Report ChronusZ's Profile

Math + anime = <3

I generally rate shows by some vague notion of "average quality", but there are some notable idiosyncracies in the way I apply this rule:
- If an anime is particularly exceptional in some respect but has a lot of flaws, then I will bound the extent to which the flaws affect my rating, so that even if the show is filled with issues I might still rate the show pretty highly.
- If the ending is particularly frustrating then I might ragequit and give the show a much lower score than it rightfully deserves based on a holistic computation.
- Conversely, if the beginning is bad but the show eventually picks up speed then I won't factor the beginning into my rating. But if a show starts out really good and then becomes mediocre then as long as it doesn't become bad I'll consider it above mediocre.
- If I feel that a show is criminally underrated then I will occasionally overrate it as compensation.
- Sometimes I overrate silly shows for the memes.
- My scoring criteria drift slowly over time and I don't rerate shows unless I rewatch them, so some shows that I watched a long time ago and never rewatched will be rated according to the mysterious criteria of a lost age.

I have a significant bias against anime set in space; I struggle to take the setting seriously. I also have trouble watching anime centered around food, since it makes me hungry (I can enjoy them while eating, but this prevents me from marathoning, and the more time I spend watching an anime the more likely it is I'll lose interest).

I rewatch anime a lot. Most shows I like I've watched at least twice, some I've watched too many times to count. I don't record rewatches on MAL, but at this point I've probably spent more time rewatching anime than I have watching shows the first time.

I started watching anime in June of 2014 (I had watched a few shows prior, but this was when it really kicked off). In early 2015 I noticed an amusing pattern in my watch time: if y denotes the number of hours I've watched and x denotes the number of days that have passed since June 1st 2014, the formula y = ln^4(x) seemed to approximately hold with quite high consistency. I continue to check this every now and then, and it seems to never diverge by more than a few % error in y. I have no explanation for this fact but I love it.