@Alzrius My condemnation of your above post might be coming off as stronger than I intend - or more specifically, more censuring of
you than I intend. I'm pretty adamantly against most of the points you make, particularly the first one, but I don't mean it to come across at all personally - not even a little.
But, let me tell you about BookScan, just to be clear as to why I'm so sure that the first point is total BS. I've had decades of experience with reading analysis of bookscan reports. They have
always been unreliable, even if they're still useful to look at. They've ALWAYS had "reporting fragments" - places where books are miscategorized, or otherwise misreported, and to get a true idea of how it's sold (even just how it's sold through locations that report to bookscan, keeping in mind that it has little to do with the entire market, much of which does NOT report to bookscan) you have to ADD UP multiple lines of Bookscan reportings for a single product.
The guy who "reported" bookscan's "thousands of copies" was only showing ONE LINE, and the LOWEST SELLING LINE of reporting for the PHB. Presumably (I've seen it hundreds of times) one could have looked through more data to find where there were more lines featuring the same book, and added them up. This happens all the time for comics.
ALSO - for example, I doubt that there are many (if ANY AT ALL) FLGSes that report to bookscan, so... even if it WAS all the sales of stores that do (and it couldn't possibly include AMAZON, for example, which probably categorized it differently) it wouldn't come close to revealing how the PHB sold IN TOTAL. It's just a useless bit of reporting. Totally not worth every thinking about ever again.
This is all not to add, that it's so obviously a ridiculous number that it would mean that MY OWN TINY LITTLE FLGS would have been responsible for like 2% of the ENTIRE SALES and my distributor would have had (I saw the pallets) something like 1/3 of all the books - for one small part of Canada ALONE. It's silly!