abraker wrote:
Current skill algorithm are not equipped to deal with this for the same reasons why some skills are over weighted or under weighted
Personally I just look at the rankings instead of the score. Why worry about weight when you already supply the numbers you need? Makes everything look pretty and diverse.
On the left are the title updates and on the right are the titles shown on the website, how gorgeous is that? All I did was use the rankings and calculated that below (sorry for the mess). I'd love to describe the numbers but frankly I just kinda went into the zen mode and crapped it out, but it came out nice. But as far as I can tell, anything under 80% is thrown away, anything over 90% sticks, and in-between I have it compared to the median.
This project is an absolute dream to a stats-drooler like myself, and if there ever releases one for Taiko, I would probably make brown. If I could, I'd love to assist in any way possible.
Edit: Original scores because why the why not. Ignore the purple and the gray text since I made this for friends, but the ones that are italicized are our highest based on the rankings, and the ones that are blue are our lowest.