Their average scores were higher for the second paper than for the first one (68 vs 58 per cent). But more than a quarter of students did worse in the second paper, despite the months of preparation.Yep, that's right. That says absolutely nothing. Who's to say that 1 quarter of students wouldn't have done worse if they'd taken exactly the same paper the very next day? 75% of the students had better grades after traditional teaching than after the new method - whose to say the rest wasn't noise? They don't seem to have a control group cramming in a different manner. According to the article "the repetition is key", but there's exactly nothing in the results that indicates this is the case - they didn't have some other group trying to cram the material in 1h 30 minutes without repeating it.
This might well be a preliminary study, and the techniques it suggests might well prove to be useful in a broader context, but at the moment what they have proven is that trying to cram lots of material in 1 and a half hours is less effective than teaching it over the course of 4 months. Amazing!!