I’ve been working on the new cruft engine, and it occurred to me that I could export my current anime ratings on this site, and perform a bit of regression on them to see what factors I like in an anime, and which ones I’m not so crazy for.

So, what does running the numbers through a spreadsheet tell me?

Apparently, I don’t care much if an anime has a chibi or realistic style, if it has a crazy ending or surreal elements, if its plot is romantic or angsty, or if the setting is Japan.

I like larger-budget productions: nothing has a larger effect on my overall rating than the “cheap” column. I’m a bit turned off by over-the-top attempts to turn me on: fanservice is also negatively related to an anime’s overall rating, but only to a small extent compared to budget. I suppose this explains how I can stand Burst Angel, which is what I’m watching right now: it’s a high-budget excuse for eyecandy. Alternatively, fanservice is used as a crutch by low-budget productions without a good plot; looking at the lowest six rated anime, I would describe all of them as cheap, and five of them as fanservicey. Nothing else has a serious negative impact on my impression of an anime.

So, that leaves things I want to see in an anime. My ideal anime apparently has an introspective but strong protagonist. The plot should be meaningful, full of fantasy, but resolved by the end of the series — and taking chances with something I haven’t seen before helps. A good soundtrack helps a little, but can’t save a show that doesn’t have anything else to offer (Arjuna? Is that you?); I’m surprised that humor doesn’t even have this small effect.

It’ll be interesting to see how other people compare once the new engine is done.

• Delegate
• Be careful what you wish for
• If a movie has sex appeal, its plot doesn’t matter
• Try new things
• Help others
• Make big plans, and then even bigger plans
• Be assertive
• Have fun
• The world is full of weird stuff waiting to be discovered
• If something you want doesn’t exist, create it

MySQL informs me that since starting FanCruft, I’ve reviewed over 100 anime. I’m just getting started, and there’s a long way to go (for comparison, ANN has information on 4735 anime titles). I was going to write tonight about the excellent Melancholy of Haruhi Suzumi, but this is “way more important.”

Of course, one review was actually written by a friend of mine: Last Exile.

So, what’s in store for FanCruft in the future? When I started, I had basically had to rely on my gut for what differed among anime. I did pretty well for just guessing — on many, many occasions, I’ve written that one anime reminds me of another one, only to have FanCruft identify the two as similar. It’s not very strong evidence, but it’s a starting point. To an extent, the changes in store for FanCruft are a reflection of changes in my worldview. Especially, I now believe that a crowd with excellent resources will usually beat a lone expert (not that I’m an expert on anime, just a dedicated fan).

So, FanCruft is going to be opened up. Big time. Eventually. I do believe that experts can help in setting up the overarching rules of a website, and that doing so is one part computers, and one part social-fu. I can’t give an ETA on this project, because it’s big.

I’ve also realized that the basic idea of rating things based on aspects is a powerful one. There are other websites that do this, of course, but they aren’t usually very transparent about where their data sources.

So, the future of FanCruft? Bigger, better, and more open. How cool is that?