Why a mock exam score from us means the same as an 11+ score
A percentage on its own tells you very little. “68% in maths” could be a triumph on a hard paper in April or a worry on an easy paper in August — and a child born in September is sitting the same paper almost a year older than a child born the following August. That is why the real 11+ does not report percentages: it reports standardised scores. Ours work the same way.
What goes into our standardised scores
Behind every mock result we adjust for three things before a standardised score is produced:
- Your child’s date of birth. Exactly as in the real 11+, younger children are not penalised for being younger. A summer-born child is compared fairly with an autumn-born one.
- When the exam was sat. A mock taken in spring is not the same event as one taken the week before the real thing, and children who sit a catch-up session on a different date are treated accordingly.
- How each paper actually relates to the real 11+. This year, 180 of the children who sat our mocks went on to sit the September 11+. That gave us nearly 1,700 matched, paper-by-paper results — the same child, the same subject, our exam and the real one — from which each mock paper is calibrated against the genuine article.
The precise recipe is ours and stays in-house — it is a large part of what you are paying for — but the outcome is simple to state: a standardised score on one of our mocks is designed to mean the same thing as the same score in September.
The result: every mock on the 11+ scale
If the standardisation is doing its job, the spread of standardised scores from each of our six mock exams should sit on top of the spread of real 11+ scores — same centre, same shape. Here are two examples; the other papers look the same:
Each coloured curve is one of our six 2025 mock exams after standardisation; the black curve is the actual September 2025 11+. The centres line up: a 110 in April means what a 110 means in September.
The evidence behind it
These curves are not an artist’s impression. Below are three of our exams, spread across the season. In each panel, the blue “ladder” is the mock’s standardised scores and the black ladder is the actual 11+ scores of the very same children — each step is one real child. Reading across at 50% gives the middle child. The curves cross in the middle and track each other along the whole range: the mock is measuring what the 11+ measures. The other exams look the same.
Three mocks, spring to September. Blue: the mock, standardised. Black: the same children’s actual 11+ scores. The pairs cross at the middle child and hug throughout — for every exam, whoever happened to sit it.
Same child, different day
The natural assumption is that an exam score simply measures how clever a child is — so the same child should get the same score every time. The data says otherwise. Plot the same children’s scores in one mock against their scores in another and you do not get a straight line; you get a cloud:
Left: 63 children who sat both maths papers — same children, same subject, weeks apart. Right: two real children from 2025 (names withheld): Child A scored 113 ± 2 across six mocks; Child B averaged higher but swung from 101 to 122.
Every child’s performance has a range, not a single value — there is no such thing as a 70% child. And mishaps do not queue politely: a gap in knowledge costs time, time pressure breeds stress, stress breeds mistakes — which is why scores lurch rather than drift. Happily the spiral runs both ways: a little more speed lowers the stress, cuts the mistakes, and produces jumps in score that look miraculous and aren’t. That range is the most important thing a season of mocks reveals, and it is why one result — good or bad — should never be over-read.
It also reframes what to aim for. The 11+ is one day. A child performing at 115 ± 5 clears a 110 bar five days out of six; a child at 120 ± 20 — “smarter” on average — fails to clear it roughly one day in three.
What parents can take from this
Having the same children’s mock and real results side by side teaches lessons that no single score can:
- Consistency beats brilliance. As the cloud above shows, scores genuinely bounce between sittings, even for able children. The children who do best in September are rarely the ones with a single spectacular mock; they are the ones whose scores stopped bouncing — a narrower ±, not a higher peak. Most of that is not intelligence — it is timing, technique, and not leaving questions blank. That is trainable, and it is where mock practice earns its keep.
- An early mock pays. The children who sat our first mock in spring went on, as a group, to strong September results — and their verbal reasoning scores ended up the most tightly bunched of any group, which is what sustained practice looks like. We cannot promise the mock causes this rather than reflecting families who plan ahead; either way, the pattern favours starting early. Exam experience itself compounds, with diminishing returns: biggest from the first sitting, roughly halving with each one after. The earlier that first jump happens, the longer it is banked before September, and first-exam nerves only happen once — better in March than in September. And a modest spring score is often nothing more than unfinished curriculum: plenty of what the 11+ tests simply has not been taught by April. The standardisation knows that, even though a raw percentage never will.
- A verbal reasoning surge is partly familiarity. Tracking the same children across the season, verbal reasoning mock scores drift upwards by roughly a dozen standardised points from spring to late summer, relative to where those children finish in the real exam. Maths shows no such drift. Verbal reasoning is simply the most practice-sensitive paper: a modest verbal score in April is not a verdict, and a late-August surge is partly familiarity rather than transformation. Our standardisation accounts for this — a raw percentage never will.
- Age really is handled. A summer-born child is not given a few courtesy points — age is factored in mathematically from date of birth, the same way the real 11+ does it, before the score ever reaches you. There is nothing left to allow for.
Prepared, not born smarter
One last picture. Take our August BWS mock and split its sitters by experience. The children sitting their very first mock scored about ten standardised points below the rest — and their September results confirmed the reading was real, not first-day nerves: as a group they were still behind in the 11+. Starting in August doesn’t just mean one rusty exam; it means arriving ten points back with three weeks left to close the gap. The children who had started earlier — even with just one previous mock — were indistinguishable from the veterans.
The same August exam, same papers, same day, split by how many mocks each child had sat before. The gap is not exam-day rustiness — it is months of preparation, made visible.
Reading your child’s next mock result
Look at the standardised score, not the percentage. Compare it with the last one, and care more about the trend and the steadiness than about any single number. A wobble is information, not a verdict — and thanks to the 180 children before yours, the number you are looking at is on the scale that matters: the real one.
A note on 2026 scores
Our standardised scores this year are calibrated against 2025: we found the percentage a child needed last year to be on course for each 11+ score, and we have applied the same rule this year. Across all our 2026 mocks, raw marks are running a little lower than 2025’s. There are two possible reasons, and we cannot tell them apart until the real results arrive. It may simply be that a wider range of children are now sitting mocks with us, earlier in their preparation. Or it may be that performance has dipped across the whole year group — and the 11+ always awards 100 to the average child of its own year, so if everyone has slipped a little, September’s exam will be slightly more generous than our mocks have been: a percentage we scored as 100 might earn a 106 on the day.
We have deliberately not adjusted for that second possibility — we would rather understate than flatter. If it turns out to be true, your child’s real score should come in a little above what their mocks suggested: a surprise, but in the right direction. We will re-measure in October when the 2026 results are in, as we do every year.
Read more
Turning the measurements into a better September — the routines to run between mocks, what to do inside the exam room, and how to shrink the ± — is covered in our companion guide, “Preparing for the 11+”. This guide on how best to prepare for the 11+ is given for free to anyone that books mock exams with us. It will be stored in the Mock Exams folder of your SEP Personal Online Drive.