Popolittz wrote:
Based on what i saw it doesn't seem like rank would be too reliable. So what would be the determining factor? A certain song being completed with a certain accuracy or score? I'd rather not presume, but is there even a definable factor?
Vialle wrote:
8000pp = Top
6000pp = Great
4000pp = Privileged to be a skrub
2000pp = Getting started
500pp = Getting to know
heh.
1. For a more elaborate measurement, you could consider things like variance, the average of the 10 highest ranked plays (or all plays, whatever you prefer) or (for senior players with lots of plays) the median.
Example:
https://osu.ppy.sh/u/6064267 (me :>)
https://osu.ppy.sh/u/MugetsuBeatz#_generalas you can see, Mugetsu has more pp than me. You can also see, I have a higher variance (due to my top-plays being better and his plays being more spread out equally throughout the 50/60/70pp-range).
Considering average pp/median/variance gives you a better idea of where the players skill levels off.
The noob-version would be to just look at the highest ranked play and taking that as benchmark, but thats a little inaccurate imo.
You could just take the players highest play and multiply its pp with 20. That gives you an approximation of how many pp the player would have, if he were to fill up his scoreboard with plays equally to his highest play. (Source: I did the math)
2. Another thing you can do, is to distinguish between skill-categories, such as "speed", "accuracy", "consistency", "stamina" etc., allocating multiple values to one player, then comparing them separately. This does more justice to the complexity of the game and the variety of different playstyles. You can also apply the idea of 1. to the different categories as mentioned in 2.