Hi, I like the Weinberg post - maybe because I'm also by nature a statistician. But there's no great surprises in any of the three posts. They all say the same thing - it's a skewed sample set being overused (the top 1.5% to 4% of books in a select few categories in Amazon ebooks if Weinberg is accurate). Hugh mentions this upfront which is the right thing to do in any statistical analysis, but I'm not sure he completely allows for this in extrapolating from his results to his conclusions. As I said before the main thing to take from Hugh's survey is that if you can make a reasonable fist of producing a good story, cover, blurb, marketing etc and you want to sell as an ebook on Amazon, you will likely make more money from this channel as an indie than as trade published. Cheers, Greg.