One of my big pet peeves in data analysis is making major inferences and broad, sweeping statements based on a very small sample size. This paper appears to be a prime offender.
Looking at this case, there's a pretty good argument that Netflix and the LoveFilm app launching in the UK, QuickFlix Streaming launching in New Zealand, The Artist (with a French director and star actor) getting a lot of attention and being released to rental, a particularly chilly winter in western Europe, and god knows what else all contributed in whole or part to the observed phenomenon. I'd love to have whatever data the authors of the paper had, to facet the data by movie studio and country - did different countries peak on different weeks, and the like. Alas, the data consists of the sales and rentals of two anonymous major motion picture studios.
That none of these potentially seriously confounding factors were mentioned or explored before reaching for a conclusion is somewhere on the scale between negligence and dishonesty, and the lack of data transparency renders the the paper useful for nothing more than kindling.
Any fellow Tartans here on HN (I know there are a bunch of us) want to sign on to a letter requesting an investigation into the integrity of the paper and IDEA?