Oh man, I remember a few years ago I wrote a Python script that used CCC data to help flip seasonal items on Amazon's own marketplace. The idea being that I would buy low from a third party seller and sell high later using FBA. Kind of a fun experiment but kind of a time sink.

Generally speaking, graduate-level textbooks (which release new editions at a far slower pace) had the largest and most predictable margin. That said, I hated that fact, since I think textbooks are also the biggest racket in the education universe, and its one of the other reasons that I stopped.

Amazon collected their fees in both directions, so I always assumed that if they realized what I was doing, they had little incentive to care.