Did you enter a contest to see how many non sequiturs you could pack into one comment, or did you just not actually read the article?Incidentally, there are lots of statistically valid marketing techniques to spotting trends. The problem in the motion picture industry is that the models aren't very reliable until there is audience test data, but by then the expense of production has already been incurred. But the marketing and distribution expenses haven't, and sometimes these are more than the cost of producing the film (easy to see why at $2000 per print). When films test badly, they go straight to video and can still sometimes make money.Anyway, this article is limited to modeling sequels only. Your first point would have been correct had you omitted the "no" at the beginning. Your second point is egregiously incorrect (see above). Your third point, while accurate, in no way contradicts anything said in the article so far as I can tell.I wouldn't normally point all of this ***** out except that you're getting dugg up pretty fast, which makes me wonder if anybody read the article. Yes people, trends can be spotted, at least early enough to make calculated decisions regarding marketing investments. Regarding certainty, you have a choice: timely or accurate, but rarely 100% of both.