regarding manga/anime formula.

I think Assertn's method is good for a rough estimate and the theory behind it is solid.

I think you could make it even more precise with the following tweak.

First break up the anime into pieces, filler and non-filler. Filler being episodes which share content 100% with the manga.

So lets say (to simplifiy) that there were only 3 non-filler sections of naruto. Section A) consisted of 20 manga issues and 5 episodes, or 4 issues/episode. Section B) consisted of 25 issues and 5 episodes, or 5 issues/episode. C) contained 20 issues and 10 episodes (damn flashbacks) or 2 issues per episode.

So we have ratios of .2 episodes per issue, .25, and .5. Given a large enough sample size i think we will see a pretty tight average and standard deviation.

As far as i can remember, every issue of manga was represented in the anime series, e.g. all fillers in manga were represented in anime.

So to find the average # of issues/episode, one would first have to find the average of the subsets. This will be the average issues per episodes for all the manga issues. Call this value "X".

But then we have to take into account filler episodes. For this i think the best method would be to look at the current ratio of filler eps to content eps and assume that the end result will be similar until the manga catches up (at which point we will be getting nothing but fillers until more content is introduced). Call this value "Y".

So the formula for A(t) would actually be:

A(t) = X*M(t) + Y*M(t)

Where the first term is anime derived from content, and the second term is episodes derived from filler.

Our final ratio equation would be A(t) = (X+Y)*M(t)

At least i think that will work. It just seemed to me to be more accurate to count the filler eps and the manga eps separately and then merge the results.