|
Is above Master, GM only, or inclusive of all Master and above?
|
No wonder I've been so mad on ladder lately :/
|
On August 13 2012 18:40 Whirligig wrote:
Is above Master, GM only, or inclusive of all Master and above?
All master and above.
Average mmr in gm would not tell you much. The result would be different if one top player just make a ladder break.
Their are just to few players.
If i track everyone who is near gm-level, that would be - 1k accounts.
But i dont have every account in my database so to few data...
|
On July 18 2012 23:26 Rick Deckard wrote:Show nested quote +On July 18 2012 23:11 Zacsafus wrote:On July 18 2012 22:47 Rick Deckard wrote:
I believe varying average MMR between races is not an indicator of imbalance.
Here's why, imagine there is a race exactly like zerg but is only be played by blind people, call it blind-zerg. This race would have a very low average MMR as blind people obviously can't play starcraft as well as sighted people. But because the blind-zerg race is the same as zerg it's no weaker than zerg. Thus a low average MMR per race doesn't necessarily imply that race is weaker.
In conclusion the statistics gathered in the study provided can't be used to make conclusions about how well starcraft 2 is balanced.
In practice the data is confounded by the fact that the average skill of of players of different races is not necessarily the same. Dividing race MMR by number of players per race to determine average race MMR doesn't change this. But blind people aren't more likely to pick one race over another, its not possible to imply that one race has worse players on it, at the highest levels everyone is competent and displays good skill so your point isnt valid at high levels, or really at any level because there is no bias between the races of which a handicapped/less-skilled player would pick It's not necessary for blind people pick one race over another for my point to be valid. I don't expect blind people to play starcraft. What I've shown is that differences in average MMR per race doesn't always imply imbalance, even significantly different average MMRs. This is proof by contradiction, it only requires a single counter example to disprove a rule. Given that (my proof is valid that) differences in average MMR per race doesn't imply imbalance in general, it also doesn't per se imply imbalance in starcraft 2. I'm not making a statement about the balance of the game at either high or low skill level. Just pointing out a logical flaw in the argument that different average race MMRs indicate imbalance. As best I can tell the author of the study has concluded because average MMR per race is different therefore the game is imbalance. I believe that logic to be flawed.
Your proof is not valid and you saying it is doesn't make it so...
|
On August 13 2012 22:54 SupLilSon wrote:Show nested quote +On July 18 2012 23:26 Rick Deckard wrote:On July 18 2012 23:11 Zacsafus wrote:On July 18 2012 22:47 Rick Deckard wrote:
I believe varying average MMR between races is not an indicator of imbalance.
Here's why, imagine there is a race exactly like zerg but is only be played by blind people, call it blind-zerg. This race would have a very low average MMR as blind people obviously can't play starcraft as well as sighted people. But because the blind-zerg race is the same as zerg it's no weaker than zerg. Thus a low average MMR per race doesn't necessarily imply that race is weaker.
In conclusion the statistics gathered in the study provided can't be used to make conclusions about how well starcraft 2 is balanced.
In practice the data is confounded by the fact that the average skill of of players of different races is not necessarily the same. Dividing race MMR by number of players per race to determine average race MMR doesn't change this. But blind people aren't more likely to pick one race over another, its not possible to imply that one race has worse players on it, at the highest levels everyone is competent and displays good skill so your point isnt valid at high levels, or really at any level because there is no bias between the races of which a handicapped/less-skilled player would pick It's not necessary for blind people pick one race over another for my point to be valid. I don't expect blind people to play starcraft. What I've shown is that differences in average MMR per race doesn't always imply imbalance, even significantly different average MMRs. This is proof by contradiction, it only requires a single counter example to disprove a rule. Given that (my proof is valid that) differences in average MMR per race doesn't imply imbalance in general, it also doesn't per se imply imbalance in starcraft 2. I'm not making a statement about the balance of the game at either high or low skill level. Just pointing out a logical flaw in the argument that different average race MMRs indicate imbalance. As best I can tell the author of the study has concluded because average MMR per race is different therefore the game is imbalance. I believe that logic to be flawed. Your proof is not valid and you saying it is doesn't make it so...
Rick Deckard makes the mistake of redefining balance.
If his example is true and only blind people play race x and this would make race x have lower mmr.
Than the game is imbalanced.
Race x is weaker in average than the other races!
The REASON for the imbalance of the data is not because of game-design in this case!
Notice that i make clear 100 times in op that i argue about the imbalance and not the REASON.
What this factors are you can not tell without big studies about the topic and even than its more sociology than math.
His theorie about blind people is a valid argument. But it is obvious not the reason.
But i think he wants to show that imbalance dont have to come from gamedesign and this is true.
However different reason for something dont change the reality of it.
You are free to argue about the reason however because of the missing big sociology study about this topic you will not come to an conclusion.
|
On August 13 2012 23:03 skeldark wrote:Show nested quote +On August 13 2012 22:54 SupLilSon wrote:On July 18 2012 23:26 Rick Deckard wrote:On July 18 2012 23:11 Zacsafus wrote:On July 18 2012 22:47 Rick Deckard wrote:
I believe varying average MMR between races is not an indicator of imbalance.
Here's why, imagine there is a race exactly like zerg but is only be played by blind people, call it blind-zerg. This race would have a very low average MMR as blind people obviously can't play starcraft as well as sighted people. But because the blind-zerg race is the same as zerg it's no weaker than zerg. Thus a low average MMR per race doesn't necessarily imply that race is weaker.
In conclusion the statistics gathered in the study provided can't be used to make conclusions about how well starcraft 2 is balanced.
In practice the data is confounded by the fact that the average skill of of players of different races is not necessarily the same. Dividing race MMR by number of players per race to determine average race MMR doesn't change this. But blind people aren't more likely to pick one race over another, its not possible to imply that one race has worse players on it, at the highest levels everyone is competent and displays good skill so your point isnt valid at high levels, or really at any level because there is no bias between the races of which a handicapped/less-skilled player would pick It's not necessary for blind people pick one race over another for my point to be valid. I don't expect blind people to play starcraft. What I've shown is that differences in average MMR per race doesn't always imply imbalance, even significantly different average MMRs. This is proof by contradiction, it only requires a single counter example to disprove a rule. Given that (my proof is valid that) differences in average MMR per race doesn't imply imbalance in general, it also doesn't per se imply imbalance in starcraft 2. I'm not making a statement about the balance of the game at either high or low skill level. Just pointing out a logical flaw in the argument that different average race MMRs indicate imbalance. As best I can tell the author of the study has concluded because average MMR per race is different therefore the game is imbalance. I believe that logic to be flawed. Your proof is not valid and you saying it is doesn't make it so... Rick Deckard makes the mistake of redefining balance. If his example is true and only blind people play race x and this would make race x have lower mmr. Than the game is imbalanced. Race x is weaker in average than the other races!
The REASON for the imbalance of the data is not because of game-design in this case!Notice that i make clear 100 times in op that i argue about the imbalance and not the REASON. What this factors are you can not tell without big studies about the topic and even than its more sociology than math. His theorie about blind people is a valid argument. But it is obvious not the reason. But he is right that imbalance dont have to come from gamedesign. I think that is the point he wants to make. However different reason for something dont change the reality of it.
Yea, I understood the gist of his point, that MMR differences do not necessarly correlate directly with imbalance, that other factors may be at work. But the example of blind people favoring Z is pretty rediculous.
|
On August 13 2012 23:08 SupLilSon wrote:Show nested quote +On August 13 2012 23:03 skeldark wrote:On August 13 2012 22:54 SupLilSon wrote:On July 18 2012 23:26 Rick Deckard wrote:On July 18 2012 23:11 Zacsafus wrote:On July 18 2012 22:47 Rick Deckard wrote:
I believe varying average MMR between races is not an indicator of imbalance.
Here's why, imagine there is a race exactly like zerg but is only be played by blind people, call it blind-zerg. This race would have a very low average MMR as blind people obviously can't play starcraft as well as sighted people. But because the blind-zerg race is the same as zerg it's no weaker than zerg. Thus a low average MMR per race doesn't necessarily imply that race is weaker.
In conclusion the statistics gathered in the study provided can't be used to make conclusions about how well starcraft 2 is balanced.
In practice the data is confounded by the fact that the average skill of of players of different races is not necessarily the same. Dividing race MMR by number of players per race to determine average race MMR doesn't change this. But blind people aren't more likely to pick one race over another, its not possible to imply that one race has worse players on it, at the highest levels everyone is competent and displays good skill so your point isnt valid at high levels, or really at any level because there is no bias between the races of which a handicapped/less-skilled player would pick It's not necessary for blind people pick one race over another for my point to be valid. I don't expect blind people to play starcraft. What I've shown is that differences in average MMR per race doesn't always imply imbalance, even significantly different average MMRs. This is proof by contradiction, it only requires a single counter example to disprove a rule. Given that (my proof is valid that) differences in average MMR per race doesn't imply imbalance in general, it also doesn't per se imply imbalance in starcraft 2. I'm not making a statement about the balance of the game at either high or low skill level. Just pointing out a logical flaw in the argument that different average race MMRs indicate imbalance. As best I can tell the author of the study has concluded because average MMR per race is different therefore the game is imbalance. I believe that logic to be flawed. Your proof is not valid and you saying it is doesn't make it so... Rick Deckard makes the mistake of redefining balance. If his example is true and only blind people play race x and this would make race x have lower mmr. Than the game is imbalanced. Race x is weaker in average than the other races!
The REASON for the imbalance of the data is not because of game-design in this case!Notice that i make clear 100 times in op that i argue about the imbalance and not the REASON. What this factors are you can not tell without big studies about the topic and even than its more sociology than math. His theorie about blind people is a valid argument. But it is obvious not the reason. But he is right that imbalance dont have to come from gamedesign. I think that is the point he wants to make. However different reason for something dont change the reality of it. Yea, I understood the gist of his point, that MMR differences do not necessarly correlate directly with imbalance, that other factors may be at work. But the example of blind people favoring Z is pretty rediculous.
It is. But he takes a extreme point to show the existens of an group of points. Thats a valid argumentation.
What he did wrong is : He said he prove that this dont show balance but what he really proved is:
"The reason for data-balance is not in 100% cases game-design."
A point i mentioned in the op already.
However the main point is, that ANY way to calculate imbalance have the this problem.
There is NO way to be 100% sure the reason for the imbalance comes from the game design.
But we have to just act like this is the case because in all other cases,
there would be no point of balancing the game.
|
On August 13 2012 23:17 skeldark wrote:Show nested quote +On August 13 2012 23:08 SupLilSon wrote:On August 13 2012 23:03 skeldark wrote:On August 13 2012 22:54 SupLilSon wrote:On July 18 2012 23:26 Rick Deckard wrote:On July 18 2012 23:11 Zacsafus wrote:On July 18 2012 22:47 Rick Deckard wrote:
I believe varying average MMR between races is not an indicator of imbalance.
Here's why, imagine there is a race exactly like zerg but is only be played by blind people, call it blind-zerg. This race would have a very low average MMR as blind people obviously can't play starcraft as well as sighted people. But because the blind-zerg race is the same as zerg it's no weaker than zerg. Thus a low average MMR per race doesn't necessarily imply that race is weaker.
In conclusion the statistics gathered in the study provided can't be used to make conclusions about how well starcraft 2 is balanced.
In practice the data is confounded by the fact that the average skill of of players of different races is not necessarily the same. Dividing race MMR by number of players per race to determine average race MMR doesn't change this. But blind people aren't more likely to pick one race over another, its not possible to imply that one race has worse players on it, at the highest levels everyone is competent and displays good skill so your point isnt valid at high levels, or really at any level because there is no bias between the races of which a handicapped/less-skilled player would pick It's not necessary for blind people pick one race over another for my point to be valid. I don't expect blind people to play starcraft. What I've shown is that differences in average MMR per race doesn't always imply imbalance, even significantly different average MMRs. This is proof by contradiction, it only requires a single counter example to disprove a rule. Given that (my proof is valid that) differences in average MMR per race doesn't imply imbalance in general, it also doesn't per se imply imbalance in starcraft 2. I'm not making a statement about the balance of the game at either high or low skill level. Just pointing out a logical flaw in the argument that different average race MMRs indicate imbalance. As best I can tell the author of the study has concluded because average MMR per race is different therefore the game is imbalance. I believe that logic to be flawed. Your proof is not valid and you saying it is doesn't make it so... Rick Deckard makes the mistake of redefining balance. If his example is true and only blind people play race x and this would make race x have lower mmr. Than the game is imbalanced. Race x is weaker in average than the other races!
The REASON for the imbalance of the data is not because of game-design in this case!Notice that i make clear 100 times in op that i argue about the imbalance and not the REASON. What this factors are you can not tell without big studies about the topic and even than its more sociology than math. His theorie about blind people is a valid argument. But it is obvious not the reason. But he is right that imbalance dont have to come from gamedesign. I think that is the point he wants to make. However different reason for something dont change the reality of it. Yea, I understood the gist of his point, that MMR differences do not necessarly correlate directly with imbalance, that other factors may be at work. But the example of blind people favoring Z is pretty rediculous. It is. But he takes a extreme point to show the existens of an group of points. Thats a valid argumentation. What he did wrong is : He said he prove that this dont show balance but what he really proved is: "The reason for data-balance is not in 100% cases game-design." A point i mentioned in the op already. However the main point is, that ANY way to calculate imbalance have the this problem. There is NO way to be 100% sure the reason for the imbalance comes from the game design. But we have to just act like this is the case because in all other cases, there would be no point of balancing the game.
Skeldark, you do realise that more than half of the discussion you end up in in this thread comes from you using the word "imbalance" in a different way than everyone else on this forum? 
Here on TL, when people talk about imbalance, they refer to game design. So as I told you several times already, if you want to avoid discussion like the one I am quoting, you need to be MUCH clearer with what you mean when you use that word. Or better, use the word the same way everyone else does, and use a different word for what you mean with imbalance, like "race dependent MMR distribution", which is a much more transparent term. So by typing a few more letter for that, you will save yourself 100 times that in avoiding replies like the quoted ones above.
I don't care if your definition makes sense or not, it is just a matter of communication. And it is not effective communication to start using a word differently from everyone else.
But if you enjoy ending up in this discussion over and over again, go ahead. 
|
On August 14 2012 08:30 Cascade wrote:Show nested quote +On August 13 2012 23:17 skeldark wrote:On August 13 2012 23:08 SupLilSon wrote:On August 13 2012 23:03 skeldark wrote:On August 13 2012 22:54 SupLilSon wrote:On July 18 2012 23:26 Rick Deckard wrote:On July 18 2012 23:11 Zacsafus wrote:On July 18 2012 22:47 Rick Deckard wrote:
I believe varying average MMR between races is not an indicator of imbalance.
Here's why, imagine there is a race exactly like zerg but is only be played by blind people, call it blind-zerg. This race would have a very low average MMR as blind people obviously can't play starcraft as well as sighted people. But because the blind-zerg race is the same as zerg it's no weaker than zerg. Thus a low average MMR per race doesn't necessarily imply that race is weaker.
In conclusion the statistics gathered in the study provided can't be used to make conclusions about how well starcraft 2 is balanced.
In practice the data is confounded by the fact that the average skill of of players of different races is not necessarily the same. Dividing race MMR by number of players per race to determine average race MMR doesn't change this. But blind people aren't more likely to pick one race over another, its not possible to imply that one race has worse players on it, at the highest levels everyone is competent and displays good skill so your point isnt valid at high levels, or really at any level because there is no bias between the races of which a handicapped/less-skilled player would pick It's not necessary for blind people pick one race over another for my point to be valid. I don't expect blind people to play starcraft. What I've shown is that differences in average MMR per race doesn't always imply imbalance, even significantly different average MMRs. This is proof by contradiction, it only requires a single counter example to disprove a rule. Given that (my proof is valid that) differences in average MMR per race doesn't imply imbalance in general, it also doesn't per se imply imbalance in starcraft 2. I'm not making a statement about the balance of the game at either high or low skill level. Just pointing out a logical flaw in the argument that different average race MMRs indicate imbalance. As best I can tell the author of the study has concluded because average MMR per race is different therefore the game is imbalance. I believe that logic to be flawed. Your proof is not valid and you saying it is doesn't make it so... Rick Deckard makes the mistake of redefining balance. If his example is true and only blind people play race x and this would make race x have lower mmr. Than the game is imbalanced. Race x is weaker in average than the other races!
The REASON for the imbalance of the data is not because of game-design in this case!Notice that i make clear 100 times in op that i argue about the imbalance and not the REASON. What this factors are you can not tell without big studies about the topic and even than its more sociology than math. His theorie about blind people is a valid argument. But it is obvious not the reason. But he is right that imbalance dont have to come from gamedesign. I think that is the point he wants to make. However different reason for something dont change the reality of it. Yea, I understood the gist of his point, that MMR differences do not necessarly correlate directly with imbalance, that other factors may be at work. But the example of blind people favoring Z is pretty rediculous. It is. But he takes a extreme point to show the existens of an group of points. Thats a valid argumentation. What he did wrong is : He said he prove that this dont show balance but what he really proved is: "The reason for data-balance is not in 100% cases game-design." A point i mentioned in the op already. However the main point is, that ANY way to calculate imbalance have the this problem. There is NO way to be 100% sure the reason for the imbalance comes from the game design. But we have to just act like this is the case because in all other cases, there would be no point of balancing the game. Skeldark, you do realise that more than half of the discussion you end up in in this thread comes from you using the word "imbalance" in a different way than everyone else on this forum? Here on TL, when people talk about imbalance, they refer to game design. So as I told you several times already, if you want to avoid discussion like the one I am quoting, you need to be MUCH clearer with what you mean when you use that word. Or better, use the word the same way everyone else does, and use a different word for what you mean with imbalance, like "race dependent MMR distribution", which is a much more transparent term. So by typing a few more letter for that, you will save yourself 100 times that in avoiding replies like the quoted ones above. I don't care if your definition makes sense or not, it is just a matter of communication. And it is not effective communication to start using a word differently from everyone else. But if you enjoy ending up in this discussion over and over again, go ahead.
I use it like everyone else, all i do is, i look deeper in it.
What most people, you included did not understand is:
I explain data - imblance dont have to be game design.
However its pointless to assume anything else.
This is valid for every game and every method to detect imbalance.
This problem have nothing to do with this thread or my data. I just pointed out that you can never now the reason for sure.
Other threads did not point this out even if its valid for everything. Blizzard patches, tournament results ect.
People just started to realise that and now act like its related to the method i use or my data.
TLDR
data balance dont have to be game design.
If its not, there is no point of balance an game.
So we have to assume data-inbalance = game design imbalance if we want to try to balance it.
PS:
but yes im tiered of this discussion 
|
On August 14 2012 10:30 skeldark wrote:Show nested quote +On August 14 2012 08:30 Cascade wrote:On August 13 2012 23:17 skeldark wrote:On August 13 2012 23:08 SupLilSon wrote:On August 13 2012 23:03 skeldark wrote:On August 13 2012 22:54 SupLilSon wrote:On July 18 2012 23:26 Rick Deckard wrote:On July 18 2012 23:11 Zacsafus wrote:On July 18 2012 22:47 Rick Deckard wrote:
I believe varying average MMR between races is not an indicator of imbalance.
Here's why, imagine there is a race exactly like zerg but is only be played by blind people, call it blind-zerg. This race would have a very low average MMR as blind people obviously can't play starcraft as well as sighted people. But because the blind-zerg race is the same as zerg it's no weaker than zerg. Thus a low average MMR per race doesn't necessarily imply that race is weaker.
In conclusion the statistics gathered in the study provided can't be used to make conclusions about how well starcraft 2 is balanced.
In practice the data is confounded by the fact that the average skill of of players of different races is not necessarily the same. Dividing race MMR by number of players per race to determine average race MMR doesn't change this. But blind people aren't more likely to pick one race over another, its not possible to imply that one race has worse players on it, at the highest levels everyone is competent and displays good skill so your point isnt valid at high levels, or really at any level because there is no bias between the races of which a handicapped/less-skilled player would pick It's not necessary for blind people pick one race over another for my point to be valid. I don't expect blind people to play starcraft. What I've shown is that differences in average MMR per race doesn't always imply imbalance, even significantly different average MMRs. This is proof by contradiction, it only requires a single counter example to disprove a rule. Given that (my proof is valid that) differences in average MMR per race doesn't imply imbalance in general, it also doesn't per se imply imbalance in starcraft 2. I'm not making a statement about the balance of the game at either high or low skill level. Just pointing out a logical flaw in the argument that different average race MMRs indicate imbalance. As best I can tell the author of the study has concluded because average MMR per race is different therefore the game is imbalance. I believe that logic to be flawed. Your proof is not valid and you saying it is doesn't make it so... Rick Deckard makes the mistake of redefining balance. If his example is true and only blind people play race x and this would make race x have lower mmr. Than the game is imbalanced. Race x is weaker in average than the other races!
The REASON for the imbalance of the data is not because of game-design in this case!Notice that i make clear 100 times in op that i argue about the imbalance and not the REASON. What this factors are you can not tell without big studies about the topic and even than its more sociology than math. His theorie about blind people is a valid argument. But it is obvious not the reason. But he is right that imbalance dont have to come from gamedesign. I think that is the point he wants to make. However different reason for something dont change the reality of it. Yea, I understood the gist of his point, that MMR differences do not necessarly correlate directly with imbalance, that other factors may be at work. But the example of blind people favoring Z is pretty rediculous. It is. But he takes a extreme point to show the existens of an group of points. Thats a valid argumentation. What he did wrong is : He said he prove that this dont show balance but what he really proved is: "The reason for data-balance is not in 100% cases game-design." A point i mentioned in the op already. However the main point is, that ANY way to calculate imbalance have the this problem. There is NO way to be 100% sure the reason for the imbalance comes from the game design. But we have to just act like this is the case because in all other cases, there would be no point of balancing the game. Skeldark, you do realise that more than half of the discussion you end up in in this thread comes from you using the word "imbalance" in a different way than everyone else on this forum? Here on TL, when people talk about imbalance, they refer to game design. So as I told you several times already, if you want to avoid discussion like the one I am quoting, you need to be MUCH clearer with what you mean when you use that word. Or better, use the word the same way everyone else does, and use a different word for what you mean with imbalance, like "race dependent MMR distribution", which is a much more transparent term. So by typing a few more letter for that, you will save yourself 100 times that in avoiding replies like the quoted ones above. I don't care if your definition makes sense or not, it is just a matter of communication. And it is not effective communication to start using a word differently from everyone else. But if you enjoy ending up in this discussion over and over again, go ahead. I use it like everyone else, all i do is, i look deeper in it. What most people, you included did not understand is: I explain data - imblance dont have to be game design. However its pointless to assume anything else. This is valid for every game and every method to detect imbalance.This problem have nothing to do with this thread or my data. I just pointed out that you can never now the reason for sure. Other threads did not point this out even if its valid for everything. Blizzard patches, tournament results ect. People just started to realise that and now act like its related to the method i use or my data. TLDR data balance dont have to be game design. If its not, there is no point of balance an game. So we have to assume data-inbalance = game design imbalance if we want to try to balance it. PS: but yes im tiered of this discussion
The reason (one of the reasons) you (and me) are tired of this discussion is because I told you the exact same thing earlier in the thread, and at that point you acknowledged that maybe it was a poor choice of word, but you seem to have forgotten that by now.
I understand perfectly what you did. We discussed it for ages earlier in the thread, and I calculated the errors for you from your data file. Remember?
What you (seem to) fail to understand is that when most of the people on TL use (or read) the word "imbalance" in this context, they refer to a design flaw. You know of all the "protoss imba" threads. They are not talking about sc2 newbs tending to choose terran, causing a tilted MMR distribution. They talk about flawed game play causing some races being easier to win with at the very highest level.
You are talking about imbalance in your data (different average MMR), which as you said, may or may not be a signal of a flawed design.
Due to these two different uses of the word "imbalance", when you talk about "proving imbalance" and so on, referring to imbalance in your data, many of the TLers skimming through your text will understand it as if you are saying that you have proven that there is a design flaw in sc2. I know that is not what you mean, but I'm telling you that that is what many understand from what they read. Which is why you end up in a lot of needless discussions. So if you are tired of discussions, it may be worth the effort to use a notation that will not be misunderstood.
I'm not saying your choice of word is technically wrong, but it is begging for misunderstandings in this context, on this forum.
anyway. cheers, gl.
|
On August 14 2012 12:59 Cascade wrote:Show nested quote +On August 14 2012 10:30 skeldark wrote:On August 14 2012 08:30 Cascade wrote:On August 13 2012 23:17 skeldark wrote:On August 13 2012 23:08 SupLilSon wrote:On August 13 2012 23:03 skeldark wrote:On August 13 2012 22:54 SupLilSon wrote:On July 18 2012 23:26 Rick Deckard wrote:On July 18 2012 23:11 Zacsafus wrote:On July 18 2012 22:47 Rick Deckard wrote:
I believe varying average MMR between races is not an indicator of imbalance.
Here's why, imagine there is a race exactly like zerg but is only be played by blind people, call it blind-zerg. This race would have a very low average MMR as blind people obviously can't play starcraft as well as sighted people. But because the blind-zerg race is the same as zerg it's no weaker than zerg. Thus a low average MMR per race doesn't necessarily imply that race is weaker.
In conclusion the statistics gathered in the study provided can't be used to make conclusions about how well starcraft 2 is balanced.
In practice the data is confounded by the fact that the average skill of of players of different races is not necessarily the same. Dividing race MMR by number of players per race to determine average race MMR doesn't change this. But blind people aren't more likely to pick one race over another, its not possible to imply that one race has worse players on it, at the highest levels everyone is competent and displays good skill so your point isnt valid at high levels, or really at any level because there is no bias between the races of which a handicapped/less-skilled player would pick It's not necessary for blind people pick one race over another for my point to be valid. I don't expect blind people to play starcraft. What I've shown is that differences in average MMR per race doesn't always imply imbalance, even significantly different average MMRs. This is proof by contradiction, it only requires a single counter example to disprove a rule. Given that (my proof is valid that) differences in average MMR per race doesn't imply imbalance in general, it also doesn't per se imply imbalance in starcraft 2. I'm not making a statement about the balance of the game at either high or low skill level. Just pointing out a logical flaw in the argument that different average race MMRs indicate imbalance. As best I can tell the author of the study has concluded because average MMR per race is different therefore the game is imbalance. I believe that logic to be flawed. Your proof is not valid and you saying it is doesn't make it so... Rick Deckard makes the mistake of redefining balance. If his example is true and only blind people play race x and this would make race x have lower mmr. Than the game is imbalanced. Race x is weaker in average than the other races!
The REASON for the imbalance of the data is not because of game-design in this case!Notice that i make clear 100 times in op that i argue about the imbalance and not the REASON. What this factors are you can not tell without big studies about the topic and even than its more sociology than math. His theorie about blind people is a valid argument. But it is obvious not the reason. But he is right that imbalance dont have to come from gamedesign. I think that is the point he wants to make. However different reason for something dont change the reality of it. Yea, I understood the gist of his point, that MMR differences do not necessarly correlate directly with imbalance, that other factors may be at work. But the example of blind people favoring Z is pretty rediculous. It is. But he takes a extreme point to show the existens of an group of points. Thats a valid argumentation. What he did wrong is : He said he prove that this dont show balance but what he really proved is: "The reason for data-balance is not in 100% cases game-design." A point i mentioned in the op already. However the main point is, that ANY way to calculate imbalance have the this problem. There is NO way to be 100% sure the reason for the imbalance comes from the game design. But we have to just act like this is the case because in all other cases, there would be no point of balancing the game. Skeldark, you do realise that more than half of the discussion you end up in in this thread comes from you using the word "imbalance" in a different way than everyone else on this forum? Here on TL, when people talk about imbalance, they refer to game design. So as I told you several times already, if you want to avoid discussion like the one I am quoting, you need to be MUCH clearer with what you mean when you use that word. Or better, use the word the same way everyone else does, and use a different word for what you mean with imbalance, like "race dependent MMR distribution", which is a much more transparent term. So by typing a few more letter for that, you will save yourself 100 times that in avoiding replies like the quoted ones above. I don't care if your definition makes sense or not, it is just a matter of communication. And it is not effective communication to start using a word differently from everyone else. But if you enjoy ending up in this discussion over and over again, go ahead. I use it like everyone else, all i do is, i look deeper in it. What most people, you included did not understand is: I explain data - imblance dont have to be game design. However its pointless to assume anything else. This is valid for every game and every method to detect imbalance.This problem have nothing to do with this thread or my data. I just pointed out that you can never now the reason for sure. Other threads did not point this out even if its valid for everything. Blizzard patches, tournament results ect. People just started to realise that and now act like its related to the method i use or my data. TLDR data balance dont have to be game design. If its not, there is no point of balance an game. So we have to assume data-inbalance = game design imbalance if we want to try to balance it. PS: but yes im tiered of this discussion The reason (one of the reasons) you (and me) are tired of this discussion is because I told you the exact same thing earlier in the thread, and at that point you acknowledged that maybe it was a poor choice of word, but you seem to have forgotten that by now. I understand perfectly what you did. We discussed it for ages earlier in the thread, and I calculated the errors for you from your data file. Remember? What you (seem to) fail to understand is that when most of the people on TL use (or read) the word "imbalance" in this context, they refer to a design flaw. You know of all the "protoss imba" threads. They are not talking about sc2 newbs tending to choose terran, causing a tilted MMR distribution. They talk about flawed game play causing some races being easier to win with at the very highest level. You are talking about imbalance in your data (different average MMR), which as you said, may or may not be a signal of a flawed design. Due to these two different uses of the word "imbalance", when you talk about "proving imbalance" and so on, referring to imbalance in your data, many of the TLers skimming through your text will understand it as if you are saying that you have proven that there is a design flaw in sc2. I know that is not what you mean, but I'm telling you that that is what many understand from what they read. Which is why you end up in a lot of needless discussions. So if you are tired of discussions, it may be worth the effort to use a notation that will not be misunderstood. I'm not saying your choice of word is technically wrong, but it is begging for misunderstandings in this context, on this forum. anyway. cheers, gl.
sure i remember and i did before i wrote the last post,
I know that i agreed with you that it was poor choice of words but i changed my opinion in some ways.
Most people use it as design flaw because they just think: data imbalance = design imbalance.
If top 16 of all tournaments for 10 years are terran, this is an imbalance of data but don't have to be from game design.
(hard example: perhaps the rules of the tournaments only allow terran)
BUT:
My mistake is to mix the this one little point about game design with the main topic, publishing my data.
in the op there is no "proven imbalance" any-more.
There is more a "take the data as what it is, if you dont understand it i dont care" ^^
I kind of regret the whole thing by now. TL is just the wrong platform for this kind of posts.
This is in no means pointed at you.
|
just found this thread... great work.
But I couldn't really find out how you calculate the MMR 
But I love the data and unbiased presentation of it, shows how close everything is.
|
|
|
Skeldark, when WOW first introduced battlegrounds (yes, ~7 years ago), Horde was winning these battlegrounds significantly more across all servers. But this wasn't because the game mechanics favored Horde in any way. It just so happened that "hardcore" players were significantly more likely to pick Horde and "casuals" more likely to pick Alliance. It's definitely very possible that the same could be happening in SC2; one race might be more likely to draw in players of weaker skill level. This would cause the average MMR of that race to be lower, even though it's actually balanced.
Some data I would LOVE to see: average win rates for Random players, broken down by race. Eg: What is the win percentage for random players when they spawn as Terran? What is the win percentage for random players when they spawn as Zerg?
|
On August 15 2012 08:48 whacks wrote:
Skeldark, when WOW first introduced battlegrounds (yes, ~7 years ago), Horde was winning these battlegrounds significantly more across all servers. But this wasn't because the game mechanics favored Horde in any way. It just so happened that "hardcore" players were significantly more likely to pick Horde and "casuals" more likely to pick Alliance. It's definitely very possible that the same could be happening in SC2; one race might be more likely to draw in players of weaker skill level. This would cause the average MMR of that race to be lower, even though it's actually balanced.
Some data I would LOVE to see: average win rates for Random players, broken down by race. Eg: What is the win percentage for random players when they spawn as Terran? What is the win percentage for random players when they spawn as Zerg?
If one race draw weaker players the race is obvious weaker than the others!
So the game is imbalanced. In this case, blizzard would try to buff this race to make the game balanced again.
Please read the post on this page. we just discussed this topic.
To the causal argument: First dataset show master player only.
Just edited the op:
This is the diffrence to the average MMR, of my Ladder-Data, per Race. Not more not Less.
You can not see on any statistic game-data, if the reason is game design or social aspects.
Not you, not me, not blizzard, not a single game designer!
So we have the choice of paying for a global sociology study to find it out (if you can call it this way in sociology ^^)
or just ASSUME it comes from game design like every game company does.
To your random data:
There are only few random players and this would be a lot of work only to find out witch race is the favourite of most random players.
I hope you dont think that favourite of randoms prove: strongest race for everyone because of game design.
You started with social reasons. I think this example show them perfect.
Random players play the game total different than all others.
My first runs with random come to the result, that random players have a -100 - -200 lower mmr than all other races.
They are the race that draw the player with less skill or it is a way harder to play different races.
They start with an automatic advantage in sc2 and have no in-game disadvantages because of race-matchups.
( they would equal out in few games) So in this special case, it is even proven the Avg-MMR-diffrence must come from an social reason.
PS:
Sorry, this part is even for me hard to read.
But i dont know how to explain it different with my limited english.
|
On August 15 2012 08:48 whacks wrote:
Skeldark, when WOW first introduced battlegrounds (yes, ~7 years ago), Horde was winning these battlegrounds significantly more across all servers. But this wasn't because the game mechanics favored Horde in any way. It just so happened that "hardcore" players were significantly more likely to pick Horde and "casuals" more likely to pick Alliance. It's definitely very possible that the same could be happening in SC2; one race might be more likely to draw in players of weaker skill level. This would cause the average MMR of that race to be lower, even though it's actually balanced.
Some data I would LOVE to see: average win rates for Random players, broken down by race. Eg: What is the win percentage for random players when they spawn as Terran? What is the win percentage for random players when they spawn as Zerg?
The data is more biased towards higher skill levels, ie. not "casuals". Please read the original post and stop posting the first thing that comes to your mind.
|
On August 15 2012 11:42 plogamer wrote:Show nested quote +On August 15 2012 08:48 whacks wrote:
Skeldark, when WOW first introduced battlegrounds (yes, ~7 years ago), Horde was winning these battlegrounds significantly more across all servers. But this wasn't because the game mechanics favored Horde in any way. It just so happened that "hardcore" players were significantly more likely to pick Horde and "casuals" more likely to pick Alliance. It's definitely very possible that the same could be happening in SC2; one race might be more likely to draw in players of weaker skill level. This would cause the average MMR of that race to be lower, even though it's actually balanced.
Some data I would LOVE to see: average win rates for Random players, broken down by race. Eg: What is the win percentage for random players when they spawn as Terran? What is the win percentage for random players when they spawn as Zerg? The data is more biased towards higher skill levels, ie. not "casuals". Please read the original post and stop posting the first thing that comes to your mind.
You do realize that skeldark is the OP right? He made this algorithm and made the entire calculation. I think he knows a bit more than u do....
|
On August 15 2012 11:47 CaptainCrush wrote:Show nested quote +On August 15 2012 11:42 plogamer wrote:On August 15 2012 08:48 whacks wrote:
Skeldark, when WOW first introduced battlegrounds (yes, ~7 years ago), Horde was winning these battlegrounds significantly more across all servers. But this wasn't because the game mechanics favored Horde in any way. It just so happened that "hardcore" players were significantly more likely to pick Horde and "casuals" more likely to pick Alliance. It's definitely very possible that the same could be happening in SC2; one race might be more likely to draw in players of weaker skill level. This would cause the average MMR of that race to be lower, even though it's actually balanced.
Some data I would LOVE to see: average win rates for Random players, broken down by race. Eg: What is the win percentage for random players when they spawn as Terran? What is the win percentage for random players when they spawn as Zerg? The data is more biased towards higher skill levels, ie. not "casuals". Please read the original post and stop posting the first thing that comes to your mind. You do realize that skeldark is the OP right? He made this algorithm and made the entire calculation. I think he knows a bit more than u do....
He quoted whacks not me. I think he posted while i wrote my answer and did not see it.
@plogamer if you answer to everyone that dont read the full op like this, you have a lot to do on this website
@CaptainCrush
Just to clear this.
I did not do the algorithm alone. The dmmr calculation and the tier offsets comes form Not_That.
I did the tier analyser ( the algorithm that try to find out in what tier a player is) and the software-program.
|
On August 15 2012 11:42 plogamer wrote:Show nested quote +On August 15 2012 08:48 whacks wrote:
Skeldark, when WOW first introduced battlegrounds (yes, ~7 years ago), Horde was winning these battlegrounds significantly more across all servers. But this wasn't because the game mechanics favored Horde in any way. It just so happened that "hardcore" players were significantly more likely to pick Horde and "casuals" more likely to pick Alliance. It's definitely very possible that the same could be happening in SC2; one race might be more likely to draw in players of weaker skill level. This would cause the average MMR of that race to be lower, even though it's actually balanced.
Some data I would LOVE to see: average win rates for Random players, broken down by race. Eg: What is the win percentage for random players when they spawn as Terran? What is the win percentage for random players when they spawn as Zerg? The data is more biased towards higher skill levels, ie. not "casuals". Please read the original post and stop posting the first thing that comes to your mind.
Easy on the nerd rage there
What you and skeldark don't seem to be getting is that there is a self selection bias in your analysis. This self selection bias could introduce a systematic error that is prevalent at all levels of the game and cannot be eliminated by averaging large sample sizes. In the wow example that I gave, I was referring to those who hit level 60 within a few months... They spent hours playing the game everyday. And yet, even within them, the higher skilled players had a bias towards one race. This had nothing to do with design imbalance since the two races were almost exactly identical.
This is why any serious study in peer reviewed journals avoid self-selection bias where at all possible. The best way to do this for SC2, would be to look at data gathered by random-race players only.
Anyway, I've said my peace. If you decide to crunch the random-race data, that would be great. If not, thanks for what you've put together so far.
|
On August 16 2012 01:45 whacks wrote:Show nested quote +On August 15 2012 11:42 plogamer wrote:On August 15 2012 08:48 whacks wrote:
Skeldark, when WOW first introduced battlegrounds (yes, ~7 years ago), Horde was winning these battlegrounds significantly more across all servers. But this wasn't because the game mechanics favored Horde in any way. It just so happened that "hardcore" players were significantly more likely to pick Horde and "casuals" more likely to pick Alliance. It's definitely very possible that the same could be happening in SC2; one race might be more likely to draw in players of weaker skill level. This would cause the average MMR of that race to be lower, even though it's actually balanced.
Some data I would LOVE to see: average win rates for Random players, broken down by race. Eg: What is the win percentage for random players when they spawn as Terran? What is the win percentage for random players when they spawn as Zerg? The data is more biased towards higher skill levels, ie. not "casuals". Please read the original post and stop posting the first thing that comes to your mind. Easy on the nerd rage there What you and skeldark don't seem to be getting is that there is a self selection bias in your analysis. This self selection bias could introduce a systematic error that is prevalent at all levels of the game and cannot be eliminated by averaging large sample sizes. In the wow example that I gave, I was referring to those who hit level 60 within a few months... They spent hours playing the game everyday. And yet, even within them, the higher skilled players had a bias towards one race. This had nothing to do with design imbalance since the two races were almost exactly identical. This is why any serious study in peer reviewed journals avoid self-selection bias where at all possible. The best way to do this for SC2, would be to look at data gathered by random-race players only. Anyway, I've said my peace. If you decide to crunch the random-race data, that would be great. If not, thanks for what you've put together so far.
You miss the point. My argument had nothing to do plogamer sentence.
Social aspects are a total different point than data-biased. If you call the data biased because of the possible existent of social aspects than every data in this world is biased.
Please reads my answer-post again.
|
|
|
|
|
|
EPT
