Exam Day Protocol Extended

Provenance: compiled 2026-09-29. Primary sourcing is research/veridian/WMA11-verified-facts.md (provenance-checked against real Pearson PDFs — see that file's own §5 for its documented WebFetch-hallucination catch) and this same folder's forge-exam-technique-guide.md (independently re-extracted the General Marking Guidance and the real January 2023/January 2024 question-paper advice text, pdftotext -layout, from the local archive). This document adds no new primary-source extraction of its own; every citation below points to a fact already verified in one of those two files, or is an arithmetic derivation of numbers already verified there (flagged as such where it applies). Not affiliated with or endorsed by Pearson Edexcel. Anything below that is general test-taking practice rather than a WMA11-specific finding is marked as such, not presented as sourced.

#WMA11 — Exam-Day Protocol (Extended)

#Pearson Edexcel IAL Mathematics, Unit 1: Pure Mathematics P1

#0. What this document is, and what it isn't

forge-exam-technique-guide.md §1 already gives the real timing arithmetic (90 minutes ÷ 75 marks = 72 seconds/mark, worked against two real series) and its own §5 already gives a compact exam-day protocol built directly from the paper's own printed advice. This document does not repeat either. It exists to go one layer deeper on exactly four things the brief for this document asked for and the existing guide does not cover: how much of the 90 minutes is actually reading time (§1), what a question-order strategy grounded in real paper mechanics looks like rather than folk wisdom (§2), a worked elapsed-clock-time checkpoint table rather than a per-question minute budget (§3), and a concrete sequence of moves for the specific situation of finding yourself behind schedule partway through (§4). Where the real material runs out, each section says so and stops rather than filling the gap with invented specifics.

---

#1. Reading time: what is and isn't actually granted

There is no separate reading-only period on this paper. The spec's own assessment-information table gives the unit's duration as a flat 1 hour 30 minutes, with no reading-time allowance listed alongside it (Spec, P1.2, content-overview table, p.7 — WMA11-verified-facts.md §2). The real January 2023 front-cover instructions, independently re-extracted for this technique hub, carry the same shape: they tell a candidate to "read each question carefully before you start to answer it" — advice about how to read once the 90 minutes is already running, not a description of extra time set aside before it starts (forge-exam-technique-guide.md §1, pdftotext -layout extraction of the real January 2023 question paper).

That is a real, checkable absence in the two series' front matter this course has actually transcribed — but only those two. Nine series appear in the underlying facts bank, and most of their own front-cover pages were never individually re-typed out; the honest claim is "no reading time in the documents checked," not "confirmed absent from every WMA11 sitting ever run" (see §5).

The practical consequence: "using reading time well" on this paper means using the opening minutes of the 90, not a bonus window. The only sourced instruction for that window is the advice box itself — read carefully before answering, attempt every question, check at the end if time allows (forge-exam-technique-guide.md §1, §5). One further move is worth naming explicitly as an extension of that advice rather than a restatement of it, and it is flagged here as general exam practice, not a WMA11-specific citation: doing a single fast pass across the whole paper before committing to Q1 lets you note each question's printed mark value in one pass (which you need anyway for the 72-second-per-mark arithmetic — forge-exam-technique-guide.md §1) and flag any printed calculator-technology-override phrase on sight, rather than discovering each one cold as you reach it. The override phrase itself is real and recurring — confirmed worded differently series to series, including "solutions relying on calculator technology are not acceptable" on multiple sub-parts across two independent series (WMA11-verified-facts.md §2) — but the "map it in one upfront pass" technique is this document's own suggestion, not something any examiner report recommends.

---

#2. Question-order strategy

No examiner report and no front-cover instruction anywhere in the facts bank recommends a specific order for attempting questions. The only printed instructions that bear on order at all are that candidates must answer all questions — a rubric requirement printed on the real January 2023 paper, not just encouragement (WMA11-verified-facts.md §2) — and the advice to read each one carefully before starting it. Order itself is left to the candidate. What follows is what the real paper's own mechanics support, not an invented ranking system.

Mark value on its own is not a difficulty signal, so "biggest questions first" or "smallest questions first" doesn't hold up against real data. Two real, verified 4-mark questions from the same series sit at opposite ends of the difficulty spectrum: January 2023 Q4 (discriminant with a parameter) sits alongside January 2023 Q9 (state the period of tan x), and the latter is the single worst-performing question this facts bank has found real statistics for — more than half of all candidates scored zero, and roughly 10% didn't attempt it at all (examiner report, Publications Code WMA11_01_ER_2301 — the full-marks/zero-marks sentence is verified verbatim in WMA11-verified-facts.md §4.3; the non-attempt figure is a fuller re-extraction of the same examiner-report paragraph, verified verbatim in that same file's trig-graphs-period-and-domain.ts lesson-audit-log entry, which states explicitly that §4.3 "previously captured only the two sentences already quoted there"). A question's printed mark tells you its time budget; it tells you nothing about how it will actually go, so an ordering strategy built purely on the numbers in brackets is building on the wrong variable.

A wrong or incomplete early part is not, on its own, a reason to abandon the rest of that question. The mark scheme's follow-through (ft) convention — confirmed verbatim across two independent series' general marking guidance — lets many A and B marks in later parts credit "correct ft" working from a candidate's own earlier value, right or wrong (WMA11-verified-facts.md §3). A candidate who fumbles part (a) of a four-part question has not necessarily forfeited parts (b)–(d); those parts are frequently gated on "their" value from (a), not the true value, and this is a real, checkable mark-scheme mechanic, not reassurance. (M marks never carry an ft tag — they credit a correct method on whatever the candidate is actually working with, so this point is specifically about the A/B marks that follow.)

A stuck low-mark question is a worse trap than a stuck high-mark one, and this paper has a real, documented example of exactly that shape. January 2023 Q9 — 4 marks, over half the cohort scoring zero, roughly 1 in 10 not even attempting it — shows what a bad time trade looks like concretely: a small ceiling on marks available, paired with a genuinely high failure rate, is an easy place to sink disproportionate minutes for little return. Contrast that with June 2022 Q5(a) (reconstructing a quadratic from its turning point, WMA11_01_2206_MS), where "1 out of 4 was a very common score" — even a question most candidates didn't fully solve still paid out real partial credit for the opening method step (WMA11-verified-facts.md §4.1). Read together, the two examples argue for the same behaviour from opposite directions: attempt the opening move of a question before judging whether it's worth more time, because partial credit for a correct first step is common on this paper even on questions that go on to be hard — but once you've made that attempt and are still stuck, the 4-mark ceiling on a question like Q9 means the cost of walking away is genuinely small.

The resulting, only-partly-sourced heuristic: attempt in printed order by default — this matches the paper's own linear "read carefully, then answer" framing rather than a scan-and-cherry-pick one — and treat your own §1 skim as the trigger for the one legitimate exception: if a specific question is visibly outside what you've prepared, it is ordinary, sound test-taking practice (flagged here as general practice, not a WMA11 citation) to note it and come back to it, rather than losing minutes to it in its printed position while questions you can actually solve sit unattempted later on the paper.

---

#3. Timing checkpoints

forge-exam-technique-guide.md §1 already gives a real, per-question minute budget for two series (January 2023 and January 2024), each summing to exactly 90 minutes by construction. What it does not give is a running clock-time figure to check yourself against mid-paper — that is what this section adds, and it is an arithmetic derivation of those same already-verified numbers, not a new primary-source finding. Pearson does not publish an elapsed-time checkpoint table; the two tables below apply the paper's own printed pacing rule ("the marks for each question are shown in brackets — use this as a guide as to how much time to spend on each question," forge-exam-technique-guide.md §1) cumulatively, question by question, so a candidate can glance at a clock after finishing a question and see whether they are ahead or behind, rather than only knowing the budget for the single question just finished.

January 2023 (11 questions, marks 5,5,5,4,6,10,10,8,4,10,8 = 75):

January 2024 (10 questions, marks 5,5,7,6,10,6,9,9,10,8 = 75):

How to use this honestly. The question count on this paper isn't fixed — it has been 10 or 11 in every series checked (WMA11-verified-facts.md §2) — and the mark spread across questions differs series to series, so neither table above is a schedule to memorise for your own sitting. The method is: after each question, add its printed mark to your running total, multiply the running total by 72 seconds, and that's the clock reading you should be at or ahead of. Both tables are two real worked examples of that one method, included so the arithmetic is checkable, not because either paper's specific shape will recur. The one series-independent anchor that does hold regardless of question count is the halfway point: 45 minutes is exactly half of the fixed 90-minute duration, so it is a reasonable single glance-at-the-clock moment on any real sitting — this specific anchor is simple arithmetic on the fixed duration, not something drawn from an examiner report, and is flagged as such.

---

#4. If you are running behind

This section assumes a checkpoint from §3 (or the equivalent running-total arithmetic on your own paper) has told you that you are behind. What follows are the concrete moves that are actually grounded in this paper's real mark-scheme mechanics, in the order they matter.

1. Secure the method mark on the question you're currently doing before you move on. The general marking guidance states directly that where a method uses a formula that has to be recalled, "the formula should be quoted first" — the method mark is then earned for quoting a correct formula and attempting to use it, even with small substitution errors, whereas skipping straight to numbers risks losing that mark to a slip with no formula on the page to fall back on (WMA11-verified-facts.md §3). Under time pressure, writing the formula line first and substituting quickly — even messily — is the single cheapest way to bank a mark you might otherwise lose entirely by rushing straight to arithmetic.

2. On a three-term quadratic, pick one of the three credited routes and commit — don't switch methods mid-question. The real "General Principles for Pure Mathematics Marking" name exactly three routes to the method mark for solving a 3TQ: factorisation, the quadratic formula, or completing the square (WMA11-verified-facts.md §3). Starting one, stalling, and restarting with a different route under time pressure costs you working time twice for a mark you could have earned once; picking whichever route you're fastest at and finishing it is the better trade when the clock is the constraint.

3. On a two-stage integration question, fix each constant immediately rather than deferring both to the end. This is a real, examiner-documented failure mode on this exact question type — January 2023 Q11(b)'s report states plainly that it was "very common to see candidates integrate twice and apply a constant only after their second integration, thus missing earlier marks" (WMA11-verified-facts.md §4.5; full worked treatment already lives in integrate-twice-two-constants.ts, not reproduced here). The time-pressure reading of that finding is specific: deferring both constants means that if you run out of time before completing the second integration, you lose the marks for both constants, including the one you had the information to fix ten minutes earlier. Fixing each constant the moment its condition is given protects that mark independently of whether you ever reach the second one.

4. Triage a stuck question by its printed mark value, not by how much effort you've already sunk into it. §2 already names the real shape this takes on this paper: a low-mark question with a documented high failure rate (January 2023 Q9 — 4 marks, over half the cohort scoring zero) is a small amount to write off compared to walking away from a 10-mark question you're merely stuck on rather than lost on. If you're behind and stuck, check the mark in brackets before deciding whether to keep going.

5. With only a few minutes left, spread a fast partial attempt across every unanswered question rather than perfecting one. Answering all questions is a rubric requirement on this paper, not optional advice (WMA11-verified-facts.md §2), and the printed advice separately urges the same thing (forge-exam-technique-guide.md §1, §5). A blank answer scores zero with certainty; a quoted formula and one line of substitution on an unattempted question has a real chance of the method mark described in point 1 above. In the last few minutes, that trade — one scored line on each of several untouched questions, rather than a completed solution to one of them — is the arithmetic the rubric itself points toward.

6. Checking is the correct thing to sacrifice first, not a question attempt. The paper's own advice places checking last and explicitly conditional — "check your answers if you have time at the end" (forge-exam-technique-guide.md §1, §5) — which means the paper's own printed priority order already puts an attempt at every question ahead of reviewing any of them. If you are behind at the checkpoint that matters (see §3's halfway anchor in particular), the time to give up first is time you had planned to spend checking finished work, not time for a question you haven't touched yet.

---

#5. Flagged gaps and what remains unverified

• §1's "no reading time" claim is confirmed only in the front-cover and general-instruction text this course has actually transcribed — the real January 2023 question paper and the general marking guidance shared across the January 2023/January 2024 mark schemes. It has not been independently re-checked against the front cover of every other series in the nine-series facts bank; treat it as "not present in the documents checked," not as a claim about every WMA11 sitting since 2019.
• No examiner report or front-cover instruction recommends a specific question order. Every ordering claim in §2 beyond the two named, cited findings (the ft mechanic, and the Q9/Q5(a) contrast) is general test-taking practice, explicitly flagged as such at the point it's made — not a WMA11-specific citation, and it should not be read as one.
• §3's checkpoint tables are this document's own arithmetic derivation of numbers already verified elsewhere (the real per-question mark totals for January 2023 and January 2024, cited in forge-exam-technique-guide.md §1 to WMA11_01_MS_2301 and WMA11_01_2401_MS). Pearson has not published, and this document does not claim Pearson has published, any elapsed-time checkpoint table.
• No grade-boundary figure appears anywhere in this document, consistent with the rest of this technique hub. The one figure that exists in the underlying facts bank (an October 2022 boundary set) is explicitly marked UNVERIFIED there and is not repeated here — see signal-frequency-and-boundaries.md §2 for the full reasoning.
• The Q9 and Q5(a) statistics used in §2 and §4 are each drawn from a single series (January 2023 and June 2022 respectively) and are cited as real, documented examples of a pattern, not as claims about how often that exact pattern recurs across every sitting.