Well my idea is not about directly playing against each other but it looks like this:Only mods should be counted since playing with mods is considered to be good/skillfull by almost everyone..
The idea is once you full combo a beatmap with a mod you usually get a higher score than players with SS no mod FC. If you done that your skill rating value or ELO would rise or fall by a few points depending on how high you are in the song ranking, only counting mod plays!
Now for an example i just took a random beatmap and looked at the scores..
19 People FC this song with mods so depending on where you are, the amount of points you get is different. In this example DarkRingSystem will get the highest amount, Uan something in the middle, Yuka-san still a bit and Rost94 almost none but still something.
So how do you drop in your skill rating ELO? Well simple, when other users get a higher place in a song than you you would drop down a rank and therefore get less "skill-rating" in that. It should update your total rating then.
Okay that was the main idea.. here are just some more things to notice:The actual value of your "skill" should be displayed as an average of .. i dont know how to say.. just an average of everything you played with mods. similar to accuracy.
This will prevent playing hundreds of maps with mods just for the reason to push your skill rating to a ridiculous high value, and thats the opposite of what a skill based rating system is about.
About the different difficulties in a beatmap set:Easy is easy to play mods and Insane is harder to play mods. Yeah thats right and none will disagree with that but every difficulty should be considered in this system. Since there are also players that love to compete in easy diffs aswell as players that love to compete in insane diffs with mods!
The actual gain of rating there has to increase as the difficulty goes up!
Example: Easy:10% Normal:30% Hard:65% Insane:100%.
The only problem here is that it would be based on the Star Rating of the song i think. Well but i think it is okay even if now people will think something like "Some 5 Star hard diffs are pretty easy to clear with mods though".
In the end it would still be the same since the gain is determined by how many players cleared it with a mod and what their ranking is compared to the others.
So if this diff is really so easy to play alot of people are actually playing it with mods and it is not as rewarding if you are the #30 out of 50 that FC'd it with mods than if you are only one of the of the few that cleared a Insane diff, and you are the #2 out of 8 there. I hope that was not too hard too read/understand because it was hard for me to write and find the words..
All in all it would be looking something like this:
Easy diffs player got a rating of 100
Normal diffs player got a rating of 300
Hard diffs player got a rating of 500
Insane diffs player got a rating of 800
Those numbers are just an example since i didnt try to calculate anything but what is important, is that it shows a clear difference on what you play and ofcourse if you play harder things your rating is higher...
Some more things i thought about:Since older maps are played less and less or almost never anymore they shouldnt be part of this. I dont think anyone wants to backtrack and play beatmaps that dont even fullfill the current ranking criteria. Also some maps are so old that only 1 or 2 players did it with hidden and the rest is just no mods play. You could just DT them and get #1 rankings, but well i dont know if anyone wants to do that... Also they are not fulfilling the standards of what is considered difficult or hard today anymore, at least in my eyes. What was hard 3 years ago is not anymore nowadays, some exceptions.
There should also be a requirement of when to get ranked. If someone just played 1 or 2 maps with mods they shouldnt be counted into this. Probably a requirement of at least 10 or 20 maps until you get valued in this.
Some players play hard or insane diffs most of the time and occasionaly easy and normal (or the other way around). Since it shows an average as a result it would falsify your rating. So only the difficulty level you played the most would count, just to keep the result to the truth as good as possible.
"Some beatmaps are so difficult allready to FC even without mods! Why are they not counted in this?" i thought that too but it just doesnt fit in and theres allways the regular scoreboard in that map for it! Also if you are the one to clear it with a mod.. well theres your reward right there lol (theres usually someone good/ crazy enough to do something like that
)
Since approved maps are often very difficult i thought about them too but meh... there are some harder approved maps and there are some that are just approved because they are either long or got a too high score. So i just left them out in my idea..
In my idea everyone is at a base level and it is impossible to go into the negative. I don't know if thats a bad thing though. You play with mods to set the difficulty for yourself higher anyways so it is okay. If you would be able to drop below the ground level it would be stupid anyways because your rating would be lower that someone that hasnt played with mods ever.
Looking at that screenshot again and since there is a recent discussion about FL... just blank that out for this. Some players will think of the one beeing harder and more difficult in skill than the other. While others think the other way around.
Also last and probably most important: No mod plays are completely left out. Players that play with mods to push them to their limits usually feel superiour to others that play without mods anyways, and they probably are better aswell. Counting no mod plays in would kind of destroy this idea. This doesnt mean you're bad or anything at all if you get S or SS on a song. Its still good but i dont know how nomod play would fit in here (it probably could but then again S or SS ranks arent a problem for the majority of players and i also think it should not be counted, because it is alot harder to play with mods anyways).
Well thats all i thought about yet...
thanks for reading and i hope i didn't give to much trouble with that