Analyzing book sales is like reading tea leaves... you can fool yourself into believing you know something, but, in truth, you've just got a pile of damp darjeeling.

Still, that's what writers do.  In the absence of real data (my royalty statements run nearly a year behind actual sales), we pore over what little data we can find and pretend to know what it means.  Anyone who has watched amazon rankings jump a million spots in an hour (based on perhaps as few as two or three book sales) will understand. 

Case in point... the New York Times bestseller list.  Pretty straightforward, right?  It's a list of the books that have sold the most copies, right?  Wrong.  The formula by which the Times generates its list is a closely guarded secret, but the parameters are pretty well established.  First, the Times pre-selects a list of books that have the potential to be best sellers (based on previous sales, or publisher hype, or author celebrity, or print run, or...).  Then it aggregates weekly sales data from a small sample of book sellers.  And the books that sold the most copies that week, in the pre-selected markets, from the list of pre-selected titles are that week's best sellers.

And why do I tell you all this?  This past week-end, I sold a few books.  Not a lot of books, not a life-changing number of books, a few books.  But they were sold by a book seller who reports his data to the New York Times.  No, I'm not a New York Times best seller.  Far from it.   But this week, somewhere in the New York Times data base, somewhere far below the best sellers, I imagine that there's a cell in the data base that says It's Beginning to Look a Lot Like Murder - 10 books.  And it makes me smile.

To read the comments or to add a comment of your own, please use this link.