It would be less entertaining, however, as everyone enjoys taking books down a peg, and it's not possible because as I discuss, GR hides the low absolute count entries from public display. Ideally you'd want all the books with 0 'abandon' tags, but GR will not show you those. It only shows books with at least 40 abandon tags. So you can estimate the least-abandoned within the most-abandoned, so to speak, but that's a pretty arbitrary thing to look at.

I’m not sure that the author was trying to find utility in the data so much as purely going through the exercise of analyzing it.

If a book is almost never abandoned, that might only indicate that it is short, or really easy, neither of which are hard to tell from a few seconds inspection. This tells you something you might not otherwise know from a few seconds' inspection.

I imagine the least abandoned would be something that comes late in a series (or at least from an author with many similar books), where basically everyone who picks it up already knows that they will like it (those who don't were filtered out by the earlier books).