|
So in a theoretically balanced game, we would say that two people with the same skill level (defined by mechanics, game sense, game knowledge, and decision-making with their respective races), given a large number of games, would each win approximately the same number of games. If one player is better than another, then played over a large number of games, the better player would win more games than the inferior player.
That said, there are different variations of A versus B played by two equally skilled players:
1) The game, given infinite APM, is imbalanced towards A. However, realistically speaking, A can never reach that level of perfection, so the win percentage happens to be about 50%.
2) The game, given infinite APM, is balanced. However, at lower levels A has a higher win percentage against B. So realistically speaking, B will never reach that level of play so the win percentage is slightly skewed towards A.
3) The early game is imbalanced towards A, but the middle game is imbalanced towards B in such a way that the win percentage happens to be 50%.
4) There are two strategies for each player. A1 wins 100% against B1, A1 wins 0% against B2, A2 wins 0% against B1, A2 wins 100% against B2 and there is no way to scout any variation. All else being equal, this happens to result in a 50% win percentage.
Which combination of these four would be considered to be 'balance'?
 
|
1) Imbalance, you said it in the first sentence. Skill set has nothing to do with balance issues.
2) By lower levels I presume you mean early stages of the game - first 5 minutes? Again, you defined balance on a variable again. What is to say B never reaches A in skill level over years of practice? I understand the concept, but if you say "realistically", why not be realistic?
3) A state of equilibrium or parity characterized by cancellation of all forces by equal opposing forces. Sc2 is complex and has a lot of different aspects to be looked at, but if you generalize it enough to your statement, then yes your statement is true by definition. Both forces have come to an equalization at one point based on any variation during the game, every time, this is balanced.
4) Imbalance. You have a strategy that beats all strategies, this is imbalanced. Even if your end result averages 50%, that is just coincidence. Let me further explain this by saying one particular strat can be used over and over again to achieve 100% win rate, thus causing the obvious imbalance. If for some reason the players continued to play their strats 1-1 causing the 50% win rate, well, you had to set a variable and the experiment is biased.
Edit: I'm glad someone is taking a different approach at the TVZ. I think that we are still at a speculation stage, and questions like yours are the correct ones to be asking. Sc2 is too young and complex to be hammering out facts on balance.
|
Those are good examples. Particularly the 4th one, as it shows it is possible - not saying it is the case - that some aspects of the game are balanced due to equal chances, rather than equal skills required. Currently all those elements are combined into the balance concerns, but Blizzard mostly does the "only" possible thing - base it all on just wins and losses on the ladder. If races win with equal %, they consider the game balanced. They can't control how many people use each race or what is their real skill level. I have no major critique for them, as I haven't come up with any better idea how to balance the game, however it must be regarded that clearly this approach is far from good, and leaves the possibility for imbalance even if the stats can't show it. For example, very few very good players are the only ones who play some difficult race, and make it look balanced (on stats) with the other races, even though it's weaker.
|
There are too many factors involved for true theoretical balance: build orders, spawn locations and maps really throw a wrench in the works. In something like a fighting game, the balance is the same every game; in Starcraft you could always proxy or 14cc and everything changes.
|
Maybe I should have said 'happens to be a 50% win percentage'. I'll edit it in to make it more clear.
On August 23 2010 00:25 Chunkybuddha wrote:
1) Imbalance, you said it in the first sentence. Skill set has nothing to do with balance issues.
2) By lower levels I presume you mean early stages of the game - first 5 minutes? Again, you defined balance on a variable again. What is to say B never reaches A in skill level over years of practice? I understand the concept, but if you say "realistically", why not be realistic?
3) A state of equilibrium or parity characterized by cancellation of all forces by equal opposing forces. Sc2 is complex and has a lot of different aspects to be looked at, but if you generalize it enough to your statement, then yes your statement is true by definition. Both forces have come to an equalization at one point based on any variation during the game, every time, this is balanced.
4) Imbalance. You have a strategy that beats all strategies, this is imbalanced. Even if your end result averages 50%, that is just coincidence. Let me further explain this by saying one particular strat can be used over and over again to achieve 100% win rate, thus causing the obvious imbalance. If for some reason the players continued to play their strats 1-1 causing the 50% win rate, well, you had to set a variable and the experiment is biased.
Edit: I'm glad someone is taking a different approach at the TVZ. I think that we are still at a speculation stage, and questions like yours are the correct ones to be asking. Sc2 is too young and complex to be hammering out facts on balance.
1) and 2): For realistic play, in the context of SC1, of perfect macro (like barracks constantly making units, every larva being used, minerals never going above 400, perfectly splitting marines in arcs) and in SC2 perfect use of Spawn Larva or marine micro against Banelings. You are never going to be 100% on top on Spawn Larva timings and it's never going to happen in a real game. The question is, should we try and get the races such that there is a 50% win percentage among equally skilled players taking that into account, or should we balance it according to 'perfect' play with perfect Larva injects?
3) I'm envisioning a scenario where in the early game, A wins the game outright 40% of the time but B wins the game 83% of the time past the 10 minute mark. A will have a winning percentage of (0.4 + 0.17 * 0.6 = 50.2%) while B will have a winning percentage of 49.8%. Should we consider that balance? Or is balance only when both players have an equal chance to win at all stages of the game?
4) There is no strategy that beats all strategies?... A1 loses to B2 and A2 loses to B1. There is no dominant strategy for either player.
On August 23 2010 01:09 Redmark wrote:
There are too many factors involved for true theoretical balance: build orders, spawn locations and maps really throw a wrench in the works. In something like a fighting game, the balance is the same every game; in Starcraft you could always proxy or 14cc and everything changes.
The problem is to really distill what 'balance' is to its most simple components, and not just in terms of Starcraft.
|
|
|
|
|
|
EPT
