on the day you buy; the figures below were vendor-reported on 20 July 2026 and change often.
- Free or low-cost. The Royal College's own syllabus, sample questions and marking scheme are free, and iatroX offers free UK-core practice you can use as an unseen measurement layer. This is enough to calibrate to the real format and to expose your gaps, but not enough volume to carry the clinical block on its own.
- One premium resource. A single paid Q-bank subscription. For illustration, PassMRCPsych's Paper B tier was vendor-reported at £99 for three months or £149 for six months, with a separate statistics course at £49 for six months; iatroX's Paper B bank was listed from £29 a month or £99 a year. Choose the one whose strongest job matches your weakest block.
- Comprehensive stack. One teaching-led resource, one high-volume bank, a dedicated critical-review course and an unseen-measurement bank. This is worth it for retakes or candidates with real gaps in more than one block; for most first-timers it is money spent to feel busy.
Time bands: what to build, and what to leave out
The commonest planning error is trying to run a full stack in too little time. So state explicitly what you will omit.
| Weeks to exam | Build this | Omit this |
|---|---|---|
| Under four | Daily timed mixed blocks from your primary bank; a focused critical-review drill; two full mocks under exam conditions; official sample questions saved for last | New content courses; a second Q-bank; exhaustive note-making |
| Four to twelve | One coverage pass of weak domains, then a transition to mixed timed blocks; a weekly critical-review session; mocks in the final fortnight | A third resource; re-reading explanations you already know |
| More than twelve | Add one teaching source if foundations are weak; build coverage domain by domain; keep a protected unseen pool for the final month | Cramming statistics at the very end; leaving mocks until the last week |
The decision matrix: one job per resource
The skill is assigning each resource a single job, not collecting subscriptions. No product below does every job well.
| Resource | Best job | Format fit for Paper B | Verify on 20 July 2026 |
|---|---|---|---|
| RCPsych syllabus, sample questions and marking scheme | Official calibration and standard-setting | Definitive for format and blueprint | Free; current sample set on rcpsych.ac.uk |
| SPMM Course | Structured teaching and high-yield notes | Strong on clinical block and critical-review theory | Current price and inclusions on spmmcourse.com |
| MRCPsychMentor | Question volume and breadth | Good clinical-block drilling | Current count and price on mrcpsychmentor.com |
| PassMRCPsych | Audio teaching, Q-bank and a statistics course | Covers clinical block and critical review; full mocks | £99/3mo, £149/6mo Paper B; stats £49/6mo (vendor-reported) |
| BMJ OnExamination | Question volume and explanations | Clinical-block drilling | Current count and price on onexamination.com |
| A dedicated critical-appraisal/statistics course | The critical-review third | Targets the highest-leverage weak spot | Verify current provider and price |
| iatroX | Unseen, blueprint-stratified measurement and a Socratic tutor | Mixed SBA and EMI, blueprint-aligned | From £29/mo or £99/yr (vendor page) |
Match the job to your weakest block. A candidate strong on clinical knowledge but weak on statistics should not buy a second clinical bank; they should buy the critical-review tool and reserve an unseen bank to measure whether it worked.
The cannibalisation guardrail: a hub, not a duplicate
This article is the hub for one question — "which Paper B resource?" — and it deliberately does not reproduce the granular evidence on each platform. The platform-specific workflow guides in this series (for example, the PassMRCPsych Paper B workflow) carry the detailed walk-throughs, and the iatroX comparison hub holds side-by-side product facts. Read this to choose a branch; read the child article to run the chosen tool. If you find yourself reading three platform reviews, you have already broken the minimum-stack rule.
Three worked candidate profiles
Profile A — first attempt, ten weeks, moderate budget. One premium Q-bank plus the free official material. Weeks one to four: coverage passes of the four weakest domains, three sessions a week, plus one weekly critical-review block. Weeks five to eight: switch to mixed timed blocks; keep an unseen pool untouched. Weeks nine and ten: two full mocks under exam conditions, then official sample questions to calibrate. Exit criteria: every domain above your personal floor on unseen items, pace under roughly seventy seconds a question, and a critical-review subscore no longer dragging the total.
Profile B — retake, six weeks, failed on critical review. Do not buy another clinical bank. Add one dedicated statistics and critical-appraisal course and drill it daily; use your existing bank only for mixed timed blocks. Weeks one to four: critical-review course plus three mixed blocks a week. Weeks five and six: two mocks, then targeted revision of the specific statistical concepts you still miss. Exit criteria: critical-review accuracy on unseen items pulled up to match your clinical accuracy.
Profile C — busy CT2 revising around nights, eight weeks. Protect a twenty-minute daily retrieval habit rather than chasing hour-long sessions you cannot sustain. Use one bank on your phone for short mixed blocks between clinical duties, plus one weekly ninety-minute critical-review session on a day off. Omit teaching courses entirely. Exit criteria: consistent first-attempt accuracy across your commonest domains and at least one full timed mock completed without a caffeine crash.
How to weigh the evidence when resources disagree
When two resources contradict each other, use a hierarchy. Official material wins on format and standard — only the Royal College defines what Paper B is. Primary clinical guidance wins on content: NICE, CKS, SIGN, the Maudsley Prescribing Guidelines and the SmPC/eMC for medicines. Vendor pages are the source for product facts — question counts, prices, features — but only when labelled vendor-reported and dated, because they change without notice. Independent user testing (cohort feedback, trainee forums) is useful for user experience only, not for clinical truth. When a Q-bank explanation conflicts with current guidance, the guidance wins and the item is flagged, not memorised.
Reading your results without fooling yourself
Your overall Q-bank percentage is the least useful number you own, because it blends easy seen items with hard unseen ones and hides the domain where you will actually lose the exam. Watch four things instead: domain-level accuracy on unseen items, first-attempt accuracy (not post-review), your pace per question, and your critical-review subscore in isolation. Three mistakes this decision tree is designed to stop: chasing completion percentage as if finishing a bank equalled competence; hoarding banks so no single score is trustworthy; and letting the comfortable clinical block crowd out the critical-review third that decides borderline candidates.
Frequently asked questions
How do I know whether I have covered the full MRCPsych Paper B blueprint? Completion is not coverage. Map the Royal College syllabus into a simple matrix — every clinical domain plus each critical-review topic as its own row — and mark each row only when you can score reliably on unseen questions in it, not when you have merely read the notes. Gaps hide in the rows you keep skipping because they feel uncomfortable, which for most candidates means statistics, forensic and old-age psychiatry. You have covered the blueprint when no row is blank and none sits below your personal accuracy floor on fresh items.
Can one question bank be enough for MRCPsych Paper B? For a first-time candidate with solid foundations, yes — one well-matched primary bank plus the free official material can carry the clinical block, provided you also reserve an unseen pool to measure yourself and give the critical-review third dedicated attention. One bank stops being enough when it cannot do a job you need: if your chosen bank is thin on statistics, add a critical-review tool rather than a second clinical bank, and keep one bank untouched as your measurement instrument.
What should I measure instead of my overall Q-bank percentage for MRCPsych Paper B? Measure domain-level accuracy on unseen, timed questions; your first-attempt accuracy before you review the explanation; your pace against roughly seventy seconds a question; and your critical-review subscore separately from your clinical subscore. A trainee sitting at 74% overall but 55% on critical review is not a 74% candidate — they are a borderline one hiding behind an average. Your overall percentage is not your exam score, and treating it as one is the single most common calibration error.
When should I stop doing new MRCPsych Paper B questions? Stop opening new topics once every blueprint row is above your floor on unseen items and your pace is stable; at that point new volume adds little and revisiting your own errors adds a great deal. In the final two to three weeks, switch from acquisition to consolidation: mixed timed mocks, targeted review of persistent error types, and official sample questions for calibration. Doing new questions on exam eve tends to shake confidence without changing competence.
Which MRCPsych Paper B resource should I use for my weakest component? Match the tool to the component. For a weak clinical block, a high-volume bank (MRCPsychMentor, BMJ OnExamination or PassMRCPsych) drills breadth. For weak critical review — the highest-leverage gap for most candidates — a dedicated statistics and critical-appraisal course beats more clinical questions. For weak pacing, no new content helps; you need timed mixed blocks and mocks. For blind-spot detection across the whole blueprint, an unseen, blueprint-stratified bank such as iatroX shows you where you actually fail rather than where you feel weak. Buy for the component, not for the brand.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 20 July 2026. Product figures (question counts, prices, access periods and features) are vendor-reported and change frequently; verify each on the provider's own page before you rely on it. Disclosure: iatroX operates a competing MRCPsych Paper B question bank; its role in this article is confined to the jobs a decision hub can honestly do — unseen, blueprint-stratified measurement and blind-spot detection — not to a claim of being the single best resource, which this article deliberately refuses to make. Corrections are welcome via the feedback route on iatrox.com.
References: Royal College of Psychiatrists — preparing for exams and Papers A and B marking scheme (rcpsych.ac.uk); SPMM Course (spmmcourse.com); MRCPsychMentor (mrcpsychmentor.com); PassMRCPsych (passmrcpsych.com); BMJ OnExamination (onexamination.com); iatroX MRCPsych Paper B bank (iatrox.com/mrcpsych-paper-b); the iatroX comparison hub (iatrox.com/compare); the two-Q-bank rule and blueprint-coverage matrix framework pillars, and "Your Q-Bank Percentage Is Not Your Exam Score" (iatrox.com/blog).