| Forum Index > SC2 General |
TheYango
United States47024 Posts
On August 12 2010 11:06 roymarthyup wrote: what i wanna know is why this guy played against 5 diamonds in his placements Its possible that Mewtwo was still in placements as well at the time, and when he was actually placed, it affected InStink's placement accordingly. | ||
|
{ToT}ColmA
Japan3260 Posts
| ||
|
paralleluniverse
4065 Posts
On August 12 2010 15:02 {ToT}ColmA wrote: who cares <: if u wanna compete, play tournaments.... or is it any satisfaction being #1 in a ladder which doesnt mean shit? That's the problem. | ||
|
nextstep
Canada705 Posts
sc2ranks and Kespa rankings are based on numbers, while power ranking usually includes various other factors that might influence a player's rank. | ||
|
paralleluniverse
4065 Posts
On August 12 2010 15:52 nextstep wrote: maybe we can look at sc2ranks as Kespa ranking, and this top 200 thing as Power Rankings. sc2ranks and Kespa rankings are based on numbers, while power ranking usually includes various other factors that might influence a player's rank. Or maybe there should only be a single, official and correct ladder rank, since both are Blizzard's ranks. The problem with 2 different ranks is that they can't both be right. | ||
|
virgozero
Canada412 Posts
On August 12 2010 14:38 paralleluniverse wrote: The ranks are not based on points, Only the in-game ranks are based on points. But the in game ranks are not right, the online ranks are right. okay I seriously cannot make this any simpler. the in game ranks represents points the top200 list represents top200 of na server (in terms of unspecified variables) GET IT??????? read that 2x They both MEAN different stuff. Your assuming they MEAN the same and they don't. | ||
|
paralleluniverse
4065 Posts
On August 12 2010 15:59 virgozero wrote: okay I seriously cannot make this any simpler. the in game ranks represents points the top200 list represents top200 of na server (in terms of unspecified variables) GET IT??????? read that 2x They both MEAN different stuff. Your assuming they MEAN the same and they don't. No, I'm saying they should represent the same thing: the best estimate of who is the better player. If the top 200 represents the top 200, but the in game ranks don't, then the in-game ranks should be changed so that they are capable of ranking who is the better player If the points used to rank in the game are suboptimal in actually ranking who the better players are, then the ladder is a charade, and it needs to be changed to the website ranks because they correctly rank players. | ||
|
kzn
United States1218 Posts
On August 12 2010 14:40 paralleluniverse wrote: Then points should converge to whatever rating is used to derive the top 200. So that ranks based on points are correct, after enough games are played. And so that there's 1 correct, official ladder, not 2 ladders, of which 1 is right, and the other is wrong. Well done, you repeated your conclusion without offering any further argument in support of it. Again: No, it shouldn't, with regards to casuals. | ||
|
paralleluniverse
4065 Posts
On August 12 2010 16:36 kzn wrote: Well done, you repeated your conclusion without offering any further argument in support of it. Again: No, it shouldn't, with regards to casuals. I don't see how this would ostracize casuals. | ||
|
kzn
United States1218 Posts
Bonus pools are mutually exclusive with a rating that is an accurate metric of skill. | ||
|
paralleluniverse
4065 Posts
On August 12 2010 16:41 kzn wrote: Because casuals need a reason to come back and play again, a reason beyond "I want to improve", and the bonus pool provides that reason. Bonus pools are mutually exclusive with a rating that is an accurate metric of skill. I don't see how my suggestion is mutually exclusive with the bonus pool. It is trivial to incorporate the bonus pool into points, it would be trivial to incorporate the bonus pool with this new rating. | ||
|
kzn
United States1218 Posts
On August 12 2010 16:44 paralleluniverse wrote: I don't see how my suggestion is mutually exclusive with the bonus pool. It is trivial to incorporate the bonus pool into points, it would be trivial to incorporate the bonus pool with this new rating. The very idea of a bonus pool is mutually exclusive with ratings "converging" to any point (except, technically, infinity). The only way it wouldn't be is if the bonus pool also applied to rating losses, which would defeat the casual-baiting point of the bonus pool in the first place. | ||
|
paralleluniverse
4065 Posts
On August 12 2010 16:47 kzn wrote: The very idea of a bonus pool is mutually exclusive with ratings "converging" to any point (except, technically, infinity). The only way it wouldn't be is if the bonus pool also applied to rating losses, which would defeat the casual-baiting point of the bonus pool in the first place. Let r be the rating that is used in the ranking of the top 200. Suppose the total bonus pool, P accrued by the d-th day after the start of the ladder season is given by: P(d) = 100 + 12d. Then on the i-th day since the start of the ladder season, your points should converge to r + 100 + 12i. Trivial. This is probably how the ladder already works just with r replaced by MMR. This type of setup would ensure that points in game will continue to inflate endlessly with the bonus pool, yet the ranks based on points would be consistent with the correct method used to rank the top 200. | ||
|
kzn
United States1218 Posts
On August 12 2010 16:53 paralleluniverse wrote: Let r be the rating that is used in the ranking of the top 200. Suppose the total bonus pool, P accrued by the d-th day after the start of the ladder season is given by: P(d) = 100 + 12d. Then on the i-th day since the start of the ladder season, your points should converge to r + 100 + 12i. Trivial. Yes, so over time ratings are converging to an infinite value. So the value of a given point of displayed rating falls over time (which is really precisely whats going on anyway). And this assumes that bonus pools aren't abusable, which is false. Someone who plays 3 games a day is going to see a much more significant boost from bonus pools than someone who plays 300 games a day, regardless of the skill levels of the two players. | ||
|
paralleluniverse
4065 Posts
On August 12 2010 16:56 kzn wrote: Yes, so over time ratings are converging to an infinite value. So the value of a given point of displayed rating falls over time (which is really precisely whats going on anyway). So what? The absolute value doesn't matter for the purpose of ranking, And this assumes that bonus pools aren't abusable, which is false. Someone who plays 3 games a day is going to see a much more significant boost from bonus pools than someone who plays 300 games a day, regardless of the skill levels of the two players. If the bonus pool is abusable go make a post about it. This has nothing to do with whether the bonus pool is abusable. It has to do with 2 inconsistent methods attempting to do the same thing. Your example doesn't show anything. Everyone gets the same bonus pool. Suppose that both players get 12 bonus pool. Then the player who plays 300 games will see, on average, the following change in rating: +24 - 12 + 12 - 12 + 12 - 12 + .... - 12 = 12. The player who plays 4 games will, on average see the following change in rating: + 24 - 12 + 12 - 12 = 12. EDIT: Completely off-topic: The bonus pool is NOT a psychological "bonus" that makes casuals feel better. It's because of the bonus pool that your rank DECREASES every time you log in. Therefore, the bonus pool is as much a penalty as it is a reward. | ||
|
jiabung
United States720 Posts
The Blizzard ranking was probably done in the current state of things, meaning AFTER the players had already played all their games. THEN, they made the top 200 list. You obviously can't replicate this in a ladder system. It is like asking why the Power Rank doesn't match up with the players ELO every time or the Kespa rankings. It's just different arbitrary systems all trying to accomplish the same thing. | ||
|
paralleluniverse
4065 Posts
On August 12 2010 17:33 jiabung wrote: There isn't any ladder system that will perfectly rank the players according to their actual skill level and the points system is merely one method that gives a general roughness of who is better than another. The way Blizzard made the top 200 list obviously takes into account points and rank, but also probably factors in things like win ratio and who you have beaten. There isn't a way to accurately represent this on the ladder because people who play more will obviously have more points. Everyone already knows that more points does not definitively equal more skill, but is only a general indicator of skill level. The Blizzard ranking was probably done in the current state of things, meaning AFTER the players had already played all their games. THEN, they made the top 200 list. You obviously can't replicate this in a ladder system. It is like asking why the Power Rank doesn't match up with the players ELO every time or the Kespa rankings. It's just different arbitrary systems all trying to accomplish the same thing. It's already been shown that at the top of the ladder, there is a negative correlation between games played and points, meaning that if you play more games, its *worse* for your points. So please stop spreading misinformation. "AFTER the players had already played" is a meaningless statement, there is no end to when players play. If they can pull the ranks at the time they did, they can pull the ranks after every game you play. | ||
|
cArn-
Korea (South)824 Posts
On August 12 2010 14:09 paralleluniverse wrote: Everyone is missing the point. It does NOT MATTER HOW THEY CALCULATED THE TOP 200. What matters is that how they calculated the top 200 on the website is DIFFERENT from how rankings are calculated IN THE GAME. Therefore, the RANKINGS IN THE GAME ARE WRONG. This should be fixed. And you really don't seem to get how this system is supposed to work. Given enough time, the ranking blizzard has and the ranking displayed will start to become close, it's just a matter of time so players actually play enough games so the rankings are getting accurate. Of course it'll never be 100% accurate, but I believe it will still be accurate enough so you can take a look at it to have an idea of someone's level. | ||
|
imperator-xy
Germany1377 Posts
being first in your div just tells you how baller you are but you dont really know how good you are | ||
|
DeCoup
Australia1933 Posts
We have no way to see the hidden skill rating which is used for match making. That skill rating is the true ranking system which should be used to rank players. Since we don't have access to it we use points as a crutch to give a ranking. If you win and loose an equal number of games you gain ranking, even without bonus points. With bonus points your points go up even further. This is why the points are not an accurate system for use for total ladder position. The title of the OP thread alone is enough to make me disagree. Clearly blizzard has a better idea of our true ranking than we do. | ||
| ||
StarCraft 2 StarCraft: Brood War Britney Dota 2  39229Jaedong  600Shuttle 478Larva 318Leta  118Hyuk 96Hyun 86Mong 68Shine 29Sexy 25[ Show more ] League of Legends Counter-Strike Other Games Gorgc3641 Grubby2889 Liquid`RaSZi2171 FrodaN1526 Beastyqt717 ceh9351 DeMusliM284 RotterdaM269 Fuzer  242JimRising  235ArmadaUGS210 KnowMe158 B2W.Neo139 Sick106 QueenE94 Mew2King42 UpATreeSC32 |
|
OSC
SOOP
SHIN vs GuMiho
Cure vs Creator
The PondCast
Sparkling Tuna Cup
IPSL
DragOn vs Sziky
Replay Cast
Wardi Open
Monday Night Weeklies
|
|
|
EPT
