But the Wilson score is not idea. It does not take into account the number of page views. it also does not directly take into account that people are more likely to up vote than down vote. Without evidence, I think down votes should be multiplied by 5.
I am starting to support a ratio of (upVotes - downVotes)/viewCount. I multiply the result by 1000, just to make it easier to reed. Next, I need to figure out the statistically correct version of that algorithm. (It has been a while since I studied statistics at MIT) Any help would be appreciated.
I would like to work with open source projects to create a branch of the tree with all of the best videos for your open source project. Please send me an email if you are interested.