07-08-2015
|
110 |
Registered User
Join Date: Jul 2014
Posts: 779
|
When I implemented ranks on Classic, it was a ****ty system (at least based on some code I probably wrote forwarded to me upon request). It was also used for PKs, but I haven't re-analyzed the PK bit as much. The number of spar points relied solely on the rank of the player you defeated and declined logarithmically. It makes sense until you consider that certain cliques spend a lot of time sparring and could run away with the most spar points. I still like the concept of logarithmic point assignments, but think that the gains should be based more on difference in rank between two players rather than solely the rank of the opponent. That way there's diminishing returns for the top ranks and the maintenance of greater returns for lower ranks defeating top ranks. Then again, I'm pretty sure my ranks were a competitive failure, and before I had the code forwarded to me, I thought that this was how I implemented it. Have to ask, how's it implemented these days? |
07-08-2015
|
112 | |
Enguard & Alumni
Join Date: Sep 2011
Posts: 5,773
|
Currently, you gain a number of point's equal to your opponent's ratio multiplied by 10. It's extremely simple. The previous leaderboard system was practically the same as Graal's default rating system. There was some glitch where players could log on to multiple accounts and reset their rating deviation back to /350, spar a few noobs, then reach the #1 spot pretty effortlessly. It basically rewarded players for avoiding the best sparrers, camping after you reach a number of points, and abusing a glitch. I reported it plenty of times, but staff claimed they couldn't fix it. iClassic sparrers aren't competitive and mature enough to let a leaderboard like that flesh out naturally. They will always abuse it and turn it into a contest where the most crafty player who avoids the best sparrers and uses the most tricks wins. It's dumb. The leaderboard many are claiming would be better in this thread is just a flat-out wins leaderboard. Zero measure at all put towards the skill of your opponent. I don't think that they realize that it'd be literally the exact same as the current leaderboard, except without any consideration at all of your opponent's skill. I honestly think people are just bored and looking for something to complain about. In no way is converting a point gain of roughMeasureOfOpponentSkill*10 to a point gain of 1 going to do anything for sparring. No one in this thread seems to be explaining it, just claiming "it's better." I don't understand. Many who seem to support it seem to be the players with the highest amount of all-time wins. Are you surprised? The current leaderboard achieves a lot of things that is healthy for the spar community in general, it just drags out for too long. It naturally encourages people to spar boosters/selective sparrers for their points. It doesn't punish players for losing, so people can queue freely and comfortably. It encourages and rewards players for sparring actively. And guess what? There's never been a mess of glitch abusers or alternate accounts ranked near the top. It should be daily or weekly by default, not seasonally. But at the end of the day, like I said:
|
|
07-08-2015
|
113 |
Dr. Professor Moderator
Join Date: Jun 2015
Location: I’m always located somewhere
Posts: 1,205
|
I was thinking of adding something for losing, like the ratio for winning but in reverse. If you lose to someone good, you lose less points. Lose to someone bad and you lose more points. I know that the part of the point of the new system was to have players not worry about losing, but I kind of feel like there should be some sort of penalty. You could always weight it so the points you lose won't compare, but that could help fix the complaint of there being less skill on top of the leaderboard. Plus no one would go threw the effort of losing hundreds of times on a noob to try and wreck other people's scores. Everyone has already pretty much agreed that the season leaderboard is too long, so I won't bother. Unless a really dominant idea comes out that would far surpass the current system, there won't be any change anyway. |
07-08-2015
|
115 |
Emperor
Join Date: Sep 2011
Location: New York
Posts: 1,000
|
no idea how we're still gonna pretend the current leaderboard "both promotes activity and shows skill" when there are never even close to 20 positive ratios sparring at the same time (activity) and the #3 on the idevice side is nøøb with a 2500-4500 record and no signs of improvement (skill). a simple leaderboard with wins and losses would attract people to spar because they understand that leaderboard and want to be on it (activity) and with enough decent people sparring, hopefully the side rooms will welcome more positive ratios who don't want to wait for the active streak room lines and we can witness the near-death of selective sparrers playing for a record (skill). and if this is implemented and brings enough activity to spar, maybe, just maybe we can bring in that ranking system described in the other thread (even more skill shown). don't see how there could possibly be an issue with any of this. don't pretend a win board wouldn't attract people to the arena. |
07-08-2015
|
116 | |
Registered User
Join Date: Mar 2014
Posts: 888
|
|
|
07-08-2015
|
117 | |
Registered User
Join Date: Jul 2014
Posts: 779
|
Let N be total number of people who have sparred during the season. Let R be the current rank position of the defeated player compared to all the other people. Let S be (N+1)/R; this provides the relative measure of skill for the defeated player. Let P be 100*ln(S); the natural logarithm provides diminishing returns on the points gained due to the opponents' measure of skill; the multiplier of 100 (arbitrary) increases the spread of possible point gains. The people you'd expect to win spars were ranking high. Why I consider it flawed: people with a higher rank can fight one another almost exclusively and boost off each other. Instead, if S were a measure of the comparative level of skill between the defeated opponent and the spar winner, it would make a lot more sense. Aside from that there could be scalability issues in relying on overall rank, which essentially means maintaining an ordered list of all of the sparrers, as opposed to only the top sparrers. Maintaining a full list can become computationally expensive as it grows. The system you described can be gamed in the following way: Spar a bunch of newbs to get your seasonal ratio high, then exclusively spar with other higher-ranked players. You'll get diminishing returns the more you fight each other, but if each players' ratio is boosted then each one will get a higher number of points by sparring amongst themselves as compared to the general population. |
|
07-10-2015
|
119 | |
Dr. Professor Moderator
Join Date: Jun 2015
Location: I’m always located somewhere
Posts: 1,205
|
Now how that list would be decided would be more difficult. Win ratio? Activity? Total points in the previous system? Community consensus?(okay, probably not that one). I just don't know how you could start it off. |
|
07-14-2015
|
120 | |
Registered User
Join Date: Jul 2014
Posts: 779
|
1. Add the winner to the tail of the list if they're not already there. 2. Add the loser to the tail of the list if they're not already there. 3. Assuming there's nobody else in the ranking lists yet: N=2, R=2, S=1.5, 100*ln(S)=41 (rounded) 4. Update the winner's position in the list If nobody else spars, they do round 2, and the previous loser wins: they get 109 points. If they do round 2 and the loser loses again: it's another 41 points for the winner, or 82 total. As long as neither player is dominant, they can flip positions all day long. The major problem would be elitism in selecting spar opponents. If neither player fights player 3, then player 3 will never be able to catch up to them. Perhaps number of opponents would be a good factor to throw in? If there's an upper limit on the number of ranked players: 1. If there are qualification slots, add the winner to the tail of the list. 2. If there are still qualification slots, add the loser to the tail of the list. 1. R=N+(arbitrary) for wins against unranked players. 2. Update the winner's position in the list. |
|