Though there are plenty of sources out there from which one can learn about it, the inner workings of the US anime market are often characterized by what we don’t know. I track US anime releases on amazon because I’m curious about said market. That curiosity has several points of origin, but I chasing after one primary question; do fans buy the titles one would think they buy, to the extent one would think, from their presence on social media?
But tracking US amazon rankings is only useful insofar as these rankings have the potential to correspond to real sales data. And even that’s not particularly useful if the “real” data has the potential to be off by a factor of 100 due to unpredictable factors. One obvious potential contributor to the potential for underestimations are long tails, the combined contribution of totals from all weeks the release doesn’t make the threshold on a given set of charts. These are a very familiar foe when it comes to trying to compare sales figures, and are definitely worth addressing. There’s plenty of reason to believe long tails might be a factor in the US home video market – per-episode prices are much lower than they are in Japan, and thresholds are much higher.
I previously used a trial account to access the lifetime home-video sales totals of various anime movies in Nash Information Services’ OpusData database. Recently, I found out that the companion website to the service TheNumbers, stores years worth of back-data on its weekly charts, though its thresholds are, by design, severely limited compared to those measured on OpusData. Using those charts, it was a trivial task to go back to each movie’s release date and check how much of its lifetime total was accumulated while it was on the BD20/DVD30 toplists. Note that 2 of the 12 movies in the original sample never made said toplists and thus cannot be compared. Results are shown after the break.