This blog post tries to calculate the books that are most likely to be abandoned after a reader has started them using data from GoodReads.