LOL, I like your mathematical model...
I think you're right on the mark with the series-position factor --- it seems to me that people check the first and last of a series, and only if they're really appealing will they 'bother' with the in-between pics.
Another factor is the title of the image and/or the 'blurb' that the poster puts with it. If the poster indicates that something particularly exciting is going on in the image, that image tends to be viewed more often.
One thing you have to remember, though, is that this stuff predicts how often the pic will be downloaded but says
nothing about its "quality and appeal" (as you put it). If someone doesn't download the pic, they don't see it (not surprisingly
) and are therefore not in a position to judge its appeal.
Think of it as packaging on breakfast cereals or something: take a kid to the store and they grab the most colourful box, the one that promises the best prizes inside, but that has nothing to do with how good the cereal will taste when they actually get it home.
Personally, I rarely revisit pics --- if there's one I particularly like I'll download it and view it that way. Not everyone does that, though, as a lot of people have to worry about leaving traces on their computers. Those people do revisit pics they like, therefore, which I suppose does give you some kind of indication of appeal; but I think that most pics accumulate more views from people randomly surfing through than from revisitation.
(Just guessing here
)
Any other thoughts on the matter?
--- sweetstuff