Filter papersTierBoardsCalchover to highlight the panels each one changes
Deep-dive topic↳ only changes the Topic deep-dive tab
ⓘ How to read this — sources & caveats
Built from 246 real OCR / AQA / Edexcel GCSE (9–1) Maths papers (June & November 2017–2024), every question-part tagged to a topic and cross-checked against the mark scheme. Marks on multi-topic (synoptic) questions are split across their topics. Spec strands are assigned by a topic classifier. The overall Assessment Objective mix is mandated identical for every board (Ofqual: F 50/25/25 · H 40/30/30) and the estimate is calibrated to match it (verified against AQA's official AO grid). The per-topic "How it's assessed" split is calibrated from AQA's official per-question AO grid by strand & question size and applied across boards (AO follows question type) — a model, not official per-paper tags. "Longest gap since last seen" reflects past frequency, not a prediction. Trends from few sittings (n<6) are greyed as weak evidence.
🔓
Unlock the full dashboard
The Topic deep-dive and Papers & search tabs, board comparisons, trends over time and CSV export are free for registered teachers.
Free, takes a few seconds. The Overview tab stays open to everyone.
Signed in as you. We'll email occasional teacher updates (new analysis, free resources) — unsubscribe anytime. No spam.
Marks by spec strand
Total marks per strand across the filtered papers.
Assessment Objective mix estimated
AO1 (routine) / AO2 (reason) / AO3 (problem-solve) split by marks. The overall AO balance is mandated identical for every board (Ofqual: F 50/25/25 · H 40/30/30) and this estimate is calibrated to match it (verified against AQA's official published AO grid, which lands exactly on the mandate). For how a specific topic is assessed, see the per-topic "How it's assessed" card, which is AQA-calibrated by question type.
Hardest topics that come up a lot real data
High-frequency topics where students actually score low in real papers — the best return on teaching time. Ranked by marks-per-paper × difficulty (from Question Level Analysis: AQA + Edexcel). Click a topic to deep-dive.
The boards phrase it differently real data
We counted every command word across 244 real GCSE papers held in full text. Data no one else publishes.
AQA says "work out" 967 times and "circle your answer" 557, but "explain" 0 times across its papers. Edexcel and OCR use "explain" freely. Knowing which words a board favours is worth marks.
Command words, counted from the real wording real data
Every command word counted across the corpus. The same skill wears different words on different boards: knowing that is worth marks.
Command
Total
OCR
AQA
Edexcel
Work out
2223
499
967
757
Find
882
502
2
378
Write down
670
259
139
272
Circle / tick (MCQ)
582
25
557
0
Complete
577
249
194
134
Show that
394
177
114
103
Estimate
345
101
112
132
Draw
326
108
89
129
Calculate
297
231
3
63
Solve
276
99
90
87
Give a reason
252
72
73
107
Explain
234
166
0
68
Simplify
233
60
75
98
Describe
133
75
26
32
Sketch
120
48
46
26
Plot
92
46
17
29
Vocabulary demand & the wordiest questions real data
Most frequent hard non-technical words (a general-English barrier, separate from maths vocabulary): accurately, calculate, estimate, possible, decimal, different, minutes, significant, coordinates, figures, average, represents, statement, rectangle, following, centimetres, represent, integers, transformation, proportional, another, interval. Ranked by word count (a plain count, not a reading-age formula).
Find every task type / mark-scheme note matching a word (e.g. "Venn", "iteration", "show that").
Foundation vs Higher
Selected topic — marks per sitting on each tier (ignores the tier filter).
Marks over time
Compare two topics
Total marks per sitting (filtered) — topic A vs topic B.
Top 10
Reacts to the board & tier filters.
Most marks by topic
Click a bar to load it above.
Rising & falling topics
Linear trend of a topic's average marks per paper, year on year (years where the topic didn't appear count as 0) — based on past frequency, not a prediction. n = sittings behind the fit; topics with fewer than 6 sittings are not shown.
▲ Rising
▼ Falling
Which board leans in
Average marks per paper, selected topic.
Calculator split
Where the selected topic appears.
How it's assessed AQA-calibrated
AO1 (routine) / AO2 (reason) / AO3 (problem-solve) split for the selected topic, calibrated from AQA's official per-question AO grid. Where AQA has enough data for this exact topic it uses a topic-specific fingerprint (so e.g. indices reads AO1-heavy while surds reads AO2-heavy); otherwise it falls back to the topic's strand × question size. A model calibrated from the one board that publishes reliable per-question AO — not official per-paper tags.
How hard is this topic? real data
The average % of marks students actually score on this topic in real papers (Question Level Analysis: AQA both tiers + Edexcel Higher). Lower = more demanding. A descriptive past-performance figure, not a prediction.
Common mistakes examiner reports
What examiners report students get wrong on this topic, paraphrased from OCR/AQA/Edexcel examiner reports. ⭐ marks a grade discriminator (separates stronger from weaker students).
How they ask it
Specific task types for the selected topic (board & tier filtered), with the mark-scheme answer/method.
Question sizes
How many marks each question is worth, selected topic (filtered).
Pairs with…
Topics the selected topic is combined with in synoptic questions (all papers).
Where it sits in the paper
Marks by position, selected topic — early questions are usually easier, later ones harder.
Command words
What students are asked to DO across the filtered papers — solve / prove / explain… Inferred from each question's description, not the verbatim exam command, so read it as indicative; "Other" groups phrasings that didn't match a standard command word.
Top quick-win question types
The most common 1 & 2-mark question types (whole questions and sub-parts) — the quick marks. Reacts to the board filter.
Single-paper profile
Pick any paper to see its topic breakdown by marks.
Longest gap since last seen
Regular topics (in ≥5 of the filtered sittings) ranked by how long it's been since they last came up. This describes the past record — it is not a prediction that they're "due", and exam boards don't work to a rota.
Most common topic pairings
The synoptic combinations that come up most across all papers — from the ~8% of question-parts the crawl tags with two or more topics, so treat these as the strongest signals, not an exhaustive list.
Source: MathedUp crawl of real OCR / AQA / Edexcel GCSE (9–1) Maths papers, mark-scheme cross-checked. Owner preview.
Exam Paper Analysis