The algorithm last year was atrocious, and it's wrong to pretend otherwise. In particular, it took input from more or less everybody apart from the students themselves -- past performances at the school, national smoothing, teachers' estimated rankings of their class, etc. Everybody, in short, other than the students' own work. It was quite correctly scrapped; the only mystery is why it was allowed to get so far, especially after the same thing happened in Scotland with the exact same pacing.