One of the quirks of A-level results is the extent to which teacher-assessed grades were or were not used. Minority subjects (which favour private schools) were more likely to get better results. Fewer than 5 students in a school cohort meant they used the teacher grade.

Between 5 and 15 students doing a particular subject meant the centre-assessed grades still played a role. But if there were more than 15 students, it appears the individual student rankings were just laid over the standardisation model to allocate the grades.

This mess has also illustrated pitfalls of reforming A-levels. In Wales, they've still been able to use AS-levels as a gauge. But in England, AS-levels were decoupled from A-levels & system switched to everything resting on final-year exam grades, with coursework mostly scrapped.

It meant there was little external evidence of individual student achievement to go on when the exams were cancelled. That's why we've arrived at the rather odd suggestion of using internal mock results as an alternative. Bring back AS-levels and coursework!