Every summer, GCSE Maths results arrive alongside a grid of grade boundaries — the raw mark needed for each grade, on each paper, for each exam board. These numbers move from year to year, sometimes by several marks, and that movement causes genuine confusion for students and parents trying to interpret a mock result or a past paper score.
The confusion is understandable, because it looks as though the target is being moved arbitrarily. In fact the boundaries are the output of a deliberate process designed to do the opposite: to keep a grade meaning roughly the same standard of achievement from one year to the next, even though the difficulty of the papers themselves cannot be made perfectly identical every series.
How boundaries are actually set
GCSE Maths is assessed by exam only — there is no non-exam assessment component — across three equally weighted papers per tier. Once all three papers have been sat nationally, examiners and senior officials at the awarding body review how the cohort performed, alongside statistical evidence about that year group's prior attainment. Where a paper turns out to have been harder than intended, boundaries are set lower to compensate; where a paper was more accessible, boundaries rise.
This process — often called comparable outcomes — means the exam board is not trying to award a fixed proportion of each grade regardless of performance, but nor is it marking against a completely fixed percentage every year. It sits between the two: broadly stable outcomes nationally, adjusted paper by paper for genuine differences in difficulty.
Why boundaries move so much between years
- A paper judged to be more demanding in its wording, structure, or the difficulty of its problem-solving questions will typically have lower boundaries than an equivalent paper the following year.
- Changes in cohort-wide prior attainment (for example, disruption to earlier years of schooling) can shift the statistical evidence used in setting boundaries.
- Foundation and Higher tier boundaries are set independently, so movement on one tier does not imply the same movement on the other.
- Boundaries differ between exam boards, because each board writes its own papers to its own specification — a mark on an AQA paper is not directly comparable to the same mark on an Edexcel or OCR paper.
What this means for how you use past papers
The most common mistake is treating a fixed percentage — 'I need 80% for a grade 7' — as a stable target across every paper you attempt. Because boundaries move, that percentage will be roughly right on some papers and noticeably wrong on others. A paper with a grade 7 boundary of 62% and one with a boundary of 71% are both real, published examples of the same grade on the same tier.
| Series | Approx. grade 7 boundary (out of 80) |
|---|---|
| Series A | 56 (70%) |
| Series B | 51 (64%) |
| Series C | 58 (73%) |
The practical response is to always mark a past paper against that specific paper's own published boundaries, available from the exam board alongside the mark scheme, rather than a general rule of thumb. If you cannot find boundaries for a particular paper, treat the raw mark as informative but the inferred grade as approximate.
Using boundaries to set a sensible target
Because you cannot know in advance whether the paper you eventually sit will be judged harder or easier than average, the safest approach is to aim comfortably above the boundary you want, using a spread of past papers rather than a single year's data.
- Collect boundaries for your target grade across the last several series, for your specific exam board and tier.
- Note the highest and lowest raw marks that have achieved that grade, not just an average.
- Set your working target at or above the higher end of that range, so a harder-than-usual paper does not put the grade out of reach.
- Recheck this target after each mock, since your own performance data over time is more useful than any single boundary.
Grades below and above the traditional pass
It is worth being clear on what the numerical grades represent, since this affects how boundaries should be read. Grade 4 is broadly treated as a standard pass and grade 5 as a strong pass; many post-16 courses require a 4 or 5, while more selective routes and A-Level Maths itself typically expect a 6 or higher. A boundary conversation about 'just getting a pass' is really a conversation about the grade 4 boundary specifically, and it is worth knowing exactly where that sits for your board.
Higher and Foundation tier boundaries are not the same scale
Foundation tier covers grades 1 to 5, and Higher tier covers grades 4 to 9 with a safety net grade 3 available in some circumstances. Because the two tiers are different papers assessing overlapping but different content, their boundaries cannot be compared directly — a raw mark that gives a grade 5 on Foundation tells you nothing about what that same raw mark would represent on Higher, since the two papers are not testing the same material at the same difficulty.
What boundaries do not tell you
A published boundary tells you the mark needed for a grade on a specific paper in a specific year. It does not tell you which topics cost you marks, and it is not a substitute for going through your own paper question by question. Two students can arrive at the same total mark by losing marks in completely different places — one through arithmetic slips under time pressure, another through never having secured a whole topic such as similar shapes or compound interest.
That is why boundary-chasing should never replace the more basic discipline of marking your own past papers strictly against the mark scheme and classifying every lost mark by cause. The boundary tells you whether you passed a particular hurdle; the mark scheme tells you what to do next.
The short version
- Boundaries are set after each exam series based on how that paper actually performed, not fixed in advance.
- They move year to year and cannot be compared across tiers or exam boards.
- Always mark a past paper against its own published boundaries, not a general percentage rule.
- Use a range of past boundaries, not a single year, to set a safe working target.
- A boundary tells you if you cleared a hurdle; only your own error analysis tells you what to fix next.