ESAT preparation guide
ESAT in October or January: what the official data can settle
- Is the choice yours
- One sitting per cycle and no January resit; a Cambridge course with an October UCAS deadline means October
- Is it the same scale
- Yes. Both events used the same scaling constants, so scaled scores are comparable across events within the cycle
- Where the distributions differ
- October: median 4.5 and 90th percentile 7.0 on all five modules. January Mathematics 1: 4.1 and 5.4
- Is one sitting easier
- No basis for saying so. The scale already removes form difficulty, so the gap describes the cohorts rather than the papers
Every figure here comes from a printed table or sentence in UAT-UK's 2025/26 ESAT technical report and its 2025-26 annual report. Nothing is read off a chart image and nothing is derived across tables.
- Constraints come before data: one sitting per admissions cycle, no resits in January, and a Cambridge course with an October UCAS deadline leaves October as the only option. This cycle 10,116 sat in October and 4,077 in January, 71% to 29%.
- In October the median is 4.5 and the 90th percentile 7.0 on every single module, because the scale is pinned to the October ability distribution. That column is the reference frame by construction.
- January sits lower and tighter: Mathematics 1 has a median of 4.1, a 90th percentile of 5.4 and an SD of 1.03, against 4.5, 7.0 and 1.59 in October. Physics and Mathematics 2 show the same shape.
- This is not a difficulty gap. The same scaling constants were used for both events, and the report states this was done so that scaled scores would be comparable across events.
- Module coverage is a hard constraint: only 16 candidates took Biology in January and the report prints NA across that entire row, while Chemistry had 332. For those two modules January offers almost no benchmark.
Both sittings, module by module
10,116
Candidates who sat in October 2025, 71% of the ESAT cycle
4,077
Candidates who sat in January 2026, 29%, against 23% a cycle earlier
4.5 / 7.0
October median and 90th percentile on every module, where the scale is anchored
One set of constants
Shared by both events so that scaled scores are comparable across them
| Whole cycle | October 2025 | January 2026 | |
|---|---|---|---|
| Mathematics 1 (compulsory) | 14,193 · mean 4.58 (SD 1.47) · median 4.4 · 90th 6.7 | 10,116 · mean 4.73 (SD 1.59) · median 4.5 · 90th 7.0 | 4,077 · mean 4.19 (SD 1.03) · median 4.1 · 90th 5.4 |
| Mathematics 2 | 12,113 · mean 4.40 (SD 1.63) · median 4.2 · 90th 6.5 | 8,082 · mean 4.73 (SD 1.66) · median 4.5 · 90th 7.0 | 4,031 · mean 3.75 (SD 1.33) · median 3.7 · 90th 5.4 |
| Physics | 11,290 · mean 4.25 (SD 1.71) · median 4.1 · 90th 6.5 | 7,717 · mean 4.58 (SD 1.70) · median 4.5 · 90th 7.0 | 3,573 · mean 3.55 (SD 1.51) · median 3.4 · 90th 5.5 |
| Chemistry | 3,146 · mean 4.61 (SD 1.74) · median 4.5 · 90th 7.0 | 2,814 · mean 4.72 (SD 1.72) · median 4.5 · 90th 7.0 | 332 · mean 3.63 (SD 1.62) · median 3.5 · 90th 5.7 · max 8.9 |
| Biology | 1,547 · mean 4.44 (SD 1.87) · median 4.5 · 90th 7.0 | 1,531 · mean 4.46 (SD 1.86) · median 4.5 · 90th 7.0 | 16 candidates · the whole row is marked NA, with no statistics published |
Why October lands on 4.5 and 7.0 every time
The ESAT 2025/26 scale
Pinned to the October 2025 ability distribution, then reused unchanged in January
Anchor 1: median ability maps to 4.5The 50th-percentile ability at the October event, module by module (Table 2)
- Mathematics 1 ability 0.4049
- Physics 0.3454 · Chemistry 0.2960
- Biology 0.5970 · Mathematics 2 0.0159
Anchor 2: 90th-percentile ability maps to 7.0The second row of the same table, again October only
- Mathematics 1 ability 2.7701
- Physics 1.8273 · Chemistry 1.9269
- Biology 1.6789 · Mathematics 2 1.6918
A line through the two anchors gives the constantsTable 3, with a different constant and multiplier per module
- Mathematics 1: constant 4.0720, multiplier 1.0570
- Physics 3.9173 / 1.6870 · Chemistry 4.0463 / 1.5329
- Biology 3.1205 / 2.3107 · Mathematics 2 4.4763 / 1.4917
The same constants are applied in JanuaryThe stated purpose: to keep scaled scores comparable across events
- Transformed scores are capped at 1.0 and 9.0
- The scaling is re-fixed each cycle, so cycles are not comparable
Tap a branch to unfold
The same score sits in a different place in each sitting
Mathematics 1, 90th percentile
7.0 in October, 5.4 in January
25th percentile 3.6 in both; SD 1.59 against 1.03
Mathematics 2, 90th percentile
7.0 in October, 5.4 in January
25th percentile 3.6 against 2.9; SD 1.66 against 1.33
Physics, 90th percentile
7.0 in October, 5.5 in January
25th percentile 3.4 against 2.5; SD 1.70 against 1.51
Chemistry, 90th percentile
7.0 in October, 5.7 in January
The January maximum is 8.9, the only printed maximum in Table 5 below 9.0
Two different rooms: what the annual report prints about each cohort
Bookings against tests actually delivered
The two windows are growing at very different rates
UK and overseas bookings by sitting
Access arrangements and bursaries (all three tests in that window)
Using this to decide which window to sit
Check which window your course accepts
Look at your module combination
Weigh readiness against the calendar
The fee is the same either way
Frequently asked questions
Is there an easier sitting?
Is a 5.4 in January the same thing as a 7.0 in October?
If October goes badly, can I sit again in January?
I need Biology or Chemistry. Is there any January benchmark?
Can I compare this year's scores with last year's?
Official sources (5)
- UAT-UK ESAT Technical Report 2025-26, Table 5: scaled score statistics by module and sitting
- Same report, Tables 2 and 3: October anchor abilities and the shared scaling constants
- Same report, section 4.1 (candidate numbers by event, module combinations, the Cambridge October rule) and section 5 (reliability)
- UAT-UK Annual Report 2025-26, section 3 (booking rules, window length, fees) and Tables 1, 2, 3, 5 and 6 (bookings, geography, bursaries and access arrangements by sitting)
- UAT-UK Annual Reports index (public entry point for the 2025-26 reports)
More ESAT guides
UAT-UK, Pearson, OCR, the University of Cambridge, Imperial College London and other names and logos belong to their respective owners.
Put it into practice
74 ESAT papers on FrontierVUE, including 41 Frontier Original mocks built to the current format, every question with a worked solution. 2 past papers are free.