Addendum to the FAQ: under-the-hood changes to the charts
In order to get more meaning from my core sources (Amazon, B&N, and Borders) (or rather, to extract more meaning from the data that is already being collected) starting with the charts dated the week ending 19 July, I’ll be using a new scoring method from the 3 core charts:
#1 as found is still scored at 100.0 points, from there points awarded decrease by three tenths through the first 67 titles, by two tenths through the next 201 titles, by one tenth through the next 300 titles, then by a mere .05 through another 199 titles, and then remaining titles found score a mere tenth of a point for as long as I care to keep up the data entry — another 133-200 titles depending on if I stop at the first 900 titles found (which I most often do, gratefully) or continue on to a full accounting of the top 1000 titles found on a given source chart.
[all of which is in the image above, only in a much-easier-to-understand presentation]
Nothing changes at the top (#1 still equals 100 points) but this gives us [me] a better accounting of midlist and backlist titles that appear on source charts, and was done primarily to help me detangle (and understand) a largish chunk of the manga midlist. I don’t think this breaks any ties or materially alters the rankings — I wanted some more detail, so I made arrangements to have it.
