Talk:Anime/@comment-26039230-20150124153554/@comment-217.162.95.62-20150128071605

That's an invalid conclusion drawn from misinterpreting the real meaning of the data. What it shows is merely the ranking positions, which by itself tells us nothing. Kancolle had skyhigh ranking positions for the first week or so (it started off at #1 of all BD sales), so naturally the number of people potentially preordering now is greatly reduced simply through the fact that so many people already preordered.

Then, to take a look at the ranking position change for the day (0:00 to 23:00) of the airing of the anime:

Episode 1: 50 --> 13

Episode 2: 92 --> 60

Episode 3: 134 --> 78

Furthermore, this is only the ranking of the special edition of volume 1, there is also a regular version:

Episode 1: 129 --> 24

Episode 2: 172 --> 101

Episode 3: 331 --> 215

Of course ranking position alone tells us nothing, because it doesn't provide the data on the amount, perhaps the difference between position 1-50 is as big as the gap between 51-1000. The Amazon stalker, from where all this data is, provides an estimation based on points (a point being the equivalent of a unit sold). So, as for the point change on the day of airing:

Episode 1: +232 / +139

Episode 2: +94 / +45

Episode 3: +73 / +29

Sure, the gains are decreasing, but considering the amount of  preorders already made, that is something to be expected and not an indicator of episode 3 dropping the hammer on the sales.