There are so many variables and so much variability per person and so much delay involved between training and results that it's probably almost entirely impossible to evaluate single approaches.

Maybe it works because it's a change, and making changes keeps our minds slightly off-balance. I think sticking to a single routine for a long time runs the risk of the brain getting used to it and producing a decreasing rate of success. There are obvious counter-examples though, so "in all things, moderation, including moderation!"