|
On January 20 2011 11:41 Skyze wrote:
balance is pretty damn close right now. Any race can win in almost any situation. Maps play a small part but definitely not as big as some people claim. Without idra there to say "i only lose because that map favors terran" and etc, no one would be talking about maps really.
This is probably the best we're gonna get til HotS comes out, then we start all over again with the balance of new units, yay.
we dont complain about maps cos idra complains about maps... we complain about maps cos they are TERRIBLE
|
I believe the racial mechanics are relatively balanced, I only feel some maps give an advantage to a particular race's mechanics.
|
I think at this point balance can be accomplished through maps, more or less--maybe a few more tweaks are needed on top of that.
|
I believe the maps are really the culprits behind balance.
For the most part, I'd say balance is fine, but the game favors rushes and aggressive play over a macro game.
|
[QUOTE]On January 20 2011 11:39 Rodeo wrote:
Is balance an impossible goal? Or, why  losing to  really means is weak to  .
"I lose to all the time. Blizzard loves and hates . I quit forever."
you do realize that in the early years of starcraft (original) Terran was considered to be vastly in-superior race so ya, they were trying to even out the teams
|
I think the matchmaker aims for 50% win ratio against all races, not just in general. This may mean there is a seperate MMR for each matchup.
|
On January 20 2011 11:56 CecilSunkure wrote:
It seems that the OP is making the assumption that what is "best" or "perfect" balance-wise is both empirically verifiable and that balance can be quantified using some sort of unit-based measurement.
Actually, my point is more or less the opposite. I go through the process of "trying" to measure balance, but mostly to show that you can't. I then suggest a method that I think would make a little sense to use as a template if you had enough data. However, my point wasn't so much about what would happen under conditions of optimal play, but why this is so impossible to measure. In other words, I feel your pain.
People seem to have understood me to be complaining that the game is imbalanced. Actually, I think it is remarkably well balanced. I have my personal opinions on some of the maps (as I think everyone does) but that's not why I'm writing. My point is more to show that identifying an imbalance based on wins and losses, which is the metric that Blizzard has put out to show that their game is balanced, is meaningless. You have to use something else. But whatever Blizzard really uses (and I don't think it's those numbers), they are doing a very good job.
-edit- In response to the post just above this one, that may be true, but you run into the same problem of not having an objective measurement to compare to.
|
This is a very interesting idea. I thought though that the whole "adjusted-win-percentage" thing was to account for the fact that the MMR-system would make everything 50/50.
Like you said you can't take personal experience into account when talking about game balance because let's say that, in your example, from the Toss perspective Z>P>T. Then he is winning over 50% against T, 50 against P, and under 50 against Z. This means that the Toss would be winning ~50% of all his games, but other Toss's with a weakness against T would be balancing his win/loss spread. With all of this, even the mu stats should still be 50/50 for ZvP, ZvT, and PvT. The calculated adjusted win percentage, I thought, was supposed to account for this.
Blizzard has done online competitive support for a while, and I'm sure you aren't the first person who has proposed this concern, which is probably why they came up with this statistic (adjusted win percentage) in the first place.
|
Interesting read.
Nice first post.
|
The game is never going to be balanced by simply changing numbers and stats on units. And this is the only thing that Blizzard is doing so far. The game needs actual core changes to the way it works (maps, unit sizes, macro mechanics, unit designs in general) to eventually be as balanced as BW.
|
One of the first balance threads that I had no problem reading. It makes sense, but I believe Blizzard does have a way of correcting the MMR difference when they look at their data. To be honest, I feel like the game is close enough to balance that I would be ok with Blizzard simply stopping the patches. But I'm a Terran, which is the op race, so I probably don't count.
|
On January 20 2011 12:18 Angra wrote:
The game is never going to be balanced by simply changing numbers and stats on units. And this is the only thing that Blizzard is doing so far. The game needs actual core changes to the way it works (maps, unit sizes, macro mechanics, unit designs in general) to eventually be as balanced as BW.
Balance is like playing a game of Jenga; large changes are almost certain to cause more problems than they fix. Blizzard is only being smart and efficient by taking on balance with the policy they currently enforce, as I can say from personal experience from working on a major RTS title on a balance team. You have no evidence or backing to your claim and posts like yours are just spammy. I hope other people refrain from making posts like this. Merely asserting your opinion upon others is counter-productive to the entire community.
|
On January 20 2011 12:18 Angra wrote:
The game is never going to be balanced by simply changing numbers and stats on units. And this is the only thing that Blizzard is doing so far. The game needs actual core changes to the way it works (maps, unit sizes, macro mechanics, unit designs in general) to eventually be as balanced as BW.
Is that why every matchup has been perceived as imbalanced in both ways, shifting left and right pretty much every week? Is that why when they give Roaches a +1 range people flip the fuck out? Please. The game will not be balanced by armchair designers playing God. The game will be balanced incrementally, more by the community than by Blizzard.
|
It is incredible how many people seem to be incapable of reading a text, and understanding what it says.
For example, how you can get the impression that a thread that talks about the technical viability of imbalance statements in general based on the system used, and which specifically comes to the conclusion that even finding or confirming an imbalance is incredibly hard, because it would screw up different data which would lead to strange results, is a zerg whining thread, is beyond me.
Basically, most people whining about game balance do so because they feel like their own skill level is higher than the results indicate, and thus believe that there must be a flaw in the system giving the results. This usually manifests in either thinking that the ladder system does not work right, or that their chosen race is weaker then the others. As the OP stated, it is very hard to either verify or falsify such a statement empirically, which makes it a question of faith, which can not be solved rationally. This is probably also where this "Everyone knows the game is imbalanced" mentality comes from that usually does not have anything solid to back up its claims except for gut feelings.
However, it should not be impossible to prove or disprove such a statement scientifically. It is just not very obvious, and the obvious statistics might be misleading. So if you want to claim something like this, you need to take a very close look at your methology. As has been stated in the OP, pure ladder win percentages at anything but the very highest level do not give any good data directly. At the very highest ratings, there are not enough people to form a statistically significant group, i believe.
The often stated statistics with the overall tournament wins is obviously flawed, again. First of, most of these tournaments did not even take place in the same version of the game we play at the moment. Secondly, the skill levels of these tournaments are sometimes vastly different. Also, you would obviously need to normate the results relating to the actual participations of the races.
It would be very easy to make statements about balance if it were possible to directly measure the skill of the players, since players of equal skill should play 50/50 against each other. However, as the OP stated, since the most common definition of "equal skill" is "playing about 50/50 against each other", this circularly leads to nothing. This means that it is technically not possible to give each person which exactly defines their skill if you don assume that all matchups are balanced.
Another important problem is generalisation. An imbalance is not necessarily existant throughout all skilllevels.
Also, it is really nice to see someone writing something up that makes actual sense and shows a working scientific mind instead of taking rumours for facts, and basing general statements only on personal feelings.
@Elkram, do you have any sources to that "Adjusted win percentage" thingy? It sounds interesting, especially to know by what they adjust the win percentage.
|
The fact that few people read your text due the "wall of text thread" fear, does not stop of making it an amazing ladder analysis. One of the most interesting threads i ever read in TL.
|
This is a very interesting post Rodeo! Thanks for this.
I have a general question and would be interested in peoples thoughts as to how it would change the situation.
Would the problem be easier to analyze for Blizzard if we had separate MMR ratings for each match up? This seems like it would be easy to implement (it is all behind the scenes) but may be able to separate balance issues.
For example I have been doing really well in PvZ these days. And I have noticed that my PvT has been terrible (I'm pretty confident it is because of the reasons you mentioned above). If I was playing Z and Ts at separate MMR ratings, I would (should) have 50% win rate against both. Then Blizzard could just look at the sum of everyone's (probably in masters league) MMR ratings for the match ups to decide what they should do.
Thoughts?
|
great OP. I think the most useful conclusion to draw is that trying to determine imbalances is useless except at the very highest of play. the problem then becomes that the sample size is so small. basically, locating sources of imbalance is hard as fuck.............................and that's not even worrying about fixing them
|
United States6046 Posts
To quote IdrA:
+ Show Spoiler +if statistics get you hard make one of those ladder analysis pages or something, but stop interfering with balance discussions.
Ladder should never be mentioned in balance discussion.
|
O-o, that is alot of words
what a great read, i agree with you at almost every point.
|
I don't know if your beginning analysis is supported by the facts, but I like the thought process. It seems logical, and it's actually quite profound. Thank you for the thoughts! It would truly be interesting if one matchup imbalance could snowball around in just such a way. I would tend to think no, but I'd have to work through some more things in my head.
|
|
|
|
|
|
EPT
