Level Threshold Guide

Provenance: written 2026-09-29. Primary sources: research/veridian/WME01-verified-facts.md §3 (mark-scheme conventions) and §4/§9 (the Oct 2022 Q3 and Jan 2023 Q4 audit entries, each independently re-verified against a primary Pearson PDF by a prior session — provenance re-checked here, not re-fetched); content/veridian/technique-hub/wme01/forge-exam-technique-guide.md §1–§2 (the fresh January/June 2025 mark-scheme read, including the Q2(a) graph-mark breakdown quoted in §2.2 below). Before writing, this pass read in full every other document already in this folder — codex-formula-and-method-bank.md, all nine codex-deep-dive-*.md files, all eight signal-topic-brief-*.md files, signal-frequency-and-boundaries.md, forge-battle-card.md, forge-error-pattern-tracker.md, forge-vocabulary-precision-list.md, forge-exam-day-protocol-extended.md, forge-forty-eight-hour-checklist.md, forge-mock-paper-protocol.md — plus every lesson in content/veridian/curriculum/wme01/ and content/veridian/question-bank/wme01.ts, specifically to find what threshold material this large, already-crowded hub had not yet built, rather than restate any of it. Not affiliated with or endorsed by Pearson Edexcel.

#WME01 — Level/Threshold Guide

#Pearson Edexcel IAL Mechanics 1 — mark-band boundaries, by real example

#0. What this document is, and an honest note on what it found

A Mechanics paper has no essay-style "Level 3 vs Level 4" descriptor ladder — every WME01 mark is a discrete M, A, B or DM/DB point, awarded or withheld one line at a time. This course's own WMA11 sibling document (content/veridian/technique-hub/wma11/forge-level-threshold-guide.md) already states this for Pure Maths, and it holds just as fully here: a genuine "threshold guide" for this paper has to work at the same grain — for a specific real question, exactly what, on the page, moves a script from one mark to the next, and what a real, examiner-confirmed response looks like on each side of that line.

The honest finding this pass has to report before anything else: the WME01 technique hub is unusually well-developed already. Nine CODEX deep-dives and eight SIGNAL topic briefs, most written earlier the same day this document was, have between them already built full mark-by-mark ladders — several already using "just misses / just clears" language — for: Jan 2023 Q4's two-tilting-condition moments question (moments-rods-on-two-pivots.ts, codex-deep-dive-moments-rods-on-two-pivots.md); Jan 2022 Q1's equilibrium question, including its real M1/A1/A1 split and the DM1 elimination step (equilibrium-of-a-particle-coplanar-forces.ts, codex-deep-dive-equilibrium-of-a-particle-coplanar-forces.md); Jun 2024 Q2's resultant-of-two-forces question, including its second, independent structural M-mark (resultant-forces-and-bearings.ts); Jun 2024 Q3 and Jan 2022 Q7's "must be in terms of mg" DM1/A1 pair (forge-vocabulary-precision-list.md, forge-error-pattern-tracker.md, connected-particles-pulleys-and-lifts.ts); and more besides — §4 below maps the rest. Per this course's standing rule against padding, none of that is rebuilt here.

What was left, once all of that was set aside, is smaller but genuinely real: one general rule this hub states but has never illustrated by contrast (§1); one rich, multi-mark real question whose specific mark-void condition has not been laddered anywhere in this hub (§2.1); and one real graph question whose own three independent B-marks have been counted but never worked through as a threshold in their own right (§2.2). §3 names what was checked and found genuinely thin, rather than invented to fill the gap.

---

#1. A rule this hub states but has never illustrated by contrast: two different floors

forge-battle-card.md already states the misread rule in one line: a misread that doesn't change the question's character costs "two marks off the A/B marks in that part, not the whole part." forge-exam-technique-guide.md already states, separately, that omitting a mass from a resolution or a length from a moments equation costs the method mark itself. Read side by side, against the real, verbatim mark-scheme wording each is drawn from, they describe two structurally different kinds of floor:

• A genuine misread — copying a number wrong from the question, without changing what the question is actually asking — is capped. The general marking guidance's own words: "deduct two from any A or B marks gained," in the part affected. Every M mark survives; every A/B mark above that two-mark deduction survives too.
• A structurally broken equation — one missing a mass, a length, or (per the record's own live example) an entire acceleration term — has no such cap. Because A marks cannot stand without their own M mark ("M0 A1 is impossible," per the general M/A/B convention already quoted in forge-exam-technique-guide.md §2), a void M mark voids everything chained to it: the A marks that depend on it, and any DM mark that depends on those. January 2025's own examiner report records a live instance already quoted in this hub: a connected-particles equation with every term a genuine, correctly-signed force, but no ma anywhere in it — an equation that (if solved) describes a system in equilibrium, not one accelerating, and scores nothing for the method it was supposed to demonstrate, however tidy the algebra that follows looks.

The practical threshold this produces, not stated as its own contrast anywhere else in this hub: two answers that are both "wrong" can lose very different amounts, and which floor applies depends on why the number is wrong, not how far downstream the consequences run. A script that misreads a given value but reasons validly from it is protected at minus two in that part. A script that reasons from a structurally incomplete equation is protected by nothing — it can lose the method mark, the accuracy marks chained to it, and any dependent mark built on top, from one missing term.

Honestly flagged, not papered over: unlike the structural-error side, which the record illustrates with a real, quoted equation, no concrete WME01 script survives in the material reviewed for this course that shows the two-mark misread deduction actually being applied to real numbers. The rule itself is real and verbatim, repeated identically across three independently-checked mark schemes (facts bank §3); a live WME01 illustration of it costing exactly two marks is not yet part of this course's verified material. This is the same gap the WMA11 sibling document found for its own paper's equivalent rule — not a coincidence of one paper, but, on this evidence, a genuine hole in what this course has mined from either paper's misread lines so far.

---

#2. Two threshold ladders not yet built anywhere else in this hub

#2.1 Oct 2022 Q3(b) — the M0 floor for carrying a value from the wrong part into this one

The question (Paper WME01/01, Publication Code WME01_01_2210_MS, re-verified directly against wme01-01-rms-20230112.pdf and the question paper wme01-01-que-20221018.pdf by a prior session this pass re-checked the provenance of): a particle on a plane inclined at angle α to the horizontal (tan α = 3/4, so sin α = 3/5, cos α = 4/5), mass 2 kg, coefficient of friction μ = 0.5, held by a horizontal force of magnitude X. Part (a), 3 marks: "Show that when X=14.7 there is no frictional force acting on P." Part (b), 8 marks, moves to a different, unstated value of X — the smallest one for which the particle stays in equilibrium, with friction now genuinely at its limiting value.

The real ladder for part (b), verbatim mark-scheme structure (S is this mark scheme's own label for the normal reaction — the same quantity this course's friction lesson elsewhere calls R):

Eight marks, all accounted for. Solving the two equations together (independently checked, by hand, against the mark scheme's own stated answer): X cos α + 0.5(X sin α + 2g cos α) = 2g sin α → 0.8X + 0.3X + 7.84 = 11.76 → X = 3.92/1.1 = 4g/11 ≈ 3.56 N. The arithmetic matches the source's own quoted answer exactly, confirming the reconstruction above of what X physically is (a horizontal push into the slope) and how the two equations combine.

The floor. The mark scheme carries an explicit note the other two equations don't need: "M0 for both equations if they put X = 14.7 anywhere." This is not an arbitrary penalty — carrying part (a)'s own special value into part (b)'s equations doesn't just give a wrong number, it makes the two equations mutually contradictory. Substitute X = 14.7 into the along-plane equation: 14.7(0.8) + F1 = 11.76 → F1 = 0 (exactly part (a)'s own finding — unsurprising, since 14.7 was defined as the value that makes friction zero). But F1 = 0.5S then forces S = 0, while the perpendicular equation, fed the same X = 14.7, gives S = 14.7(0.6) + 2(9.8)(0.8) = 24.5 — a direct contradiction between what the two equations say S is. The mark scheme's M0 note is catching a script that has, in effect, tried to answer part (b) using an equilibrium condition that only holds in part (a)'s own, different scenario.

Just misses vs just clears, concretely:

• Just misses (B1 only, 1/8): a script that writes "X cos α + F1 = 2g sin α" and "S = X sin α + 2g cos α" but substitutes X = 14.7 into one or both before solving — perhaps out of habit, having just found that value in part (a), or from not registering that part (b) asks about a different X. Both M marks are void by the scheme's own explicit instruction; since A marks cannot stand without their M, the two A2 pairs and the final A1 are all unreachable regardless of how the remaining arithmetic is handled. Only the B1 for F1 = 0.5S survives, because it needs no reference to X at all.
• Just clears (B1 M1 A2 M1 A2 A1, 8/8): the same two equations, with X left as the unknown throughout, solved simultaneously to X = 4g/11.

A family resemblance worth naming, not rebuilding: this is the same shape of error as two traps already documented elsewhere in this hub — a value true in one part or one scenario, carried into a part where the physical picture has actually changed (the "stale force kept after a support loses contact," Jan 2025 Q4, in codex-deep-dive-moments-rods-on-two-pivots.md; a given tension substituted where a different, to-be-found value belongs, Jun 2024 Q3, in codex-deep-dive-connected-particles-pulleys-and-lifts.md). Those two documents already own their own worked treatments; this entry adds a third, independently-confirmed real instance of the same general failure mode — reusing a number instead of re-deriving it once the question has moved on — rather than a new mechanism.

#2.2 Jan 2025 Q2(a) (corroborated Jun 2025 Q3(c)) — three independent B-mark floors on one graph

forge-exam-technique-guide.md §2 and forge-error-pattern-tracker.md both already name the "continuous line" trap on this question. Neither works through what the three marks actually are, side by side, as a threshold in its own right — and codex-deep-dive-kinematics-constant-acceleration-two-stage-motion.md explicitly flags that this exact question's own qualitative content was never deep-mined beyond confirming it carries real marks. This closes that specific gap, at the grain the marks themselves are awarded.

The real ladder (January 2025 Q2(a), 3 marks, all B — independent of each other and of any M mark, since B marks need no method):

The floor that matters most here is the shape mark's own, and it is unusually concrete: the mark scheme states the shape "must start on the vertical axis and not at the origin," and its own B0 condition fires on "any extra continuous lines." The examiner report confirms this in practice — a solid line added where the real motion should show a distinct corner cost candidates exactly this mark.

Just misses vs just clears, on the shape mark specifically:

• Just misses (this B1 lost, whatever the other two do): a graph drawn as one unbroken sloped line where the real motion has a genuine corner between phases (e.g. acceleration meeting deceleration without a change of gradient shown), or a graph that starts at the time-origin on the speed axis when the object's own initial speed is not zero. Because this is a B mark, there is no downstream method step that can rescue it — it is either drawn correctly or it isn't, independently of every other mark in the question.
• Just clears: the same underlying journey, drawn with a genuine corner at each phase change and starting on the vertical axis at the correct nonzero speed.

Independently corroborated, same family, a second series: June 2025 Q3(c) awards a B1 for "a longer 2nd line and the vertex on the t-axis," with its own explicit B0 for a continuous vertical line drawn at the wrong point, plus a separate follow-through B1 for the graph's three numerical values. Two consecutive series building the identical shape-mark floor into their own mark schemes is real, named corroboration — not an estimate.

Honestly narrower than it might look: the shape mark is the one this pass can illustrate concretely, because the mark scheme's own B0 condition is stated in exactly those terms. The other two marks in the same 3-mark line (time-axis labels, speed-axis labels) are real and independently awarded — a script can lose the shape mark while still banking both label marks, since none of the three depends on either of the others — but the material reviewed for this pass does not give a comparably specific wrong-script/right-script pair for either label mark beyond the general "total time mistaken for one stage's own duration" trap already named in forge-exam-technique-guide.md and forge-error-pattern-tracker.md. That trap is not rebuilt here as a fourth, thinner ladder.

---

#3. Genuinely checked, genuinely thin — flagged rather than filled

• The misread floor (§1) is real, verbatim, and repeated across three independently-checked mark schemes — but no concrete WME01 script illustrating the actual two-mark deduction survives in the material this course has reviewed. Stated as a real threshold with no live example, not invented one.
• cso ("correct solution only") is a real, verbatim WME01 marking convention (facts bank §3), and one concrete cso-tagged WME01 mark is on record — Oct 2019 Q2(b), h = 0.625 m (noted in this hub's own modelling-lesson audit trail) — but no instance has been recovered anywhere in this course's WME01 material of a script that reached the right final number through a working error that happened to cancel out, and lost the cso mark anyway. The tag is confirmed; a live example of it actually biting is not.
• Jan 2025 Q2(a)'s own two label marks (§2.2 above): real, independently awarded, but not laddered here with their own concrete wrong-script pair, for the reason stated there — the material reviewed doesn't yet support one at the same grain as the shape mark's.

---

#4. Where the rest of this paper's threshold material already lives

Checked directly against every document in this folder before writing anything above:

This document ends here rather than re-describing any row in that table in its own words. What it adds — §1's misread-vs-structural-error contrast, §2.1's Oct 2022 Q3(b) M0 floor, and §2.2's Jan 2025 Q2(a) shape-mark floor — is the genuine remainder once everything already covered in this now-substantial hub is set aside.