Hello there.

After hearing a particular complaint about how actual ranking system works...
I gave up ranking because it depends only on how much you play, and doesn't rely to much on your skill.
I thought about this new way of ranking up~


As we all know, current ranking depends on the amount of points you gather after playing and beating ranked beatmaps.
The more maps you play, more points will you accumulate and a higher ranking you will achieve.

Simple, and easy.

But what about personal skill? People that play good but dont have the time to play every single beatmap out there?
They should be considered also.

So here is my idea, hope you like it.

Before i start saying how does this idea works, i will have to say this:

Not every player has the same skill lvl, and thats why there are different beatmap difficulties. Make them count.

We will start by dividing the ranking in 3 difficulty sections: (Separate difficulty rank)

Expert - Normal - Easy

This means that players will be able to be the Number 1 in rather Expert, Normal or Easy difficulties.
At the beginning new players will choose what difficulty to play, and will be able to change their decision later on and start
ranking up on a different difficulty if they wish so.
There is going to be a limit of beatmaps needed to play before starting the rank up. (Players will be informed of this when they start playing for the first time)
(You will understand this more after you read the upcoming paragraphs)

Now for the ranking up~

As we all know, every player gets a rank when plays and beats a ranked beatmap. This rank is for that specific beatmap and its assigned depending on the player's performance.

Good, sounds fair. Now...

We can do somthing really interesting with this.

This is:
Calculate the RANKS that players have gotten on EACH beatmap and making a % of them to get one single overall rank for that specific difficulty.
This is the main idea.

Now ill type all the possible questions that you may have, and answer them here:

1.- Theres not only Expert, Normal, and Easy
difficulties out there, but also other personalized difficulties. How will this work?

Remember that the majority of the beatmaps have different difficulties,
but each difficulty has been assigned a number of stars. When dividing the three difficulties for ranking them
separatedly, the stars will play an important roll when choosing which difficulty is rather [/color]Expert, Normal, or Easy.


2.- What if a player plays more than me? Will he get a higher rank?

He may have played more than you, but if your skills are better, he wont likely be higher than you.
Why?
A player may have played lots of beatmaps but his ranks at them were not good at all, so this is where
your skill comes into play. You may have played less, but you may have gotten better ranks,
and that way you got a higher rank over all.
Play count might play its roll later on, when you will have to compete against trully skilled players and the only
way of improving your rank, (if you want to / need so, because you might have a high rank already) will be to play
more beatmaps in order to get a good rank and increase that overall rank %
Example: *Playing 400 beatmaps and ranking 300 or below on them < playing 20 and getting 299 rank or above.*



3.- What happens if a new player plays only 1 beatmap and he ranks 1st? What ranking is he going to get?

Despite being really difficult for a new player to rank 1st on his very first beatmap, this is a good question.
(Because a player can always create a new account or retry his first beatmap till he hits 1st place)

I thought about a way that may help with this issue. (Basically for new players) Its called:


Player lvl

This can be used like a tutorial for new players (a big tutorial).
Asking the players what difficulty would they like to start ranking up with, and then, the tutorial shoots a certain amount of beatmaps they need to pass to reach a certain lvl and start ranking up.
Players will be told that the beatmaps they play will keep their score disabled till they reach that lvl and when they reach it, all of the ranked beatmaps he played (which were "score disabled") will "activate" letting know the player that the Ranking Up has started!. The player will get his rank and will be allowed to retry them for a better rank as usual.
(I repeat, players will be able to decide starting with an specific difficulty lvl but the can also change to a different difficulty and start ranking there, but if they do so, i think it would be nice that another tutorial pops, making them play a specific amount of beatmaps and then give them their ranks for that difficulty.
All of the "tutorial" beatmaps, should be ranked beatmaps that people already played.)




This is my idea of the new rank up system. I know its a huge change, maybe its not the best, maybe its wrong, i dont know, i just thought it up and wanted to post and see what people think.

Good luck all and thx for your patience and attention.

Bye~

hmmm. after reading it, it actually could work, even though there may be someone who could break it somehow.
its a pretty good idea, but i doubt it would replace the current ranking system. maybe it could be implemented as a 2nd ranking system.

Interesting idea. This is definitely going to require a lot of new coding though which might be a little extreme. D=

However, I do understand your point. However, a few questions:

1) Does the actual "meat" of these beatmaps count towards rank? Compare...How about Akai Tsuki to luvotomy. There's a GIANT difference between those two. But if you SS the former, is it worth the same as an SS against luvotomy?

Or rather...That's a silly question.

How about SSing luvotomy once, but then SSing all of those small beatmaps that are worth so tiny amounts of points...

There...

This map has been deleted on the request of its creator. It is no longer available.

This map has been deleted on the request of its creator. It is no longer available.

I think he just forgot about those things. He's just asking for a ranking that's an average of your overall rankings, and dependent on the difficulties you play. I'd say this would be a great second ranking, not a primary one though. Why?

Saturos would be up there again. We're trying to drop him down out of the top 100, this would go against that goal. Though seriously, you can already tell who gets lots of high rankings. I see the same people at the top of many maps, but they aren't necessarily the highest ranks. Great idea though.

Also, keep in mind that for this to work, it would be best to have a map, once ranked, have like 5 days before it's calculated into that sort of ranking, for the rankings to get sorted out.

EDIT: Thought about it. Would make a more competitive ranking overall actually. Instead of FCing a map, then dumping it, you'd strive to improve!

Only flaw, would be not finishing a map because your ranking would go down. There are ways around this though, because your ranking will never go up if you never finish playing a map.

I really don't see this as a good solution. It's not too intuitive and I think a better system can be designed.

Indeed Ekaru.

My bad Peppy.

Wanted to say that if you SS and incredibly hard beatmap, then you wouldnt have to worry about losing 1st place because the beatmap would be so hard that no1 will be able to pass it with any mod.

And Gemi. Yes im sry i mean, i just typed what i thought it could be a solution. Didn't think about it deeply.
It's a little bit difficult to think about another ranking style thats intuitive

And about the split rankings.
Some people may not like it indeed... But i think players that dont really play too good, will find themselves more attracted to the game because they can actually get an average ranking, rather than being like.. 8000+ or somthing on the general rank.

Ekaru <--- as i said before, you can retry a beatmap and pass it again with better score and improve your rank again. Just like everyone does.

I prefer the current one, I know there are many ranked maps and its hard to do them all, but people spend days online play MMORPGs, and i don't mind osu! having this. This idea will keep people competing over the same maps over and over. I see it close to ranking by accuracy.

I am warm to this idea - at least to give it a go. I think it does have merits, but needs some refining.

*readreadreadskim*

You know what I think? You're making things more complicated than they need to be. I guess an alternative ranking balanced around accuracy and the difficulty of the maps you play would be rather neat though.

By the way, I apologize, but I read the general explanation and then once you got to countering potential problems, the first thing to flow through my head was Okay, this is starting to get stupid. You held my interest for a bit and then you just kind of lost it... KISS principle, my friend.

peppy wrote:

Sir Minelli wrote:

Forget about points.
If you SS that incredibly hard song, then you will obtain the 1st place of that one and will be sure that no one will ever take that away from you.

Remember new ranking system doesnt work with points.

???????????????? I thought I understood what you were trying to say until this unlogical reply. SS doesn't mean you will hold 1st place forever. Someone might XS or XH it or get a higher spin score and then they would take your ifrst place.

I think that he means that everything is relative.
For instance, it may be easy to SS something like Akai Tsuki, but then, it's easy for everyone, thereby making it just as hard to get and maintain 1st place as it is a longer song.




On the topic of the thread:
Oddly enough, I was thinking about the exact same thing not too long ago, though it was just some thinking on the concept and I didn't develop the idea as much as you have here.

I think, though, that there should be a system which takes iinto account play counts of each song to "weight" the song based on how many people have played it. For example, a song that has just been ranked has 100 plays, and it should be very simple to get within the top 6 with just an FC, not being careful to SS or high-90% it. However, has time passes, more and more people will try the song, and it's likely that the top ranks will be filled with people looking to get them, adding mods and high accuracy in there, pushing your previous top-6 score into the 40's or 50's. By the time your score is pushed to there, let's say that the song has 2000 plays. This can be seen in songs like Hare Hare Yukai, where the top 50 is dominated by people S'ing the song with Hidden or Hard Rock, where that normaly constitutes only the top few. As playcount for the song increases, as does the difficulty of obtaining a higher rank, and I think the weight of the song in one's overall rank should reflect this.


Also, here's my proposed formula:

a = playcount of song
b = average playcount of all songs
c = total number of rank scores for song
x = achieved rank



Average of all y values for all songs is the ranked score.

In my mind, the best and most obvious way to determine ranking is the way that it is already determined.

Any alterations take away from those who really wish to move up.

Yes, ranking up is dependent on how much time is spent on osu! and the points earned in that time. In my opinion, if rank is important, a player will spend the time to play each and every song, evens ongs the player does not like, in order to rank up.

That is what I did, and I know I'm not alone in this. I've played tons of maps I've hated and spent hours on them just to get the points to move up. People are in the top rankings for two reasons:

1) Skill
2) Dedication

If you can't put the time in, then just accept your rank as it is. That's what I'm doing now. I don't have time to play, and so I'll let my rank fall. It was fun while it lasted, and maybe I'll move up again this summer.

Seems like this thread belongs in the "Gameplay/Rankings" section.

This map has been deleted on the request of its creator. It is no longer available.

I have this idea: A small percent of the ranked mapped gets beyond ranked level (Golded maps) maybe 10%, golded maps would be determined on how popular they are or just between the bats. They would have a total ranked score of there own, or gets more weight than the other score (20x ?), if they maps would be the same all the time or changed on time basis. But this would make people shun off other ranked maps.

tl;dr: Golden maps, 20x score multiplier.

Atmey wrote:

I have this idea: A small percent of the ranked mapped gets beyond ranked level (Golded maps) maybe 10%, golded maps would be determined on how popular they are or just between the bats. They would have a total ranked score of there own, or gets more weight than the other score (20x ?), if they maps would be the same all the time or changed on time basis. But this would make people shun off other ranked maps.

tl;dr: Golden maps, 20x score multiplier.

I don't really understand why this is a good idea.

Main Idea:

Sounds mostly good, though it feels a bit off. I don't like it taking into consideration the ranks you get on all of your songs.

Why? - Because there are some songs that I play just to play. I'll get a low ranking on it, but I don't want to worry about my ranking going down because I didn't try multiple times on that song, or I didn't want to play that song much.

----------------

A system where the harder songs weigh more would be something I'd like to see.

Hmm...

---

Maybe give certain maps difficulty bonuses. A really difficult map worth 5mil might give 20mil worth of ranking points, while an easy map worth 10mil would give just the 10mil....

Then, take a total score based on the top 25 or 50 scores that one person has, and base the ranks off of that.

*shrug*

Finding a balanced way to base it off of skill is hard.

Harder songs already do weigh more. They are 99% of the time worth vastly more points than ordinary songs.

Download: justnobody - Liverne - Necrofantasia [Lesjuh] (2011-07-05) CatchTheBeat.osr

This map has been deleted on the request of its creator. It is no longer available.

I think before doing this the current star-system (for determining a maps difficulty) needs to be tweaked a bit. For one thing, Insane of map A could be significantly harder than the Insane of another map. A star system somehow needs to more accurately gauge the difficulty of each map, then categorize into groups of difficulty. I've played maps that the system gauged as 4.92 difficulty, which were quite easy, but other 3.0 star maps that drove me crazy (wide spacing and fast pacing, despite larger hit circles gives lower difficulty rating, but this is misleading).

I think this ranking idea might be going somewhere, but it needs a lot of refining, and we need a better way of truly measuring how *hard* a map is.

The existing system's main problem is that as more and more maps are ranked, it truly turns into a competition of who has the time to actually play every map. So, an updated system to rank everyone's skill at this game is a great idea, and it's good, I think, that people are thinking about this.

james039 wrote:

I think before doing this the current star-system (for determining a maps difficulty) needs to be tweaked a bit. For one thing, Insane of map A could be significantly harder than the Insane of another map. A star system somehow needs to more accurately gauge the difficulty of each map, then categorize into groups of difficulty. I've played maps that the system gauged as 4.92 difficulty, which were quite easy, but other 3.0 star maps that drove me crazy (wide spacing and fast pacing, despite larger hit circles gives lower difficulty rating, but this is misleading).

I think this ranking idea might be going somewhere, but it needs a lot of refining, and we need a better way of truly measuring how *hard* a map is.

I completely agree here, there really is something odd with the current difficulty measurement. I ran into this on my Office map, I had to raise the HP drain on Insane to make it have a higher star difficulty than Hard, when every aspect of Insane was of higher difficulty (accuracy, circle size, distance snap...)

Back on topic though, maybe a rating of average leaderboard position would work? You could weight it by difficulty of course, but really, I think its almost as hard staying in the top 40 of an easy map as it is for hard maps...

I still completely disagree with the arguments here.

I don't mean to be rude, but.. I still believe that people who put in the time to play every ranked beatmap should and WILL be in the top rankings. CMC and Chocobo may be insanely difficult, but everyone has the same chance on those particular maps. Everyone has equal chances for every single map. Sure, if someone plays more maps, they'll be ranked higher. Because they spent more time.

If you want to pass them, put in the time. Play more maps. And if you've played the same amount of maps, play them better. It's just that simple.

I'm thinking the general consensus is that the new ranking system (if implemented) would be secondary, not a replacement of the current system...

0_o wrote:

I'm thinking the general consensus is that the new ranking system (if implemented) would be secondary, not a replacement of the current system...

I hope that this is really the case. I would be okay with a supplemental ranking system.

ThePassingShadow wrote:

Atmey wrote:

I have this idea: A small percent of the ranked mapped gets beyond ranked level (Golded maps) maybe 10%, golded maps would be determined on how popular they are or just between the bats. They would have a total ranked score of there own, or gets more weight than the other score (20x ?), if they maps would be the same all the time or changed on time basis. But this would make people shun off other ranked maps.

tl;dr: Golden maps, 20x score multiplier.

I don't really understand why this is a good idea.

Raise the gap for too-lazy-to-play-all-maps people. And why does it suck so bad?

This map has been deleted on the request of its creator. It is no longer available.

Ivalset wrote:

I agree with TPS in that people who put in the time to play and score well on every ranked song deserve to be at the top, but at the same time, I do think the scoring system is inherently flawed. There are many maps that I don't think are worth the points they give out.

That's understandable, but everyone knows about those maps and has an equal chance to get the points.

Although I do agree most songs that score 10m+ should be at least as hard as SSI, sometimes they aren't.

Here's my thoughts:

I think that, no matter what, the people who put the time in to play and practice will inevitably always be the ones at the top. Hense, any scoring system that requires (stated or not) the time to play, whether trying to get ranked points or doing better on maps you've already played, will be flawed if those with less time are the ones in mind.

peppy wrote:

I am warm to this idea - at least to give it a go. I think it does have merits, but needs some refining.

^