|
On January 19 2011 17:58 teamsolid wrote:Show nested quote +On January 19 2011 17:56 wherebugsgo wrote:On January 19 2011 17:52 teamsolid wrote:On January 19 2011 16:41 iEchoic wrote:On January 19 2011 15:12 IdrA wrote:On January 19 2011 14:49 branflakes14 wrote:
The GSL proves that when the game is pushed, it's fairly balanced. A bunch of Terran and Protoss wins at sub-Code A tournaments isn't evidence of imbalance. ... TWO zergs earned a spot in the ro16 and we both had relatively easy ro32 groups Clearly you feel that this is a result of imbalance. What specific matches so far do you think were won by the lesser skilled player, with the Z player being eliminated due to imbalance? (this question is for anyone else, as well, since I doubt you're reading TL an hour before your match) I don't think there have been any big upsets this whole season so far, except for Fruitdealer and Maka's elimination. MC might have been a favorite to win, but I don't honestly consider it a huge upset because MK and Jinro are also sick good. Imbalances that exist if any, are probably more subtle than just straight up "I lost because I played Z". Also, there are now far less BitByBitPrime's in Code S. Fruitdealer and Maka losing were not upsets. They just played badly. That's like saying Flash getting eliminated from the MSL/OSL was not an upset, he just played badly. Yes, that might be true, but it's still an upset.
Flash has played BW for years, and has been top 3 in the world easily since 2008. His emergence onto the scene (by cheesing Bisu) was well over 3 years ago, in September 2007. The comparison does not exist. Maka has not won anything, and there are tons of players better than Maka. Flash's only true competition in the last few years have been Jaedong, Stork, and Bisu.
Fruitdealer, on the other hand, has been shown to be inconsistent. He won the first ever global tournament of SC2, sure, but he has not done anything significant since then, which was four long months ago. It was not an upset for him to lose in this GSL. It was an upset when he lost to Foxer in RO32 season 2, sure, but not this time around.
|
I didn't have time to read the whole thread, but in the part that I read there was one answer or aspect kind of missing or just represented through discussion about the map-pool. From my personal experience and watching lots of streams I'd make an argument that points in the same direction, but isn't bound only to the map pool, though it surely is one of the most contributing factors.
My feeling is that Z is the... most shaky of the the three races. I'm far away from thinking they are underpowered, but the margin for error seems a lot smaller than those for the other races. Mistakes usually do not only have more painful consequences but are harder to recover from. This stems from the different mechanics of the three races - the drone-mechanic is great, but often a walk on the tightrope as well, over-reacting in any direction is usually deadly, while this balance the Z player has to search anew in every game is basically build into the other two races.
Another mechanic is the low cost-efficiency, micro-bility and harass-incapability of Z units. When in a dire situation careful leap-frogging, dropplay, forcefield- or blink or marinemicro can easily save the day. The Z units are rather blunt and don't really allow for fancy maneuver that pull you back into the game.
This idea especially makes sense because it could be an answer to why terrans perform so extremly strong (even compared to protoss). Good Terrans have awesome defenses, the mule and very strong single units that can make huge differences even in small numbers (marines, medivac drops, blue flame hellions, banshees, tanks) to fall back onto which give them possibilites to recover from former errors. This doesn't mean they are overpowered, the race just allows for a more stable gameplay.
This kind of sounds like a balance-rant, but isn't ment to be one. But besides other facts that have been pointed out already (map-pool - its just awful that cups like go4sc2 or zotac still force you to play maps like jungle basin, delta and steppes. Minority of high caliber foreign z players that participate in those tourneys) I think it contributes to the weak perfomance of Z players that the race just is not build for comebacks. Especially in small tourneys where the pre-rounds have to be played on awful maps in a BO-1 format.
|
On January 19 2011 12:09 iEchoic wrote:Show nested quote +But the claim that Zerg is played less than Terran and Protoss is, in America and EU, false http://sc2ranks.com/stats/league/all/1/allWhat are you talking about? Zerg is the least-played in every single league. What I said was 100% correct. I don't care to argue how much less it is or in what region or what league or blah blah blah. What I said wasn't false.
That's a great standpoint to have... if you don't care then why are you posting.
Based on the statistics that the op posted 8% of tournaments are won by zerg, 25% toss, and 63% Terran. These statistics (depending on accuracy of course) show that Z is up simply because we know that Z is more than 8% of the tournament population. Using the 85 tournaments that were in this statistic I would like to see the top 248, top 16 and top 4 in addition to these winners so we know how many of each race started the tournament and when in the tournament we saw drop offs (this would really just show the percentages of the lesser skilled players race choice). I would almost guarantee that in the top 248 there is going to be more than 8% Z and in the top 16 as well (8% of 16 is like 1.28 per tourney) whereas it is within realm of reason that there aren't any Z in the top 4 (or any race for that matter).
While Z is the least played race, it is not so underrepresented that 8% is not worth noting.
I actually don't think that Z is up, because proving up/op is extremely hard to do (glaring things like reaper rush's and feeling completely helpless prior to patch 1.1 aside). I think that with a better map pool we will see where the real balance is at.
|
To the guy calling shenanigans on this OP because 2 zergs have won GSL:
GSL4 Code A Final 4: T, T, T, T
GSL4 Code S Final 4: Z, T, T, T
You still want to argue that terrans aren't crushing everyone in the GSL?
|
These stats are meaningless. They ignore the most important factor, the person behind the keyboard.
|
On January 20 2011 11:47 RiB wrote:
These stats are meaningless. They ignore the most important factor, the person behind the keyboard.
Following this logic you'd had to declare every statistic ever made to be meaningless -_-
|
we need more good maps imo, bigger maps = more fun games and no less cheesy allins
|
On January 20 2011 11:47 RiB wrote:
These stats are meaningless. They ignore the most important factor, the person behind the keyboard.
So acording to you, the fact that in every tournament top8, +50% are terrans, is just because terran players are superior??
Any person with some knowledge in math statistics knows that an analize like that really means something.
This is the thing. Terrans players are NOT superior human beings skill wise....there is something that makes them superior in the game.
|
On January 20 2011 07:46 jdseemoreglass wrote:Show nested quote +On January 20 2011 07:26 Raid wrote:
I can name a lot of talented terran players but only few protoss and zerg... Maybe that is why.. I don't understand why people have to bring this up when terran hasn't even won a gsl yet. I don't quite understand why everyone throws away TONS of evidence regarding something because of a few games played by a handful of pros... That is all that GSL is: a very small sample of games between top players. How can that negate all the evidence? What evidence? The stats from tournaments since the game was released, the stats from TL's player index, the stats from Blizzard's countless Top 200's, the statements from the pro-gamers themselves from ALL the races... All these facts are pointing to the same conclusion. A conclusion that everyone wants to reject because of a handful of games between a handful of players. Do you want links to both Korean and NA polls stating the game is imbalanced? Do you want links to the comments from players like IdrA, FruitDealer, Artosis, Dimaga, Morrow, Drewbie, TLO, Sen, Sheth, MasterAsia, Psy, and countless others? Do you want the stats from sc2ranks.com? All the evidence is there. But many people like to ignore the flood of evidence in favor of the exceptions, to make the tail wag the dog. The GSL's do not dissprove anything, especially when the most recent results have Terran's completely dominating the GSL with 87.5% of the Ro8 being Terran. And no, this can't be explained by more Terran players. Obviously Terran's do not come close to making up 87.5% of either players or professionals. I wish people would just admit they don't give a damn about the evidence, they just don't WANT to accept imbalance so that they can accuse those who do of being whiny, poor players.
The thread should really end with jdseemoreglass, who very concisely summed up the attitude of many of the people now entering. Could we get a sticky on a generalized version of this, so maybe people will understand better how to use statistics as evidence?
|
On January 20 2011 05:13 MorroW wrote:
just like i said some months ago. statistics will get abit silly when u add in the foreigner scene because we dont have many good zerg players who compete in tournaments
also zerg has been a very weak race for a long period of time. its clearly not as imba as it used to be
im not really sure why you keep saying that
me ret dimaga haypro you machine sheth *should* easily be as good and accomplished as the group of players playing terran/protoss
the reason zerg seems underrepresented is because so many new players have risen up and had success, to varying degrees, with t and to a lesser extent with p. the fact that essentially no one has done that with z says something in itself. but to say that the people playing z, especially outside of korea, dont deserve to be winning is absurd.
|
The IEM racial distribution is just depressing to see
|
To be fair, a lot of those are from a time where the strategies and patches are now considered obsolete.
Lets do some math on the OP's statistics, shall we? Lets assume that all 3 races have an equal chance of winning a tournament. This is a binary distribution with each race having a 1/3 chance of winning a tournament, ideally. The likelihood of any given outcome where a race, in this case zerg, wins only 7 tournaments can be expressed as
Zerg: (1/3)^7*(2/3)^(85-7=78) = 8.4145 * 10^-18. Of course there are 85C7=4,935,847,320 ways this could occur. The probability of this event is .000000042. If all possibilities were equal one would expect an average probability of 1/85 = .01176. In short, it is extremely unlikely that this would happen by chance.
Terran: Now what is the probability of a race winning 54/84 games by chance. The likelihood of any situation where a race wins 54/85 games is (1/3)^54*(2/3)^(85-54=31)= 5.97892*10^-32. There are 85C54=1.48409980330845*10^23 possible outcomes with this result. The final probability of this happening is .000000009 compared to an expected average of .01176.
Skill might add some randomness, but i doubt one could argue that Zerg players are that skilless in general. The same could be said about Terran, given that a significant portion of sucessful Terran players don't do much besides SCV all ins and bunker rushes. How can one argue that the races are equal when the probability of Zerg winning only 7 games by chance is less than 1/10,000 the expected average probability. Less than 1/100,000 in the case of Terran.
Edit: As it's been pointed out. In the masters league the distribution of the races is close to equal.
|
Another mechanic is the low cost-efficiency, micro-bility and harass-incapability of Z units. When in a dire situation careful leap-frogging, dropplay, forcefield- or blink or marinemicro can easily save the day. The Z units are rather blunt and don't really allow for fancy maneuver that pull you back into the game.
Low cost efficiency? Absolutely.
Micro-bility? Through the roof.
Harass-incapability? Mutalisks and baneling drops would like to have a word with you.
Leap-frogging is a necessity. Same goes for marine micro, forcefields, and blinks. You can compare these to flanking with zerg, or using FG.
|
One of the worst things to happen to balance discussion has been the 2 zergs winning the GSL.
I don't know how anyone, in a level headed non-emotional balance discussion can point to any single victory by any race to represent balance from one tournament. I'm sorry, but when the finals is 3 Terran and 1 Zerg that to me is much more telling of balance, than if 1 Zerg manages to win it.
First, you can pretty much at this point throw away ANY statistics involving GSL 1. The game is so far removed from what we knew, and how it was played, plus we've gone through 2 patches(although not major they've still change the game enough) that really using it to show anything is pretty pointless. I immediately dismiss anyone's logic as soon as they bring up FD winning GSL 1.
I think you can reasonably go back 1 patch as this last patch hasn't change anything significantly, especially the zvt aspect.
I really don't see how anyone, who takes a non-bias look at the current state of the game doesn't see that there is something pretty off and in favor of terran at the moment. And this could shift with new strategies for sure. But as it is right now, with the current patch and current meta game terran is just rolling.
|
I don't know how anyone, in a level headed non-emotional balance discussion can point to any single victory by any race to represent balance from one tournament. I'm sorry, but when the finals is 3 Terran and 1 Zerg that to me is much more telling of balance, than if 1 Zerg manages to win it.
Yes, one single victory won nestea the second GSL. He definately didn't have to push through his qualifying brackets, and tournament brackets to get there. Same with MC. They both just played the finals.
First, you can pretty much at this point throw away ANY statistics involving GSL 1. The game is so far removed from what we knew, and how it was played, plus we've gone through 2 patches(although not major they've still change the game enough) that really using it to show anything is pretty pointless. I immediately dismiss anyone's logic as soon as they bring up FD winning GSL 1.
Illogical to dismiss an argument based on one small part of an opinion, but I agree with you, in that nobody can pull any analysis from GSL 1.
I really don't see how anyone, who takes a non-bias look at the current state of the game doesn't see that there is something pretty off and in favor of terran at the moment. And this could shift with new strategies for sure. But as it is right now, with the current patch and current meta game terran is just rolling.
And what should we do? Ask for patches, or try to change the meta game in order to compete? I'd vote for option two. It might take longer, or even be a futile effort, but I think it's a better choice in the long run. Terran had to deal with the same thing in BW.
|
Unless youve seen all these games you cant judge balance. Im sorry.
|
Any balance discussion is inherently flawed since you'd have to assume:
1.) The skill level of all players are the same, only the race is different which is obviously false. Even in the GSL there is a HUGE range of skill level between the top 8 and the rest. There are even bigger differences in skill in other tournaments.
2.) There is an even number of Zergs, Toss, and Terrans in each tournament. Also false, the numbers will vary per tournament but all of them combined don't have a 33/33/33 split.
3.) Patches have affected balance alot, Roach range was huge along with other changes, so old tournament data isn't as valid, thus reducing the amount of data we have.
4.) The map pool and map voting procedures are the same in every tournament. Which of course is false, it has changed alot within the GSL and every tournmanet has their own procedure.
The basic of statistics is that if you start with bad data you will get bad results. The data we have is terrible at the moment.
|
On January 21 2011 03:01 Leviwtf wrote:
Any balance discussion is inherently flawed since you'd have to assume:
1.) The skill level of all players are the same, only the race is different which is obviously false. Even in the GSL there is a HUGE range of skill level between the top 8 and the rest. There are even bigger differences in skill in other tournaments.
2.) There is an even number of Zergs, Toss, and Terrans in each tournament. Also false, the numbers will vary per tournament but all of them combined don't have a 33/33/33 split.
3.) Patches have affected balance alot, Roach range was huge along with other changes, so old tournament data isn't as valid, thus reducing the amount of data we have.
4.) The map pool and map voting procedures are the same in every tournament. Which of course is false, it has changed alot within the GSL and every tournmanet has their own procedure.
The basic of statistics is that if you start with bad data you will get bad results. The data we have is terrible at the moment.
No one is trying to scientifically prove anything here. The point is, if you look at tournament results, ladder results, top 200 results, community polls and pro-gamer comments, you see a general trend clearly exists below the number.
I created a thread just now discussing the problem inherent in point 1) you made. Here is the link.
Game Balance and The Uncertainty Principle
|
And what should we do? Ask for patches, or try to change the meta game in order to compete? I'd vote for option two. It might take longer, or even be a futile effort, but I think it's a better choice in the long run. Terran had to deal with the same thing in BW.
I agree. I never said anything had to be change right now or that it wouldn't change over time. However if you take a snapshot of the current game I find it hard to argue that there isn't a balance in the favor of terran. Not only that, but the race continues to improve simply because the current design allows for so many more options. The game to me has always felt like terran is 95% complete, and the other 2 races are only 75% complete and unfinished. Of course that doesn't mean a new strategy or build or idea doesn't pop up which can cause a shift in that balance. I have to imagine the expansions will help z and p in terms of options and units more so than terran.
The question is does blizzard just sit back and give it time and let the players figure out how to balance the game out, which they will eventually, or do they try to do something to help speed that process up some with patches.
Personally, I like to look at tournaments from the RO8. That to me is the point where more often then not the players that are there could win the tournament as long as they get some good luck and don't just play horrible. It is pretty rare these days to look at a tournament at that point and not have it be like 5 terran, 2 toss and a zerg, or 6 terran 2 toss.
The best argument people have for Zerg being fine is basically the 2 GSL winners. Anytime I see someone try to say Zerg is fine they always go to that point. That point, which held little weight in the first place, gets more and more meaningless each passing day.
|
On January 21 2011 01:51 Duban wrote:To be fair, a lot of those are from a time where the strategies and patches are now considered obsolete. Lets do some math on the OP's statistics, shall we? Lets assume that all 3 races have an equal chance of winning a tournament. This is a binary distribution with each race having a 1/3 chance of winning a tournament, ideally. The likelihood of any given outcome where a race, in this case zerg, wins only 7 tournaments can be expressed as Zerg: (1/3)^7*(2/3)^(85-7=78) = 8.4145 * 10^-18. Of course there are 85C7=4,935,847,320 ways this could occur. The probability of this event is .000000042. If all possibilities were equal one would expect an average probability of 1/85 = .01176. In short, it is extremely unlikely that this would happen by chance. Terran: Now what is the probability of a race winning 54/84 games by chance. The likelihood of any situation where a race wins 54/85 games is (1/3)^54*(2/3)^(85-54=31)= 5.97892*10^-32. There are 85C54=1.48409980330845*10^23 possible outcomes with this result. The final probability of this happening is .000000009 compared to an expected average of .01176.Skill might add some randomness, but i doubt one could argue that Zerg players are that skilless in general. The same could be said about Terran, given that a significant portion of sucessful Terran players don't do much besides SCV all ins and bunker rushes. How can one argue that the races are equal when the probability of Zerg winning only 7 games by chance is less than 1/10,000 the expected average probability. Less than 1/100,000 in the case of Terran.Edit: As it's been pointed out. In the masters league the distribution of the races is close to equal.
You do have to remember that GSL 4 seeded a ton of Terrans into both S class and A class from earlier tournaments, and that the early tournaments were before some of the Terran nerfs/Z and P buffs.
|
|
|
|
|
|
EPT
