A study of the arcane

In the joyous fashion of the SA postal system, my royalty statement arrived. For the period to June 2008. Hey what's a year? I should be grateful. They arrived, not a thing that is always gauranteed -- I've yet to recieve a single ARC, although they're on sale by various collectors, and as often as not I don't get my author's copies until the book has been out for six months, if i get them at all. On one occassion I got 5 (not 10). The country has its joys but there are a few tempering factors... Anyway, back to the arcane - it's only when -as a statistician - you peer at these documents that you realise that understanding it involves the dark arts (more like necromancy than statistics or accounting)
Here is the theory HC:PB ratio will be around 1:4
Here is the fact - they range 1:2.1 to 1:7.3
Theory: 95% of sales of a new book occur in the first 3 months
Fact: The Forlorn - 1999 - sold 123 copies in the last 6 months. It's sold more than 25% more than the initial report. It's not on many shelves.
Theory: Sequels will sell more or less 20% less than the first book.
Fact. huh? Whut? sometimes the second HC outsells the first, seldom loses more than 5%, and the paperback figures are all over, but mostly a lot lower... because (theory by Dave proved by his loco... uh, local bookstore ) no one wants to by the second book in a series when the first isn't for sale.
Theory: names draw sales. Actuality - names draw hardcover sales. except A MANKIND WITCH -solo - hardback, with an awful cover, outsold anything by Eric and I combined. The PB on the other hand sold badly, barely getting 1:2.1 ratio. PB sales I think are cover and content driven - and distribution naturally. Actually the distrib might be important for the hardcover sales too.

I can't make head or tail of it. I'll stick to writing with occasional breaks to grumble about it.