|
On January 12 2011 11:57 Impervious wrote:On ICCUP, the distribution was more like an f distribution than a normal distribution. It's probably similar for SC2. If you used that as an assumption instead of a normal distribution, it would show how much of a joke diamond, and it would be even clearer. EDIT - f distribution: ![[image loading]](http://www.visualstatistics.net/statistical%20distributions/F-Distributions%207.jpg)
Yeah, i only posted the iccup information if they wanted to know about a distribution of skill between players. Imagine if 800,000 noobs joined iccup. Then it would be basically the same. The only thing I really care about is that people see the difference between diamond/masters/pro players.
As far as ICCUP goes, their ranking system is terrible and does not accurately reflect the rating distribution. The problem with iccup is the cutoff and how the points system works. The entire system works for itself to keep like 80% of people in D, whereas the one on battle.net breaks people up into divisions of 20%. The distribution of skill is always normal (or assumed to be normal) in any game. The fact that their rating system had a terrible distribution is an example of how bad their system is.
|
On January 12 2011 11:16 LuckyFool wrote:
masters league is a joke I got promoted out of my diamond division that had idra, select, drewbie, tt1 and now I'm with a bunch of nobodys it sucks.
I guess you'll fit right in then with your new friends. Master's is still the microstakes of SC2.
|
On January 13 2011 02:30 darmousseh wrote:Show nested quote +On January 12 2011 11:57 Impervious wrote:On ICCUP, the distribution was more like an f distribution than a normal distribution. It's probably similar for SC2. If you used that as an assumption instead of a normal distribution, it would show how much of a joke diamond, and it would be even clearer. EDIT - f distribution: Yeah, i only posted the iccup information if they wanted to know about a distribution of skill between players. Imagine if 800,000 noobs joined iccup. Then it would be basically the same. The only thing I really care about is that people see the difference between diamond/masters/pro players. As far as ICCUP goes, their ranking system is terrible and does not accurately reflect the rating distribution. The problem with iccup is the cutoff and how the points system works. The entire system works for itself to keep like 80% of people in D, whereas the one on battle.net breaks people up into divisions of 20%. The distribution of skill is always normal (or assumed to be normal) in any game. The fact that their rating system had a terrible distribution is an example of how bad their system is.
i don't get how the system works to keep the majority of people in D. you only need to win 1/3rd of your games to make it to C- since you get -50 for a loss and +100 for a win, and even less than than that if you play only on MOTW.
|
Well the numbers for iCCup D accounts may be skewed because it may include every unplayed smurf or otherwise unplayed account.
|
On January 13 2011 03:56 EchOne wrote:
Well the numbers for iCCup D accounts may be skewed because it may include every unplayed smurf or otherwise unplayed account.
In the thread it explicitly says that it did not include unplayed accounts.
|
On January 13 2011 03:27 rauk wrote:Show nested quote +On January 13 2011 02:30 darmousseh wrote:On January 12 2011 11:57 Impervious wrote:On ICCUP, the distribution was more like an f distribution than a normal distribution. It's probably similar for SC2. If you used that as an assumption instead of a normal distribution, it would show how much of a joke diamond, and it would be even clearer. EDIT - f distribution: Yeah, i only posted the iccup information if they wanted to know about a distribution of skill between players. Imagine if 800,000 noobs joined iccup. Then it would be basically the same. The only thing I really care about is that people see the difference between diamond/masters/pro players. As far as ICCUP goes, their ranking system is terrible and does not accurately reflect the rating distribution. The problem with iccup is the cutoff and how the points system works. The entire system works for itself to keep like 80% of people in D, whereas the one on battle.net breaks people up into divisions of 20%. The distribution of skill is always normal (or assumed to be normal) in any game. The fact that their rating system had a terrible distribution is an example of how bad their system is. i don't get how the system works to keep the majority of people in D. you only need to win 1/3rd of your games to make it to C- since you get -50 for a loss and +100 for a win, and even less than than that if you play only on MOTW.
Simple - the frequent resets, coupled with the large number of people who simply don't play a lot of games, and people who use a few different accounts, and you have a disproportionately high number of "D" accounts. Even if they are active accounts.
|
On January 13 2011 05:36 Impervious wrote:Show nested quote +On January 13 2011 03:27 rauk wrote:On January 13 2011 02:30 darmousseh wrote:On January 12 2011 11:57 Impervious wrote:On ICCUP, the distribution was more like an f distribution than a normal distribution. It's probably similar for SC2. If you used that as an assumption instead of a normal distribution, it would show how much of a joke diamond, and it would be even clearer. EDIT - f distribution: Yeah, i only posted the iccup information if they wanted to know about a distribution of skill between players. Imagine if 800,000 noobs joined iccup. Then it would be basically the same. The only thing I really care about is that people see the difference between diamond/masters/pro players. As far as ICCUP goes, their ranking system is terrible and does not accurately reflect the rating distribution. The problem with iccup is the cutoff and how the points system works. The entire system works for itself to keep like 80% of people in D, whereas the one on battle.net breaks people up into divisions of 20%. The distribution of skill is always normal (or assumed to be normal) in any game. The fact that their rating system had a terrible distribution is an example of how bad their system is. i don't get how the system works to keep the majority of people in D. you only need to win 1/3rd of your games to make it to C- since you get -50 for a loss and +100 for a win, and even less than than that if you play only on MOTW. Simple - the frequent resets, coupled with the large number of people who simply don't play a lot of games, and people who use a few different accounts, and you have a disproportionately high number of "D" accounts. Even if they are active accounts.
that's hardly how the system itself is actively attempting to keep players in D.
|
masters league is top 2%???
and really this all just depends on what your definition of baller is doesn't it. We're also assuming normal distribution which is a bit dodgy but if we take it as correct the numbers are useful =]
To put this another way, regardless of what distribution we're actually looking up, we're all tightly wrapped up here in our ball of ignorance communicating with other people who are ACTUALLY GOOD AT THE FRIGGING GAME. Masters league is still what it says on the tin, it is the top 2% of ACTIVE MULTIPLAYER players in your region. Being a player on starcraft who is active in multiplayer already puts you above a lot of people who own starcraft and the majority of the general population (not that that's relevant.) Being in the top skilled 2% of THAT is a further acheivement, whether it makes you a baller or not is down to discretion; here? (on teamliquid) probably not. Socialising with your RL friends that play starcraft? nearly definetely. In your *insert nerdy subject* lecture group? Absolutely.
|
On January 12 2011 07:46 darmousseh wrote:
Statistically speaking master's league represents 1% of all players.
...
So before saying stuff like "I'm a 2500 master" consider the possibility that statistically speaking, you are still really bad compared to the pros.
...
Yeah wow, you really hit the nail! Nobody has been sucessful in stopping players from posting their rank but surely this time you have brought down forum justice and hard as well,
I guess I just didn't realize that even tough I've played alot and owned loads of people, I am still really bad compared to the pros. Man I should kill myself how could I not realize this!! I thought I was the zerg master god of an entire galaxy in space but no, i'm really bad. Compared to the pros.
Really, you are correct because since grand master is so giga-large, it serves no purpose! There really should be a single noob league, and when you move to Korea and are at 20.000 victorious glorious games against progamers, you will get promoted to pro league. It's the only way.
|
On January 13 2011 06:23 rauk wrote:Show nested quote +On January 13 2011 05:36 Impervious wrote:On January 13 2011 03:27 rauk wrote:On January 13 2011 02:30 darmousseh wrote:On January 12 2011 11:57 Impervious wrote:On ICCUP, the distribution was more like an f distribution than a normal distribution. It's probably similar for SC2. If you used that as an assumption instead of a normal distribution, it would show how much of a joke diamond, and it would be even clearer. EDIT - f distribution: Yeah, i only posted the iccup information if they wanted to know about a distribution of skill between players. Imagine if 800,000 noobs joined iccup. Then it would be basically the same. The only thing I really care about is that people see the difference between diamond/masters/pro players. As far as ICCUP goes, their ranking system is terrible and does not accurately reflect the rating distribution. The problem with iccup is the cutoff and how the points system works. The entire system works for itself to keep like 80% of people in D, whereas the one on battle.net breaks people up into divisions of 20%. The distribution of skill is always normal (or assumed to be normal) in any game. The fact that their rating system had a terrible distribution is an example of how bad their system is. i don't get how the system works to keep the majority of people in D. you only need to win 1/3rd of your games to make it to C- since you get -50 for a loss and +100 for a win, and even less than than that if you play only on MOTW. Simple - the frequent resets, coupled with the large number of people who simply don't play a lot of games, and people who use a few different accounts, and you have a disproportionately high number of "D" accounts. Even if they are active accounts. that's hardly how the system itself is actively attempting to keep players in D.
The players themselves are part of the system..... A ladder cannot exist without players.....
If everyone played 1k+ games in a season on every one of their accounts, there'd be a shitload of people in the blue ranks. A basic understanding of how the ladder works would lead you to that conclusion.
But they don't.
You can make multiple accounts to spread your games around, you can clear your stats to start at the bottom again, and you don't have to play 1k+ games either..... Those rules/abilities help leads to the system having a lot of D players.
|
On January 14 2011 00:20 osten wrote:Show nested quote +On January 12 2011 07:46 darmousseh wrote:
Statistically speaking master's league represents 1% of all players.
...
So before saying stuff like "I'm a 2500 master" consider the possibility that statistically speaking, you are still really bad compared to the pros.
... Yeah wow, you really hit the nail! Nobody has been sucessful in stopping players from posting their rank but surely this time you have brought down forum justice and hard as well, I guess I just didn't realize that even tough I've played alot and owned loads of people, I am still really bad compared to the pros. Man I should kill myself how could I not realize this!! I thought I was the zerg master god of an entire galaxy in space but no, i'm really bad. Compared to the pros. Really, you are correct because since grand master is so giga-large, it serves no purpose! There really should be a single noob league, and when you move to Korea and are at 20.000 victorious glorious games against progamers, you will get promoted to pro league. It's the only way.
hahaha, the sarcasm is well received. Out of personal experience though there are plenty of people who think they are "the bomb" or "baller" or any other number of phrases used to describe something but in reality it's not very good. Here's a good example.
Let's say there are 40 million americans who have played baseball at some point in their life.
The major leagues has 1280 players
The minors has roughly 1500 players
2% of 40 million is 800,000. Are those 800,000 people baller? No, are they good? Yes. Will they ever make the majors? maybe a few hundred of them at most.
The problem exists that people who are directly below semi-pro think they are baller. There are men's baseball leagues that exist and they all think they are good enough to be pro. The same thing happens in sc2. There are tons of people who are in masters league now and they will use that as an excuse to be BM and say dumb things like "Terrans basically already exhausted all of the options for attempting macro management play...to no avail." without having competed at the top level. If a minor league player said something like "the strike zone is too large" would anyone take him seriously? No.
Everything is about perspective and yes I think being in masters is a huge accomplishment. I hope to be in masters and realize the amount of practice it will take to get there, but for me it is a personal accomplishment, I wouldn't go spewing out on the forums my achievement and write terrible strategy ideas.
Note: I need to revise my graph since they upped it from 1% to 2%. Will be doing that.
|
United Kingdom3685 Posts
On January 14 2011 00:20 osten wrote:
There really should be a single noob league, and when you move to Korea and are at 20.000 victorious glorious games against progamers, you will get promoted to pro league. It's the only way.
I realize your post was sarcastic, but in all honesty, this might be a good idea. Especially the 'glorious' part. Winning with 4-gate isn't glorious and therefore doesn't count towards the 20,000 victories.
|
Ok I updated the graph from 1% to 2% and that really widened the league by over 25% of the std dev (or about 100 points). Masters league is a lot bigger that I thought.
|
lol seeing the stats, when i got a stats exam 2morrow better get revising. -_-
|
Still looks like a bunch of sour grapes.
Honestly if I were in master league I'd be feeling pretty fucking badass. Calling the top 1% "terrible" because they can't hold a candle to people who play 6 hours a day every day for money is ridiculous.
|
Using arbitrary numbers proves nothing.
I can tell you that you can get into masters league with less than 2.3k points, it just depends on your ell
for example, I know of an account that went from roughly 30-20 win-loss ratio to 80-30, which boosted the ell enough to start regularly facing top 200 players. This account had a very low point total, but a high ELL.
Obviously its not special to be in masters league, if it really is 2%, of the TOTAL, that just means theres a fuckton of players on the ladder throughout all the divisions.
You used a few too many assumptions in the numbers if you ask me, and quite frankly, I know where I stand in SC2 and I'm not going to say, yeah I can kick *Insert Progamer here* in a BO9 just because of the fact I'm in masters league, rather, I'll say I can beat them if I fucking match up and face them and beat them on the ladder.
On January 12 2011 11:40 Raeleigh wrote:Show nested quote +On January 12 2011 11:16 LuckyFool wrote:
masters league is a joke I got promoted out of my diamond division that had idra, select, drewbie, tt1 and now I'm with a bunch of nobodys it sucks. They were all promoted to Masters League too most likely.. Just because you're not with the same people doesn't make it a joke. That doesn't even make sense.
Thats not his point, his point is that going from a diamond league where you could actually compare yourself was nice because thats the only ladder division you'll regularly see. That and the top 200 listings which goes by players ELL meaning he'll regularly face people in that division or people in the top 200.
Now he gets moved into masters league and its filled with 85 nobodies who won't be in the top 200 and will never ladder against him because the matchmaking won't make up that skill differential so easily. When there are 85+ of those people in the ladder, it means that the basis of comparison is noticeably worse.
Amirite Luckyfool?
On January 14 2011 06:04 Newbistic wrote:
Still looks like a bunch of sour grapes.
Honestly if I were in master league I'd be feeling pretty fucking badass. Calling the top 1% "terrible" because they can't hold a candle to people who play 6 hours a day every day for money is ridiculous.
I don't think that the OP was calling masters league terrible, if they were then whats it matter, the odds (see what I did there) of the OP being in masters league is pretty fucking low. They know some statistics and finds it growing of thy holy e-peen by saying that the top 2% is bad. Which is kinda ironic, since being in the top 2% still means you're better than 98% of the fucking population of SC2 lol, obviously theres a lot of people who are bad, but being in 2% is still meaning you get 99 people you're better than 98 of them.
|
On January 14 2011 07:30 ZlaSHeR wrote:Using arbitrary numbers proves nothing. I can tell you that you can get into masters league with less than 2.3k points, it just depends on your ell for example, I know of an account that went from roughly 30-20 win-loss ratio to 80-30, which boosted the ell enough to start regularly facing top 200 players. This account had a very low point total, but a high ELL. Obviously its not special to be in masters league, if it really is 2%, of the TOTAL, that just means theres a fuckton of players on the ladder throughout all the divisions. You used a few too many assumptions in the numbers if you ask me, and quite frankly, I know where I stand in SC2 and I'm not going to say, yeah I can kick *Insert Progamer here* in a BO9 just because of the fact I'm in masters league, rather, I'll say I can beat them if I fucking match up and face them and beat them on the ladder. Show nested quote +On January 12 2011 11:40 Raeleigh wrote:On January 12 2011 11:16 LuckyFool wrote:
masters league is a joke I got promoted out of my diamond division that had idra, select, drewbie, tt1 and now I'm with a bunch of nobodys it sucks. They were all promoted to Masters League too most likely.. Just because you're not with the same people doesn't make it a joke. That doesn't even make sense. Thats not his point, his point is that going from a diamond league where you could actually compare yourself was nice because thats the only ladder division you'll regularly see. That and the top 200 listings which goes by players ELL meaning he'll regularly face people in that division or people in the top 200. Now he gets moved into masters league and its filled with 85 nobodies who won't be in the top 200 and will never ladder against him because the matchmaking won't make up that skill differential so easily. When there are 85+ of those people in the ladder, it means that the basis of comparison is noticeably worse. Amirite Luckyfool? Show nested quote +On January 14 2011 06:04 Newbistic wrote:
Still looks like a bunch of sour grapes.
Honestly if I were in master league I'd be feeling pretty fucking badass. Calling the top 1% "terrible" because they can't hold a candle to people who play 6 hours a day every day for money is ridiculous. I don't think that the OP was calling masters league terrible, if they were then whats it matter, the odds (see what I did there) of the OP being in masters league is pretty fucking low. They know some statistics and finds it growing of thy holy e-peen by saying that the top 2% is bad. Which is kinda ironic, since being in the top 2% still means you're better than 98% of the fucking population of SC2 lol, obviously theres a lot of people who are bad, but being in 2% is still meaning you get 99 people you're better than 98 of them.
Actually I make no assumptions that the ladder rating system doesn't already make which is a somewhat normal distribution of players. The only question that remains is how much better the top 2% is compared to the other 98%.
I haven't played any games since the patch came out, i was in diamond with 1600 bonus points and a 800 rating so I have no idea if i'm good enough, but most likely not. I would love to be in masters league and would definitely be proud of my accomplishment, but I would still think I was really bad. I have a somewhat negative view of things, but I consider anyone not in the top 10 of the region bad. So far there are few if any players who can really play the game effectively. Everyone else seems to make a ton of mistakes that cost them the game. Eventually I think players will get to the point where they only lose by the strategic decisions they make because of their amazing execution, but that time is a long way away.
As far as I'm aware, there is no division offset so as long as you have no bonus pool and your points has reached the level of your MMR, then you can compare, but if you don't mass hundreds of games, the system is still completely obscure.
In the end, 2% is still a HUGE range of skill. 2% of players in baseball would be the difference between alex rodriguez and a guy who plays in a sunday league for the local city. I think we should cheer for each other's accomplishments and that everyone should strive for it, but like all competitive sports, everyone needs to keep perspective. If you are in masters league and you are trying to qualify for the TSL and you get crushed, it's because of the difference in skill. The amount of practice needed to get you from the top 2% to the top 200 is insane.
So yeah, like i said before, be proud to be in masters, but don't make terribad threads about strategies claiming authority because you are in masters.
|
While I can agree, Most people will use master league as a way to "epeen" alot about there prowess as a baller sc2 player. However an achievement is still an anchievement and you gotta realize there are people who are still unable to break out of silver/gold/plat and they make up the majority of the gamers out there. Same thing with chess. There are ALOT of 1500ish chess players. Im so they would feel good having made it to 2.3k after months of practice if not in the case of chess years. Why bother making these statements people who want to use it as a dick measuring stick will and will do so without your permission. Let the dog have his day 
|
Pretty interesting post, I was curious about the numbers and you represented it clearly here. I still think it's pretty baller to get into Master though. You don't have to be the best of the best to be baller. 
|
Hehe, I like your style darmousseh, could you give me the number at which a guy turns 'baller', and also I would like to know at which point that guy can say stupid crazy stuff like "Terrans basically already exhausted all of the options for attempting macro management play...to no avail." and it is considered valid?
|
|
|
|
|
|
EPT
