Four chapters. Every figure from a primary source.
The 80/10/10 weighting is on contextualised GCSE — only for candidates with a cGCSE score (most UK applicants). Internationals and those without a cGCSE are ranked 50/50 on the two LNAT components, with case-by-case adjustment.
The 2025-26 Faculty-central rank is documented in the Law Admissions Report: 80% cGCSE + 10% LNAT MCT + 10% LNAT essay for candidates WITH a cGCSE score; 50% MCT + 50% essay for those without one (most internationals, some non-standard quals), with case-by-case adjustment.[12] The rank is Faculty-wide — but colleges then add up to ~0.5 candidates per place to the shortlist, reserve which shortlisted candidates they want to interview, and make their own final decisions independently after interviews. There is no central admissions meeting.
Oxford's published Law (Jurisprudence) offer is A*A*A at A-Level, with no specific subject requirements.[15] There is no preferred-subjects list and no automatic weighting against humanities versus sciences. The faculty's published guidance is that any combination of essay-based and analytical subjects works.
What the offer means:
Almost all Oxford Law offers are conditional — typically A*A*A. The faculty made 252 offers for the 2025-26 cycle, of which 2 were deferred-entry and 35 were Course 2 (Law with Law Studies in Europe).[12] Unconditional offers are exceptional and reserved for candidates who have already completed their qualifications.
An open offer is something different — it is a college-allocation outcome, not a grade outcome. About 14.8% of 2025-26 applicants applied "open" (no college preference), and the faculty allocates them via the pool.[12] Open applicants face roughly the same effective offer rate as direct applicants — the pool system smooths out college-by-college variation.
Oxford uses GCSEs at shortlisting, but never in raw form. Every applicant's GCSE record is converted into a cGCSE — a contextualised score that adjusts for the school's prior KS4 attainment.[6] A high-performing applicant from a low-performing school yields a positive cGCSE; an under-performing applicant from a high-performing school yields a negative one.
The exact cGCSE calculation is exempt from disclosure under FOIA section 43(2).[6] What is public:
The Tomkinson FOI provides 1,566 anonymised applicants' GCSE counts (2021 cycle).[5] Aggregate statistics for applicants with at least 5 declared GCSEs:
For 2024-25 specifically, the faculty publishes shortlist and offer averages for raw A* counts: 7.73 A*s for shortlisted candidates, 8.78 A*s for offer holders.[11] The 1.05 A*-grade gap means the GCSE record continues to discriminate between strong candidates even after they reach interview.
| A* count | Number of applicants |
|---|---|
| 1 | 89 |
| 10 | 57 |
| 11 | 33 |
| 12 | 8 |
| 2 | 99 |
| 3 | 103 |
| 4 | 93 |
| 5 | 83 |
| 6 | 82 |
| 7 | 83 |
| 8 | 87 |
| 9 | 68 |
The mode is around 8-10 A*s. Most law applicants come in with at least 6 A*s. That means a record of "all A*s at GCSE" — distinctive 15 years ago — is now table stakes. The variation that matters happens above this threshold (10+ A*s) or below it (4-6 A*s with a strong contextual case).
If you went to a school where most students don't sit GCSEs and you got 5-6 A*s, your cGCSE score will be high in absolute terms. If you went to a high-attaining independent school where most students get 8+ A*s and you got 7, your cGCSE will be neutral or negative.[6][14]
The precise cGCSE formula is unpublished. What is known: both the raw A* count and the school baseline matter. A candidate with a stronger LNAT and weaker raw GCSEs from a less competitive school may be on equal or better footing than a candidate with stronger raw GCSEs at a top independent.
For 2025-26 entry the Faculty Selection Committee shortlists candidates using a published, formula-driven rank — the most explicit Oxford has ever published.[12]
The 2025-26 shortlisting rank (with cGCSE):
Rank = 0.80 × cGCSE + 0.10 × LNAT MCT + 0.10 × LNAT Essay
Without cGCSE:
Rank = 0.50 × LNAT MCT + 0.50 × LNAT Essay (then adjusted on individual basis)
The 2021-22 to 2024-25 reports describe shortlisting as "equal weighting of cGCSE, LNAT MCT, and LNAT essay" — i.e. 33%/33%/33%.[8][9][10][11] The shift to 80%/10%/10% in 2025-26 is the largest publicly-documented change to Oxford Law's admissions process.
Contextualised GCSE performance now dominates the shortlist decision. The LNAT scores still discriminate between candidates with similar cGCSE scores, and the FSC review can pull in "super-LNAT" candidates outside the rank, but the centre of gravity has shifted toward GCSE attainment in school context.
After the algorithm produces the rank, the FSC reviews each candidate to consider:[10][12]
The FSC reviews bring the shortlist to ~2.0 candidates per place. Then colleges add further candidates from their own pool to reach ~2.5 candidates ultimately interviewed.
For 2025-26 applicants. The 80% cGCSE weighting makes GCSEs arithmetically more determinative of shortlisting than the LNAT for most candidates. A weak cGCSE score is hard to compensate for through LNAT performance alone.[12]
The right prep approach varies sharply with where you are starting from. The data above sorts applicants into four common profiles.
Top school · A*A*A predicted · 9+ A*s
Your weakness will be the LNAT MCQ. You're used to tests where preparation produces consistent results; the LNAT punishes that style. Most strong-on-paper candidates finish their first practice section in the 22-25 MCQ range and panic.
Strategy. Treat the LNAT like a reading discipline, not a content subject. Spend 60% of prep on dense argumentative reading (LRB, NYRB, FT comment) without timing. Layer timing on only after you can reliably hit 30+ accuracy untimed.
Mid-pack school · A*AA predicted · 6-8 A*s · MCQ mid-20s
You're the candidate the offer-rate curve is steepest for. Every MCQ mark from 25 to 30 produces meaningful shortlist gains.[1] cGCSE will help you marginally — you'll likely get a positive adjustment if 8 A*s is above your school's average.
Strategy. 50% Section A drilling, 35% essay practice with structured feedback, 15% interview prep from November. Consider Mansfield, LMH, Hertford, or Worcester.
Significant school disadvantage · strong personal record
Oxford's contextualisation is real.[6][14] A 25 MCQ from a school where most students don't sit GCSEs is treated very differently from a 25 MCQ from a top-tier school. The faculty's published commitment is reflected in the data — 76.6% state-school admits, 21.3% POLAR 1-2.[14]
Strategy. Apply. Don't self-select out. Even a 27-28 MCQ from a contextual applicant is a strong signal. Consider Mansfield, Harris Manchester, LMH, or Pembroke.
~31.7% of the 2025-26 pool
Roughly 30-35% of Oxford Law applicants are international.[12] The cGCSE measure works less neatly for non-UK qualifications; Oxford uses a separate contextualisation for international applicants, not published in detail.
Strategy. Plan the LNAT seriously — international applicants without UK context have less margin for cGCSE benefit. Sit at a Pearson VUE centre near you. Familiarise yourself with the Oxford interview format if it is unfamiliar.
Every numerical claim above ends in a [n] superscript that links here. Each link opens the primary FOI attachment, faculty-published report, or official course page.
First publication of the 80%/10%/10% (cGCSE/MCT/essay) shortlisting weighting. Without-cGCSE 50/50 fallback. Spreadsheet-of-strong-candidates rescue mechanism.
Published A-Level offer, application timeline, course structure.
1-5 interview scoring scale, GCSE-shortlisted vs GCSE-offered means, six-criterion tutor rubric, redistribution priority for deprived backgrounds.
Refusal under FOIA s.43(2) to disclose the exact contextualised GCSE formula. Confirms cGCSE adjusts raw GCSE record against the school's KS4 attainment baseline.
Tomkinson dossier — per-applicant GCSE counts, LNAT scores, and shortlisting outcomes for the 2021 cycle (n=1,567).
Demographic baseline aggregated across 2022-2024. UK applications, offers, admits by school type, ACORN, POLAR, gender, ethnicity.
Faculty-level publication. Funnel, LNAT averages, demographic split, and shortlisting weighting for the 2021-22 cycle.
Funnel, LNAT min/max and average ranges, second-interview activity for 2022. First explicit publication of the 3-point rank methodology.
Course-1 / Course-2 split, super-LNAT pull-in rule, bGCSE pull-in rule, second-interview redistribution detail.
Per-applicant Section A, essay, college, and offer outcome for all 1,814 applicants to BA Jurisprudence in the 2025 cycle.
More on the overview, LNAT, and the interview.
Headline numbers, five-cycle funnel, demographic baseline.
Open Overview →Distribution, offer-rate curve, calculator, marking criteria.
Open LNAT →Score distribution, rubric, sample questions, tips.
Open Interview →Every figure traced to its primary FOI or faculty report.
View the sources →