Hello, out of curiosity I've calculated winratios by race for winners bracket (1-5 round) and losers bracket (1-7 round) of IPL 5 Tournament, so its quite up-to-date info. If anybody interested I can send excel file in which every calculation was proceed. Just PM.
And please keep balance QQ as far as possible away from this thread! Thanks in advance.
Before we look into results, I want to explain few things. Xrace wins% formula: sum of race wins in specific MU divided by amount of same MUs. For example Zerg wins% vs Terran = 13 zerg wins against terran divided by 19 TvZs (see source) "Overall" result includes every win of specific race (even in mirrors) divided by all matches in analysed bracket. Ok, let's go!
So.. that is all! Thanks for reading, I hope that at least one person find it helpful and/or informative. Also, I will gladly hear from you what to fix or add , Cheers
Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
On December 02 2012 07:04 corpuscle wrote: Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
That's the percentage of games won by the race out of total number of games played, not an average win %.
On December 02 2012 07:04 corpuscle wrote: Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
There were many mirrors among multirace MU's, so overall result is highly affected, thus downed. Formula for zerg: (wins vT+ vP+ vZ) / (ZvTs + ZvPs + ZvZs)
On December 02 2012 07:04 corpuscle wrote: Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
That's the percentage of games won by the race out of total number of games played, not an average win %.
The numbers still don't make any sense, though. If Terran is winning 31% of their vZ, 55% of their vP, and 50% of their vT (obviously), how does it end up as 24% overall?
If I won 9-1 tvz, I have a 90% winrate. If I won 1-4 tvp, I have a 20% winrate.
I have a 66.67% (10/15) total winrate, even though the average of the 2 is 55% (110/2).
That's not how you calculate an average in a situation like that. You've played twice as many TvZs, so you use a weighted average, i.e. (2*90 + 20)/3 = 66.67
On December 02 2012 07:04 corpuscle wrote: Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
That's the percentage of games won by the race out of total number of games played, not an average win %.
The numbers still don't make any sense, though. If Terran is winning 31% of their vZ, 55% of their vP, and 50% of their vT (obviously), how does it end up as 24% overall?
Because we had much less terran games against other races, thus in general they will have less wins.
On December 02 2012 07:04 corpuscle wrote: Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
That's the percentage of games won by the race out of total number of games played, not an average win %.
The numbers still don't make any sense, though. If Terran is winning 31% of their vZ, 55% of their vP, and 50% of their vT (obviously), how does it end up as 24% overall?
Because we had much less terran games against other races, thus in general they will have less wins.
...so are you calculating "terran winrate" as "number of games terran won"/"total games played in general"? It explains your numbers, at least, but kind of makes them useless...
On December 02 2012 07:14 xPabt wrote: Including mirrors makes no sense.
Imo, not at all. Because mirrors still stands for decent number of matches, so if we want to have realistic statistics we need to consider them. Of course ZvZ always mean that Z wins but if we want to calculate specific race wins in general we need to include all matches.
On December 02 2012 07:04 corpuscle wrote: Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
That's the percentage of games won by the race out of total number of games played, not an average win %.
The numbers still don't make any sense, though. If Terran is winning 31% of their vZ, 55% of their vP, and 50% of their vT (obviously), how does it end up as 24% overall?
If I won 9-1 tvz, I have a 90% winrate. If I won 1-4 tvp, I have a 20% winrate.
I have a 66.67% (10/15) total winrate, even though the average of the 2 is 55% (110/2).
That's not how you calculate an average in a situation like that. You've played twice as many TvZs, so you use a weighted average, i.e. (2*90 + 20)/3 = 66.67
You're telling me that doing 10/15 does not calculate your total average win percentage..? Total wins / total games = total win rate. Don't need to weight anything. There is more than 1 way of doing simple calculations.
Aside from being interesting using data like this from a single tournament is pointless. Sample size way to small.
Sample size? He used the population. There can not physically be a bigger "sample".
On December 02 2012 07:14 xPabt wrote: Including mirrors makes no sense.
Imo, not at all. Because mirrors still stands for decent number of matches, so if we want to have realistic statistics we need to consider them. Of course ZvZ always mean that Z wins but if we want to calculate specific race wins in general we need to include all matches.
You don't include mirrors because it makes the numbers misleading. Whichever race plays the most mirrors (in this case, Zerg) is going to have their number somewhat arbitrarily skewed towards 50%.
If, let's say, Zerg was 15-5 vT, 15-5 vP, and 30-30 vZ, they'd have a 60% overall winrate, which doesn't sound awful, but covers up the fact that they're actually winning 75% of the time in non-mirrors.
You're telling me that doing 10/15 does not calculate your total average win percentage..? Total wins / total games = total win rate. Don't need to weight anything. There is more than 1 way of doing simple calculations.
10/15 is the winrate. I was telling you that you calculated your "average" wrong. You have to use a weight or else it's a meaningless stat.
so fucking imbalance it isn't even funny. at this point, where money is involved in a huge stake and the situation is not looked into for, what?, over a few months already shows how disappointing the lack of adjustments are in the game despite the said results.
Zenbrez said it badly, let me tell it more clearly: You win TvP 1-4 (20% winrate) You win TvZ 9-1 (90% winrate) Now If I were to make an average of these two, without weighing in the amount of games played, my winrate is 55% But actually, I've won 10 games and lost 5, that means that my winrate is 66.6% (10/15*100) the OP factored in the amount of games played, therefore, the average is skewed to one side or another, no surprise. BTW, this is grade school mathematics and I'm kind of ashamed I'm forced to explain this kind of thing to tl users, I thought you guys were smart!
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
we don't because as long as money is involved, and the situation isn't addressed, we clearly dont have enough whining threads to complain about a problem. how else can a problem be addressed
On December 02 2012 07:14 xPabt wrote: Including mirrors makes no sense.
Imo, not at all. Because mirrors still stands for decent number of matches, so if we want to have realistic statistics we need to consider them. Of course ZvZ always mean that Z wins but if we want to calculate specific race wins in general we need to include all matches.
You don't include mirrors because it makes the numbers misleading. Whichever race plays the most mirrors (in this case, Zerg) is going to have their number somewhat arbitrarily skewed towards 50%.
If, let's say, Zerg was 15-5 vT, 15-5 vP, and 30-30 vZ, they'd have a 60% overall winrate, which doesn't sound awful, but covers up the fact that they're actually winning 75% of the time in non-mirrors.
Thanks for contribution, of course you are right, but I think that Overall is still good indicator (but somewhat confusing at first glance). We can read from it how specific race deals in bracket by how many games and wins generates (even in mirror).
On December 02 2012 07:22 MicroTastiC wrote: so fucking imbalance it isn't even funny. at this point, where money is involved in a huge stake and the situation is not looked into for, what?, over a few months already shows how disappointing the lack of adjustments are in the game despite the said results.
People like you should be perma banned from internet, you dont understand stats op posted and you draw some awful conclusion just to back up your balance whining. Stupid Blizzard if only they knew that there are so many smart guys like you that they can hire and make game perfect.
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
we don't because as long as money is involved, and the situation isn't addressed, we clearly dont have enough whining threads to complain about a problem. how else can a problem be addressed
Well excuse me for being tired of reading the same thing about 100times/hour.
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
we don't because as long as money is involved, and the situation isn't addressed, we clearly dont have enough whining threads to complain about a problem. how else can a problem be addressed
Well excuse me for being tired of reading the same thing about 100times/hour.
Leave the internet, it is pretty much one long regurgitation of the same shit invented by a few smart people.
The overall is the number of games won by a race divided by the total number of games played whatever the races. for example number of games won by a Z dividede by total of games played. It is clearly defined in the OP.
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
Sigh, i wish people would fucking read for once. I NEVER said the game was balanced so please stop trying to make up i did. I know the game is imbalanced atm but i am just so tired of reading it over and over and over and over.
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
we don't because as long as money is involved, and the situation isn't addressed, we clearly dont have enough whining threads to complain about a problem. how else can a problem be addressed
you obviously have no idea how blizzard operates, they arent gonna be like "hey guys i saw 50 threads on balance this week on TL !" "nah who cares" - "this week there was 75 threads!" "MOTHER OF GOD NERF ZERG NOW"
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
We'll have to wait until the head emerges from the bunghole.
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
we don't because as long as money is involved, and the situation isn't addressed, we clearly dont have enough whining threads to complain about a problem. how else can a problem be addressed
you obviously have no idea how blizzard operates, they arent gonna be like "hey guys i saw 50 threads on balance this week on TL !" "nah who cares" - "this week there was 75 threads!" "MOTHER OF GOD NERF ZERG NOW"
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
Sigh, i wish people would fucking read for once. I NEVER said the game was balanced so please stop trying to make up i did.
And here YOU go, whining about balance.
See how annoying it is to be accused of "whining" when all you're doing is observing a recent trend that favours one race?
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
Sigh, i wish people would fucking read for once. its over one tournament........... say the top 3 players at this tournament are zerg, then those 3 players are gonna win a lot more than they lose, and will completely destroy the win rates in zergs favour.
Statistics are no measure of balance. even if you have the stats from an entire month, leenock and life probably have a combined winrate of something like 80% over 40 games or even more games this month, thats because they are 2 of the best palyers atm, that completely squeues balance stats.
The Overall calculation is combining the win rate calculation and the race distribution calculation (not shown in the OP). It is also affected by the distribution of matchups as some people have already pointed out. But I think the impact it's supposed to have is that fans of zerg watching the tournament get to see their race win in over half the games played (there are lots of zergs in the bracket and they often win) while protoss fans and terran fans each see their race win less than a quarter of the time. Of course games in which your race isn't played prevent your race from winning (terran and protoss can't get wins in a ZvZ, and zerg will just get closer to 50% after a ZvZ) but this is significant in an elimination bracket, not totally random.
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
Sigh, i wish people would fucking read for once. its over one tournament........... say the top 3 players at this tournament are zerg, then those 3 players are gonna win a lot more than they lose, and will completely destroy the win rates in zergs favour.
Statistics are no measure of balance. even if you have the stats from an entire month, leenock and life probably have a combined winrate of something like 80% over 40 games or even more games this month, thats because they are 2 of the best palyers atm, that completely squeues balance stats.
Such a bad example i can't even believe it.I can make this stupid excuse about mvp, taeja, rain, parting and etc. to the point where terran and protoss will have close to zero win ratio in tournaments.
its over one tournament........... say the top 3 players at this tournament are zerg, then those 3 players are gonna win a lot more than they lose, and will completely destroy the win rates in zergs favour.
this, this, 100 times this. i mean it's a nice thread and everything, but anyone using raw percentages from a small sample size without even considering quality of players in each race is ridiculous.
On December 02 2012 07:04 corpuscle wrote: Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
That's the percentage of games won by the race out of total number of games played, not an average win %.
The numbers still don't make any sense, though. If Terran is winning 31% of their vZ, 55% of their vP, and 50% of their vT (obviously), how does it end up as 24% overall?
If I won 9-1 tvz, I have a 90% winrate. If I won 1-4 tvp, I have a 20% winrate.
I have a 66.67% (10/15) total winrate, even though the average of the 2 is 55% (110/2).
That's not how you calculate an average in a situation like that. You've played twice as many TvZs, so you use a weighted average, i.e. (2*90 + 20)/3 = 66.67
Mirror Matches are always 50% so they bring the overall win rate closer to 50. By how much is dependent on how many mirror matches of each there was, i.e. 30 PvPs vs 60 ZvZs means the ZvZ will be pulled closer to 50% for zerg than for Protoss.
OP you're seriously so incomprehensibly stupid i don't know where to begin.
The tournament isn't even over yet but you stressed that you were looking for the "Overall" winrates? Is this a really elaborate joke or something? That's going to skew the results immensely, THERE ARE STILL GAMES THAT HAVE TO BE PLAYED.
I wish people would actually spend time playing the game instead of coming up with the most retarded ways to justify their balance complaints.
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
Sigh, i wish people would fucking read for once. I NEVER said the game was balanced so please stop trying to make up i did. I know the game is imbalanced atm but i am just so tired of reading it over and over and over and over.
Well the rest of us are actually tired of the game being imbalanced. I won't get tired of hearing that until the game is fixed tbh.
On December 02 2012 07:59 Swagasaurus wrote: OP you're seriously so incomprehensibly stupid i don't know where to begin.
The tournament isn't even over yet but you stressed that you were looking for the "Overall" winrates? Is this a really elabate joke or something? That's going to skew the results immensely, THERE ARE STILL GAMES THAT HAVE TO BE PLAYED.
I wish people would actually spend time playing the game instead of coming up with the most retarded ways to justify their balance complaints.
maybe you are incomprehensibly stupid to get so mad about non-finished tournament statistics? or maybe you are so mad because these statistics leads to a conclusion which can not be admitted by zerg players ?
On December 02 2012 07:59 Swagasaurus wrote: OP you're seriously so incomprehensibly stupid i don't know where to begin.
The tournament isn't even over yet but you stressed that you were looking for the "Overall" winrates? Is this a really elaborate joke or something? That's going to skew the results immensely, THERE ARE STILL GAMES THAT HAVE TO BE PLAYED.
I wish people would actually spend time playing the game instead of coming up with the most retarded ways to justify their balance complaints.
Thanks for your opinion, althrough expressed in not the best way Also I explained in the OP why I didn't included Final Bracket. You completely got me wrong, sir.
On December 02 2012 07:59 Swagasaurus wrote: OP you're seriously so incomprehensibly stupid i don't know where to begin.
The tournament isn't even over yet but you stressed that you were looking for the "Overall" winrates? Is this a really elaborate joke or something? That's going to skew the results immensely, THERE ARE STILL GAMES THAT HAVE TO BE PLAYED.
I wish people would actually spend time playing the game instead of coming up with the most retarded ways to justify their balance complaints.
Thanks for your opinion, althrough expressed in not the best way Also I explained in the OP why I didn't included Final Bracket. You completely got me wrong, sir.
Thanks for the thread, it's too bad about half of the people that are responding didn't even read the entire OP.
its over one tournament........... say the top 3 players at this tournament are zerg, then those 3 players are gonna win a lot more than they lose, and will completely destroy the win rates in zergs favour.
this, this, 100 times this. i mean it's a nice thread and everything, but anyone using raw percentages from a small sample size without even considering quality of players in each race is ridiculous.
IPL is the most stacked tournament to date (outside of GSL). There are a lot of top terrans, lots of top protoss (maybe parting is the only one I can think of missing), and a lot of top zergs. The distribution of top tier players of each race is pretty even. Just look at each race: http://wiki.teamliquid.net/starcraft2/IGN_ProLeague_Season_5. There is no way anyone can believe that there are more top zergs than the other races.
On December 02 2012 07:59 Swagasaurus wrote: OP you're seriously so incomprehensibly stupid i don't know where to begin.
The tournament isn't even over yet but you stressed that you were looking for the "Overall" winrates? Is this a really elaborate joke or something? That's going to skew the results immensely, THERE ARE STILL GAMES THAT HAVE TO BE PLAYED.
I wish people would actually spend time playing the game instead of coming up with the most retarded ways to justify their balance complaints.
Thanks for your opinion, althrough expressed in not the best way Also I explained in the OP why I didn't included Final Bracket. You completely got me wrong, sir.
Thanks for the thread, it's too bad about half of the people that are responding didn't even read the entire OP.
The OP doesn't even contain a single assertion of balance, and I agree that it's fairly unbiased. The OP does however ask for input regarding things that are wrong with the calculations, and one thing that is wrong about them (or misleading) is a)"Overall" includes mirrors and b)The tournament isn't over.
On December 02 2012 07:59 Swagasaurus wrote: OP you're seriously so incomprehensibly stupid i don't know where to begin.
The tournament isn't even over yet but you stressed that you were looking for the "Overall" winrates? Is this a really elaborate joke or something? That's going to skew the results immensely, THERE ARE STILL GAMES THAT HAVE TO BE PLAYED.
I wish people would actually spend time playing the game instead of coming up with the most retarded ways to justify their balance complaints.
Thanks for your opinion, althrough expressed in not the best way Also I explained in the OP why I didn't included Final Bracket. You completely got me wrong, sir.
Thanks for the thread, it's too bad about half of the people that are responding didn't even read the entire OP.
The OP doesn't even contain a single assertion of balance, and I agree that it's fairly unbiased. The OP does however ask for input regarding things that are wrong with the calculations, and one thing that is wrong about them (or misleading) is a)"Overall" includes mirrors and b)The tournament isn't over.
For b), the last three sets of games will have no impact compared to the hundreds that have already been played.
Overall is a nearly worthless stat. Reason - it is a positive feedback loop if one race is slightly imba and/or happens to win more in the first round or two. Then more of the race distribution will be of that race, and overall counts mirrors - thus that race's overall win% will go up.
On December 02 2012 07:59 Swagasaurus wrote: OP you're seriously so incomprehensibly stupid i don't know where to begin.
The tournament isn't even over yet but you stressed that you were looking for the "Overall" winrates? Is this a really elaborate joke or something? That's going to skew the results immensely, THERE ARE STILL GAMES THAT HAVE TO BE PLAYED.
I wish people would actually spend time playing the game instead of coming up with the most retarded ways to justify their balance complaints.
Thanks for your opinion, althrough expressed in not the best way Also I explained in the OP why I didn't included Final Bracket. You completely got me wrong, sir.
Thanks for the thread, it's too bad about half of the people that are responding didn't even read the entire OP.
The OP doesn't even contain a single assertion of balance, and I agree that it's fairly unbiased. The OP does however ask for input regarding things that are wrong with the calculations, and one thing that is wrong about them (or misleading) is a)"Overall" includes mirrors and b)The tournament isn't over.
Yes, I should maybe rename Overall or even remove it. As for b), convention has right, data from final bracket is irrevelant because there are 10 matches left (from over 100) with 2 terrans and 8 zerg fighting.
the results are pretty one sided. i wonder if blizzard will exclude this piece of info so they can keep saying that zerg does not have an advantage globally or whatever.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
On December 02 2012 08:31 EtherealDeath wrote: Overall is a nearly worthless stat. Reason - it is a positive feedback loop if one race is slightly imba and/or happens to win more in the first round or two. Then more of the race distribution will be of that race, and overall counts mirrors - thus that race's overall win% will go up.
It's actually a "negative" feedback loop. If Zerg wins more games, there will be more ZvZs, which, by nature, are going to always be 50% winrate overall. If a race is winning non-mirrors more than 50%, mirrors being included brings the overall winrate down, not up.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
just saying... :D
Because Europe is backwards and weird (they use commas instead of decimal points and vice versa, iirc)
On December 02 2012 08:31 EtherealDeath wrote: Overall is a nearly worthless stat. Reason - it is a positive feedback loop if one race is slightly imba and/or happens to win more in the first round or two. Then more of the race distribution will be of that race, and overall counts mirrors - thus that race's overall win% will go up.
Overall stats might be non significant but surely other stats are not (affected by race distribution)?
Aside from being interesting using data like this from a single tournament is pointless. Sample size way to small.
Sample size? He used the population. There can not physically be a bigger "sample".
As you say, the sample used is the entire population of games at IPL, so we can, based on the sample, conclusively say that Zerg has done better than Terran on average. However, when people say the sample size is too small, they mean it's too small to draw conclusions regarding the overall balance (or perhaps just the overall pro-level balance) of Starcraft 2. In this case, the tournament is only a small sample of all the games played at a professional level, and it's probably too small to draw any strong conclusions from. Combining the results from the IPL5 with those from other more recent tournaments does provide fairly strong evidence that Zerg is too strong at the moment, but then again, not many people are denying that anymore.
Regarding overall winrates, it's expected to be 33%/33%/33% if there wre equal numbers of all races. However, since there weren't an equal number of players from each race, this is not the case. A more useful statistic would be to look at how significantly each race has over or under-performed its expected rate.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
just saying... :D
Because Europe is awesome and right (they use commas instead of decimal points and vice versa, iirc, because that's the proper way to do it)
On December 02 2012 07:59 Swagasaurus wrote: OP you're seriously so incomprehensibly stupid i don't know where to begin.
The tournament isn't even over yet but you stressed that you were looking for the "Overall" winrates? Is this a really elaborate joke or something? That's going to skew the results immensely, THERE ARE STILL GAMES THAT HAVE TO BE PLAYED.
I wish people would actually spend time playing the game instead of coming up with the most retarded ways to justify their balance complaints.
User was warned for this post
This is why the op should be read in full. That said op thank you for the stats, think we could all just take a second to breath and let P and T adjust and watchthe stats even out overtime.
The overall I feel makes alit of sense to include simply because the other stats are seperated before hand. As stated by others please no more balance stuff please.
I'm not surprised at these statistics. It's this point it's laughable at how blatantly overpowered Zerg has been and still is consistently for such a long period of time. How can anyone play Zerg and publicly admit it?
After the latest "update" from Blizzard I refused to watch SC2 streams anymore, but as a hardcore fan I'm still concerned about the future of the game. Blizzard needs a new balance team.
On December 02 2012 08:40 urbaNo wrote: the results are pretty one sided. i wonder if blizzard will exclude this piece of info so they can keep saying that zerg does not have an advantage globally or whatever.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
just saying... :D
Hah, I rarely pay attention to details unfortunately (believe me, I am not proud of it :D ).
Backing to topic, I don't believe if my results tell what is op and what is up. For example: "Overall" obviously favors zerg, but what explanation stands behind? I used all games played as denominator. At IPL5 the most players are playing zerg, thus generate the most games and (eventually) wins. Liquid'NonY explained it better than I could. In case if you need further explanation - his post is somewhere in 3rd page
On December 02 2012 08:31 EtherealDeath wrote: Overall is a nearly worthless stat. Reason - it is a positive feedback loop if one race is slightly imba and/or happens to win more in the first round or two. Then more of the race distribution will be of that race, and overall counts mirrors - thus that race's overall win% will go up.
It's actually a "negative" feedback loop. If Zerg wins more games, there will be more ZvZs, which, by nature, are going to always be 50% winrate overall. If a race is winning non-mirrors more than 50%, mirrors being included brings the overall winrate down, not up.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
just saying... :D
Because Europe is backwards and weird (they use commas instead of decimal points and vice versa, iirc)
No you don't get it. His overall is actually not overall win % out of all a race's games. It's (# wins won by X race) / (#total number of games played - every game in the tourney regardless of race matchup). So actually a mirror counts as 100%, which drives the number up. Thus it's a positive feedback loop. Notice how the overall % numbers add up to 100%.
On December 02 2012 08:31 EtherealDeath wrote: Overall is a nearly worthless stat. Reason - it is a positive feedback loop if one race is slightly imba and/or happens to win more in the first round or two. Then more of the race distribution will be of that race, and overall counts mirrors - thus that race's overall win% will go up.
It's actually a "negative" feedback loop. If Zerg wins more games, there will be more ZvZs, which, by nature, are going to always be 50% winrate overall. If a race is winning non-mirrors more than 50%, mirrors being included brings the overall winrate down, not up.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
just saying... :D
Because Europe is backwards and weird (they use commas instead of decimal points and vice versa, iirc)
No you don't get it. His overall is actually not overall win % out of all a race's games. It's (# wins won by X race) / (#total number of games played - every game in the tourney regardless of race matchup). So actually a mirror oucnts as 100%, which drives the number up. Thus it's a positive feedback loop. Notice how the overall % numbers add up to 100%.
So the stats are still wrong in the opposite direction.
On December 02 2012 08:31 EtherealDeath wrote: Overall is a nearly worthless stat. Reason - it is a positive feedback loop if one race is slightly imba and/or happens to win more in the first round or two. Then more of the race distribution will be of that race, and overall counts mirrors - thus that race's overall win% will go up.
It's actually a "negative" feedback loop. If Zerg wins more games, there will be more ZvZs, which, by nature, are going to always be 50% winrate overall. If a race is winning non-mirrors more than 50%, mirrors being included brings the overall winrate down, not up.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
just saying... :D
Because Europe is backwards and weird (they use commas instead of decimal points and vice versa, iirc)
No you don't get it. His overall is actually not overall win % out of all a race's games. It's (# wins won by X race) / (#total number of games played - every game in the tourney regardless of race matchup). So actually a mirror oucnts as 100%, which drives the number up. Thus it's a positive feedback loop. Notice how the overall % numbers add up to 100%.
So the stats are still wrong in the opposite direction.
What do you mean they are still wrong in the opposite direction? His stats are correct.
edit - Though I'm not sure what his source is, as his % wins don't match up with the "source" in the OP.
On December 02 2012 08:31 EtherealDeath wrote: Overall is a nearly worthless stat. Reason - it is a positive feedback loop if one race is slightly imba and/or happens to win more in the first round or two. Then more of the race distribution will be of that race, and overall counts mirrors - thus that race's overall win% will go up.
It's actually a "negative" feedback loop. If Zerg wins more games, there will be more ZvZs, which, by nature, are going to always be 50% winrate overall. If a race is winning non-mirrors more than 50%, mirrors being included brings the overall winrate down, not up.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
just saying... :D
Because Europe is backwards and weird (they use commas instead of decimal points and vice versa, iirc)
No you don't get it. His overall is actually not overall win % out of all a race's games. It's (# wins won by X race) / (#total number of games played - every game in the tourney regardless of race matchup). So actually a mirror counts as 100%, which drives the number up. Thus it's a positive feedback loop. Notice how the overall % numbers add up to 100%.
Ah, yeah, okay, I see what you're saying. I thought you meant winrates in a more general sense, not the specific (and really odd) way that the OP did it.
OP should delete the misleading stats. This guy is right:
On December 02 2012 07:20 corpuscle wrote: You don't include mirrors because it makes the numbers misleading. Whichever race plays the most mirrors (in this case, Zerg) is going to have their number somewhat arbitrarily skewed towards 50%.
If, let's say, Zerg was 15-5 vT, 15-5 vP, and 30-30 vZ, they'd have a 60% overall winrate, which doesn't sound awful, but covers up the fact that they're actually winning 75% of the time in non-mirrors.
On December 02 2012 08:31 EtherealDeath wrote: Overall is a nearly worthless stat. Reason - it is a positive feedback loop if one race is slightly imba and/or happens to win more in the first round or two. Then more of the race distribution will be of that race, and overall counts mirrors - thus that race's overall win% will go up.
It's actually a "negative" feedback loop. If Zerg wins more games, there will be more ZvZs, which, by nature, are going to always be 50% winrate overall. If a race is winning non-mirrors more than 50%, mirrors being included brings the overall winrate down, not up.
also kinda annoying that you did all this precise calculation, but then you used commas as opposed to periods for your decimal places in your percentages??? wtf?
just saying... :D
Because Europe is backwards and weird (they use commas instead of decimal points and vice versa, iirc)
No you don't get it. His overall is actually not overall win % out of all a race's games. It's (# wins won by X race) / (#total number of games played - every game in the tourney regardless of race matchup). So actually a mirror counts as 100%, which drives the number up. Thus it's a positive feedback loop. Notice how the overall % numbers add up to 100%.
Ah, yeah, okay, I see what you're saying. I thought you meant winrates in a more general sense, not the specific (and really odd) way that the OP did it.
Yea his overall is a bit odd, but it is what it is xD
On December 02 2012 07:04 corpuscle wrote: I don't really see how a 69% vT and 61% vP can average out to 52%, for example
I don't understand either...
If I won 9-1 tvz, I have a 90% winrate. If I won 1-4 tvp, I have a 20% winrate.
I have a 66.67% (10/15) total winrate, even though the average of the 2 is 55% (110/2).
Wrong math broski. Both of the winrate above were above 52% so there is no way the average can be less when both the numbers that make it up are over 52%
On December 02 2012 07:20 corpuscle wrote: You don't include mirrors because it makes the numbers misleading. Whichever race plays the most mirrors (in this case, Zerg) is going to have their number somewhat arbitrarily skewed towards 50%.
If, let's say, Zerg was 15-5 vT, 15-5 vP, and 30-30 vZ, they'd have a 60% overall winrate, which doesn't sound awful, but covers up the fact that they're actually winning 75% of the time in non-mirrors.
Technically, what I was talking about isn't how the OP calculated the "overall" stat anyway, I kind of gave him too much credit. He did "number of games won by X race" divided by "number of games played overall" which is just a weird way to do it because it factors in games that the race being discussed wasn't even in... that's why the T and P "overall" scores are so abysmally low even though their actual winrates were much higher.
edit:
It would be nice to see the amount of games played next to each stat.
Yeah, have to echo this. OP, if you still have the raw numbers you used, that'd be really helpful, since we can calculate actual winrates and stuff.
On December 02 2012 10:05 achan1058 wrote: Seriously, if this is such a problem, there's no need to wait for blizzard. The tournament organizers can throw back in Steppes and be done.
Wait, solving balance through maps? That sounds like blasphemy to me.
On December 02 2012 10:05 achan1058 wrote: Seriously, if this is such a problem, there's no need to wait for blizzard. The tournament organizers can throw back in Steppes and be done.
Wait, solving balance through maps? That sounds like blasphemy to me.
So, if Terran plays on Steppes they win, and if they don't play on steppes they lose?
On December 02 2012 10:05 achan1058 wrote: Seriously, if this is such a problem, there's no need to wait for blizzard. The tournament organizers can throw back in Steppes and be done.
Wait, solving balance through maps? That sounds like blasphemy to me.
So, if Terran plays on Steppes they win, and if they don't play on steppes they lose?
On December 02 2012 10:05 achan1058 wrote: Seriously, if this is such a problem, there's no need to wait for blizzard. The tournament organizers can throw back in Steppes and be done.
Wait, solving balance through maps? That sounds like blasphemy to me.
So, if Terran plays on Steppes they win, and if they don't play on steppes they lose?
How is that balance?
I see more 50/50 games on huge maps with a lot of expansion which can be dropped than on small maps where zerg reinforcements comes too fast for terran to hold pushes
Because on huge maps : - Terran can harass - Terran and zerg can macro up - Zerg broodlords comes but the map is huge, and terran has time to prepare
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
Sigh, i wish people would fucking read for once. I NEVER said the game was balanced so please stop trying to make up i did. I know the game is imbalanced atm but i am just so tired of reading it over and over and over and over.
you realize you clicked this thread without being forced too. we are discussing win% and there is only 1 reason to do this, to study the balance/imbalance, if everything was always 50% no theads would be made about it. Its like you walk into a bar and saying your so tired of seeing drunk people. your a fuckin tool
On December 02 2012 10:05 achan1058 wrote: Seriously, if this is such a problem, there's no need to wait for blizzard. The tournament organizers can throw back in Steppes and be done.
Wait, solving balance through maps? That sounds like blasphemy to me.
So, if Terran plays on Steppes they win, and if they don't play on steppes they lose?
How is that balance?
I see more 50/50 games on huge maps with a lot of expansion which can be dropped than on small maps where zerg reinforcements comes too fast for terran to hold pushes
Because on huge maps : - Terran can harass - Terran and zerg can macro up - Zerg broodlords comes but the map is huge, and terran has time to prepare
True on certain map designs, not true on others, and it's unrelated to the pure size. Terran's iirc have a better winrate on say, Whirlwind (or at least Ryung does), than say Daybreak. Also, part of people's conceptions of map balance are so far off base it's insane.
People look for what they want to find, in terms of how it fits into their intuitive feeling. In the case of a racial matchup being statistically balanced, people will look at 'how the game flows' to take one example.
You can't use just stats, or just watch games. You have to do both if you're going to talk about balance. Consider the relationship between stats, and racially favoured maps. It doesn't relate to the stubborn perceptions that remain from previous eras and metagames.
Daybreak: Considered 'slightly' Zerg favoured, by something in the ballpark of 61-39 Antiga: The map from current map pool most people whine about as being Terran favoured is something like, 52-48 in favour of T
These numbers may be wrong by one or two, but the rough ratios are close, I'm nearly sure. Couldn't find the stats that I remember another OP posted.
On December 02 2012 07:28 Assirra wrote: Great, another balance whining thread. Clearly we didn't have enough of those...
70% ZvT is pretty fucking alarming. How long are we gonna watch ZvZ finals and completely 1 sided rapes until you give up trying to dispel the truth lol?
Sigh, i wish people would fucking read for once. I NEVER said the game was balanced so please stop trying to make up i did. I know the game is imbalanced atm but i am just so tired of reading it over and over and over and over.
you realize you clicked this thread without being forced too. we are discussing win% and there is only 1 reason to do this, to study the balance/imbalance, if everything was always 50% no theads would be made about it. Its like you walk into a bar and saying your so tired of seeing drunk people. your a fuckin tool
Even win percentages are correlative to balance being achieved, but don't tell you if something is balanced in how the community perceives what balance IS.
PvT is really fucking balanced these days, statistically at many differing levels. However, there's asymmetry in difficulty between how hard it is to win with T and P respectively. The rough relationship is (as some people that aren't me put it)
T- Win earlier, or outplay your opponent really hard if you're going to lategame P- Have a tougher early game, but if you survive the lategame is easier to win for you than you opponent.
I was very surprised when I woke up to watch the grand finals and the views were at 27K, when previously they had rose up to 35K. Then I saw it was zvz and understood why. The last game was epic. But this is becoming a problem. Balance is actually hurting this game.
Coming from a FG community where competitive games are almost never balanced. This stuff ruins a community.
Considering that BW was pretty balanced. This issue might be new to the starcraft 2 community. But veterans can see for themselves the kind of atmosphere this creates.
Use whatever stats/information to get blizzard to be more proactive.
On December 02 2012 16:00 KingAce wrote: I was very surprised when I woke up to watch the grand finals and the views were at 27K, when previously they had rose up to 35K. Then I saw it was zvz and understood why. The last game was epic. But this is becoming a problem. Balance is actually hurting this game.
Coming from a FG community where competitive games are almost never balanced. This stuff ruins a community.
Considering that BW was pretty balanced. This issue might be new to the starcraft 2 community. But veterans can see for themselves the kind of atmosphere this creates.
Use whatever stats/information to get blizzard to be more proactive.
Blizz has their own stats. S'long as the matchups have ~50% winrates, they're happy. Nevermind the fact the immortal sentry is way to good, and lategame zvp is way to good.
Not saying that sc2 is balance or inbalanced, but this is a completely pointless thread. Like a lot of "statistics" threads in sc2.
1. Include errors There are not that many games. The error is important. How does Op handle it? Ignore it. Well done... If you would have bother to estimate the error, you would have understood how silly it is to include 2 decimals, and how little these numbers mean. (Hint, 70.23% +- 15% is silly) If only we could have more games to look at, maybe even all games in TLPD... Oh, wait, that thread is already done on a monthly basis, and these number are just a biased subset of that data, but made with horrible analysis and presented in a horrible way... :noo:
2. Bias Would this thread be here if the percentages were around 50%? No, of course not. People don't make separate threads for win rates of each tournament. only for the tournaments that happen to be extra in favour of one race (or nationality) or the other. Which is why we should not care about these threads, but wait for the monthly summary of the month, which is a close to entirely unbiased sample.
So zerg may be OP, but it is definitely not proven by publishing selected, statistically insignificant number, and showing them without any error analysis.
On December 02 2012 16:00 KingAce wrote: I was very surprised when I woke up to watch the grand finals and the views were at 27K, when previously they had rose up to 35K. Then I saw it was zvz and understood why. The last game was epic. But this is becoming a problem. Balance is actually hurting this game.
Coming from a FG community where competitive games are almost never balanced. This stuff ruins a community.
Considering that BW was pretty balanced. This issue might be new to the starcraft 2 community. But veterans can see for themselves the kind of atmosphere this creates.
Use whatever stats/information to get blizzard to be more proactive.
A lot of people were watching through the GOM stream so those numbers didn't mean that much.
I would really like to know how many % of the winrates pvz is late game and early game. So how many % from the games did protoss win with immortal all ins (or any other all in on 2 bases with no intention taking third). And how many wins did protoss get when zerg goes infestor BL composition.
also the tvz wins include games like sculp-catz, mma-daisuki, bogus-massan and arguably more, which are basically players many leagues apart, no offence.
On December 02 2012 22:28 QuixoticO wrote: I like how it went from early WoL "there's just a lot of good pro terrans that's why they win so much" to "zerg is imba just look at them winrates".
Not that hard to concile the 2 when you see the list of nerfs to T and buffs to other races.
Also, the only players to ever have wrists problems are Terrans. Coincidence?
CranK doesn't count as his is a neck problem and was long before he even touched SC2 iirc.
On December 02 2012 22:25 topschutter wrote: I would really like to know how many % of the winrates pvz is late game and early game. So how many % from the games did protoss win with immortal all ins (or any other all in on 2 bases with no intention taking third). And how many wins did protoss get when zerg goes infestor BL composition.
You basically have to watch the games since Protoss (just like Terran) can easily win the late game if they get there ahead. A successful 2 or 3 base push does not necessarily imply killing the Zerg.
This could be done, although it would be very hard to objectively define what "getting in the late game ahead" means.
On December 02 2012 22:35 Champi wrote: surprising to see Terran did better than Protoss overall this tournament. they recently havnt been doing so well outside of GSL.
zerg doing well is standard though
why are you surprised? the T who did well are GSL T.
On December 02 2012 22:25 topschutter wrote: I would really like to know how many % of the winrates pvz is late game and early game. So how many % from the games did protoss win with immortal all ins (or any other all in on 2 bases with no intention taking third). And how many wins did protoss get when zerg goes infestor BL composition.
You basically have to watch the games since Protoss (just like Terran) can easily win the late game if they get there ahead. A successful 2 or 3 base push does not necessarily imply killing the Zerg.
This could be done, although it would be very hard to objectively define what "getting in the late game ahead" means.
Obviously you have to watch the games... if the game goes on after a failed immortal all in and the protoss still wins the late game then it's not a win because of the immortal all in. Shouldn't be too hard to judge what won the game. Usually the game ends there tho when protoss all ins.
I've always thought that PvT was slightly in favor of P but looks like I was wrong, at least when I look at this tournament. Awesome job, I really like stats like that.
On December 02 2012 22:32 raga4ka wrote: So it's Z>T>P>Z . The opposite of BW .
No it's Z > All. The opposite of a balanced game.
I remember there was once a time where balance whining got people a temp-ban.
Logical claims based on rational and public data do not equal to "balance whines".
Its still balance whining none the less, nothing rational about this information , when you look at the quality of some of the players.
As for the tournament. The quality of production has been great and there has been some fantastic games been played. I Getting really hyped for the finals this evening
On December 02 2012 22:32 raga4ka wrote: So it's Z>T>P>Z . The opposite of BW .
No it's Z > All. The opposite of a balanced game.
I remember there was once a time where balance whining got people a temp-ban.
We are not in a LR thread you know.
Even though I did expect this kind of stats, it still makes me laugh to see such abysmal winrates despite the terrans being the likes of MMA, Bogus, ByuN and Polt.
definitely not a protosstournament, and although TvZ looks abyssmal as usual, i still feel warm and fuzzy from watching polt and bomber play yesterday... terran wins look much more epic and hard working compared to zerg, horray for being subjective!
Aside from being interesting using data like this from a single tournament is pointless. Sample size way to small.
Sample size? He used the population. There can not physically be a bigger "sample".
As you say, the sample used is the entire population of games at IPL, so we can, based on the sample, conclusively say that Zerg has done better than Terran on average. However, when people say the sample size is too small, they mean it's too small to draw conclusions regarding the overall balance (or perhaps just the overall pro-level balance) of Starcraft 2. In this case, the tournament is only a small sample of all the games played at a professional level, and it's probably too small to draw any strong conclusions from. Combining the results from the IPL5 with those from other more recent tournaments does provide fairly strong evidence that Zerg is too strong at the moment, but then again, not many people are denying that anymore.
Regarding overall winrates, it's expected to be 33%/33%/33% if there wre equal numbers of all races. However, since there weren't an equal number of players from each race, this is not the case. A more useful statistic would be to look at how significantly each race has over or under-performed its expected rate.
I was going to be a little more blunt with the guy but thanks to you looks like I don't have to.
On December 02 2012 10:05 achan1058 wrote: Seriously, if this is such a problem, there's no need to wait for blizzard. The tournament organizers can throw back in Steppes and be done.
Wait, solving balance through maps? That sounds like blasphemy to me.
So, if Terran plays on Steppes they win, and if they don't play on steppes they lose?
How is that balance?
That's how BW was balanced.
Noooooooooooooooooooooooo.
lol
-_-
Look, there have been plenty of imbalanced maps in the past BW tournaments but we don't go about our business in this way. In fact, there are plenty of maps which have very short-lived runs because they are found out to be too imbalanced. Unfortunately you still have some tournaments like WCG who never get the memo who use old maps and don't change shit.
Gorky Island anyone?
It's not as simple as that.
Blizzard are terrible at designing maps and they'll say, "Oh we're just doing that because we want people to try different strategies." The reality is... they never really had a great map design team and in many cases that's because they just don't frigging understand the game yet till the pro's start doing their thing.
"Oh, so that's how the game will play out."
They will either leave it or patch it. They rarely look at the maps and it's up to our community to play Mr. Fix It because they're too busy elsewhere.
The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
I did something similar here, but not as many tournaments, but more data is visible to those who want to look more into it for the ones I did do. (Code S, BWC, Lone Star and IPL (winner bracket + final bracket)
On December 03 2012 06:15 Zooper31 wrote: Am I not the only person who came into this thread thinking race had to do with actual races, Korea, American, Swedish, etc?
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
Again, this might not be completely accurate. But even if I missed a couple of games, it shouldn't make much of a difference.
I KNEW IT
Protoss players are full of shit!
Yes, because BWC which is carrying those stats had a lot of big korean Z.
Because they couldn't qualify -_- not to mention protoss is favoured at many other of these tournaments as well.
Just admit it. Toss players just like to whine for no reason
They have some reasons. No Protoss players in Code S top4, no Protoss players in IPL Top6, MLG Top4. These are tournaments lately with highest level of players.
On December 03 2012 06:15 Zooper31 wrote: Am I not the only person who came into this thread thinking race had to do with actual races, Korean, American, Swedish, etc? lol
No, since if we talk about player's actual race, it's going to be like 95% Korean, 5% foreigner or something.
Aside from being interesting using data like this from a single tournament is pointless. Sample size way to small.
Sample size? He used the population. There can not physically be a bigger "sample".
As you say, the sample used is the entire population of games at IPL, so we can, based on the sample, conclusively say that Zerg has done better than Terran on average. However, when people say the sample size is too small, they mean it's too small to draw conclusions regarding the overall balance (or perhaps just the overall pro-level balance) of Starcraft 2. In this case, the tournament is only a small sample of all the games played at a professional level, and it's probably too small to draw any strong conclusions from. Combining the results from the IPL5 with those from other more recent tournaments does provide fairly strong evidence that Zerg is too strong at the moment, but then again, not many people are denying that anymore.
Regarding overall winrates, it's expected to be 33%/33%/33% if there wre equal numbers of all races. However, since there weren't an equal number of players from each race, this is not the case. A more useful statistic would be to look at how significantly each race has over or under-performed its expected rate.
I was going to be a little more blunt with the guy but thanks to you looks like I don't have to.
On December 02 2012 10:05 achan1058 wrote: Seriously, if this is such a problem, there's no need to wait for blizzard. The tournament organizers can throw back in Steppes and be done.
Wait, solving balance through maps? That sounds like blasphemy to me.
So, if Terran plays on Steppes they win, and if they don't play on steppes they lose?
How is that balance?
That's how BW was balanced.
Noooooooooooooooooooooooo.
lol
-_-
Look, there have been plenty of imbalanced maps in the past BW tournaments but we don't go about our business in this way. In fact, there are plenty of maps which have very short-lived runs because they are found out to be too imbalanced. Unfortunately you still have some tournaments like WCG who never get the memo who use old maps and don't change shit.
Gorky Island anyone?
It's not as simple as that.
Blizzard are terrible at designing maps and they'll say, "Oh we're just doing that because we want people to try different strategies." The reality is... they never really had a great map design team and in many cases that's because they just don't frigging understand the game yet till the pro's start doing their thing.
"Oh, so that's how the game will play out."
They will either leave it or patch it. They rarely look at the maps and it's up to our community to play Mr. Fix It because they're too busy elsewhere.
I did not suggest that all BW maps are balance, but rather that you can skew balance one way or another by putting in certain maps, and that's what I am suggesting SC2 tournaments do as well. Throw away all the bigger maps, and put in some that's really small, small enough to guarantee a Terran favorite. If Terrans are too good after that, switch a few maps back. Heck, even if we just throw away every single map other than Antiga, the game's balance would change already. Aside from balance, there are good reason to throw away some maps, namely that the game is getting stale because of the lack of map changes already. (of course, this only works for tournaments, and will suck for anyone on ladder)
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
On December 02 2012 16:05 Cascade wrote: Not saying that sc2 is balance or inbalanced, but this is a completely pointless thread. Like a lot of "statistics" threads in sc2.
1. Include errors snip..
TL:DR error analysis pl0x.
Ok, this post makes my brain bleed. While I don't find the OP particularly interesting or important, the poster calculated the actual win rates for this tournament. He took all the games that were played and just divided two numbers. There is no error.
On December 02 2012 07:04 corpuscle wrote: Um, seems to me like you're factoring in mirror matchups in the "overall" score, unless I'm missing something big... I don't really see how a 69% vT and 61% vP can average out to 52%, for example
That's the percentage of games won by the race out of total number of games played, not an average win %.
The numbers still don't make any sense, though. If Terran is winning 31% of their vZ, 55% of their vP, and 50% of their vT (obviously), how does it end up as 24% overall?
If I won 9-1 tvz, I have a 90% winrate. If I won 1-4 tvp, I have a 20% winrate.
I have a 66.67% (10/15) total winrate, even though the average of the 2 is 55% (110/2).
That's not how you calculate an average in a situation like that. You've played twice as many TvZs, so you use a weighted average, i.e. (2*90 + 20)/3 = 66.67
You're telling me that doing 10/15 does not calculate your total average win percentage..? Total wins / total games = total win rate. Don't need to weight anything. There is more than 1 way of doing simple calculations.
Aside from being interesting using data like this from a single tournament is pointless. Sample size way to small.
Sample size? He used the population. There can not physically be a bigger "sample".
lol what a great way to misuse those two words.
Yes, it is the population of all the games of the tournament. Therefore his results are 100% true within the games of this tournament.
This tournament, however, is only a very small sample of all games played within recent times, hence the other poster pointed out that it was the sample was too small to say anything about the entire population of games played.
No one is surprised that these stats are completely true within this tournament, the interesting question to follow is whether they have any external validity for the balance situation of SC2. You're acting as if he is questioning the obvious.
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
Again, this might not be completely accurate. But even if I missed a couple of games, it shouldn't make much of a difference.
I KNEW IT
Protoss players are full of shit!
Yes, because BWC which is carrying those stats had a lot of big korean Z.
Because they couldn't qualify -_- not to mention protoss is favoured at many other of these tournaments as well.
Just admit it. Toss players just like to whine for no reason
Interesting. Tell me again how is the GSL %winrate at PvZ, which is the most stacked format out there.
since when does one tournament mean anything conclusive?
Since they have somewhat the best skill/race distribution. BWC ? WCG ? For all we can see IPL5 and GSL are the only tourneys stacked enough for this to happen.
No one is surprised that these stats are completely true within this tournament, the interesting question to follow is whether they have any external validity for the balance situation of SC2.
The game never should be balanced by stats, not even big ass stats. It should be balanced around how the game is played out. Nothing else.
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
Again, this might not be completely accurate. But even if I missed a couple of games, it shouldn't make much of a difference.
I KNEW IT
Protoss players are full of shit!
Yes, because BWC which is carrying those stats had a lot of big korean Z.
Because they couldn't qualify -_- not to mention protoss is favoured at many other of these tournaments as well.
Just admit it. Toss players just like to whine for no reason
Interesting. Tell me again how is the GSL %winrate at PvZ, which is the most stacked format out there.
since when does one tournament mean anything conclusive?
Since they have somewhat the best skill/race distribution. BWC ? WCG ? For all we can see IPL5 and GSL are the only tourneys stacked enough for this to happen.
No one is surprised that these stats are completely true within this tournament, the interesting question to follow is whether they have any external validity for the balance situation of SC2.
The game never should be balanced by stats, not even big ass stats. It should be balanced around how the game is played out. Nothing else.
I strongly disagree. Stats are quantitative, "played out" is qualitative. One of these do not have viewer bias. While looking at how things play out is important, it is extremely unscientific to ignore the stats.
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
Again, this might not be completely accurate. But even if I missed a couple of games, it shouldn't make much of a difference.
I KNEW IT
Protoss players are full of shit!
Yes, because BWC which is carrying those stats had a lot of big korean Z.
Because they couldn't qualify -_- not to mention protoss is favoured at many other of these tournaments as well.
Just admit it. Toss players just like to whine for no reason
Interesting. Tell me again how is the GSL %winrate at PvZ, which is the most stacked format out there.
since when does one tournament mean anything conclusive?
Since they have somewhat the best skill/race distribution. BWC ? WCG ? For all we can see IPL5 and GSL are the only tourneys stacked enough for this to happen.
No one is surprised that these stats are completely true within this tournament, the interesting question to follow is whether they have any external validity for the balance situation of SC2.
The game never should be balanced by stats, not even big ass stats. It should be balanced around how the game is played out. Nothing else.
I strongly disagree. Stats are quantitative, "played out" is qualitative. One of these do not have viewer bias. While looking at how things play out is important, it is extremely unscientific to ignore the stats.
Anybody with braincells will disagree with you playing and watching PvZ. A match up entirely balanced by stats is awful for the game. If the "bias" is about being fun, i don't see anything wrong with it since it's a game, not the final balance of a company.
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
Again, this might not be completely accurate. But even if I missed a couple of games, it shouldn't make much of a difference.
I KNEW IT
Protoss players are full of shit!
Yes, because BWC which is carrying those stats had a lot of big korean Z.
Because they couldn't qualify -_- not to mention protoss is favoured at many other of these tournaments as well.
Just admit it. Toss players just like to whine for no reason
Interesting. Tell me again how is the GSL %winrate at PvZ, which is the most stacked format out there.
since when does one tournament mean anything conclusive?
Since they have somewhat the best skill/race distribution. BWC ? WCG ? For all we can see IPL5 and GSL are the only tourneys stacked enough for this to happen.
No one is surprised that these stats are completely true within this tournament, the interesting question to follow is whether they have any external validity for the balance situation of SC2.
The game never should be balanced by stats, not even big ass stats. It should be balanced around how the game is played out. Nothing else.
I strongly disagree. Stats are quantitative, "played out" is qualitative. One of these do not have viewer bias. While looking at how things play out is important, it is extremely unscientific to ignore the stats.
Anybody with braincells will disagree with you playing and watching PvZ. A match up entirely balanced by stats is awful for the game. If the "bias" is about being fun, i don't see anything wrong with it since it's a game, not the final balance of a company.
I never said entirely balanced by stats. I said stats must take a significant factor. Besides, whether a match up is boring or not is not remotely the same as balance.
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
Again, this might not be completely accurate. But even if I missed a couple of games, it shouldn't make much of a difference.
I KNEW IT
Protoss players are full of shit!
Yes, because BWC which is carrying those stats had a lot of big korean Z.
Because they couldn't qualify -_- not to mention protoss is favoured at many other of these tournaments as well.
Just admit it. Toss players just like to whine for no reason
Interesting. Tell me again how is the GSL %winrate at PvZ, which is the most stacked format out there.
since when does one tournament mean anything conclusive?
Since they have somewhat the best skill/race distribution. BWC ? WCG ? For all we can see IPL5 and GSL are the only tourneys stacked enough for this to happen.
No one is surprised that these stats are completely true within this tournament, the interesting question to follow is whether they have any external validity for the balance situation of SC2.
The game never should be balanced by stats, not even big ass stats. It should be balanced around how the game is played out. Nothing else.
I strongly disagree. Stats are quantitative, "played out" is qualitative. One of these do not have viewer bias. While looking at how things play out is important, it is extremely unscientific to ignore the stats.
Anybody with braincells will disagree with you playing and watching PvZ. A match up entirely balanced by stats is awful for the game. If the "bias" is about being fun, i don't see anything wrong with it since it's a game, not the final balance of a company.
I never said entirely balanced by stats. I said stats must take a significant factor. Besides, whether a match up is boring or not is not remotely the same as balance.
I'm a big fan of stats too
I would like to further add to what you said to satisfy the previous poster that you could also just make more specific stats that would cover how the game is played out better.
A simple example would be to calculate winrates for games at the 10-20 min mark and then one for 20 and beyond.
You could also make stats for when a particular united is fielded, how many bases a player has relative to the other, openings and so on and so. Do toss mostly win with timings or by dragging the game out...We all have an idea about that but having the actual stats would be so helpful
It's just most of the time that this data is not readily available and would require alot of work to obtain, but I really do feel that they could aid us alot in the judgement of the true balance issue!
It's so much easier to support your claims with stats. In the field of research, while qualitative studies are acknowledge as a scientific method, it is still at the bottom of the food chain when it comes to making actual claims about how things truly work.
I'm pretty sure the terran skill cap is way higher than the zergs, but even with a near-perfect terran play, the zerg still wins because the race is simply better and easier. That's bad. Leenock does perform consistently at foreign tournaments, he's frequently getting 1st/finals/etc. but is it because of his skill or just balance? It's very hard to tell, but I can safely tell you infestors aren't hard to use.
And the problems with interpreting the statistics is that, obviously it's a small sample. Another thing is, in SC2, it's much more random than any other game. Consistency is barely even there. In WC3, you KNEW who the top players were, and they ALWAYS got to the finals. That was SORT of the case back when MVP and Nestea were winning GSL/other big events, but even then, you'd see them lose to people all the time in other events. SC2 contains a lot more elements of randomness. Obviously, scouting and corner-cutting plays a large part of this, but another thing may just be that players do not perform that well at live events. Chess doesn't contain nearly the same amount of randomness as SC2 does, as you can pretty much see what they are doing always and make the best decisions. We all hear about how players play the best at home or when practicing, so doesn't that mean the BEST games are during those times? Do live events even MEAN anything if they do not put out the best performances???? The reason we have "live" events in sports is because there is no other way to do it, but with SC2, there isn't even LAN, so what is even the point??????? Just to imitate real sports? If zerg is more forgiving or easier, maybe it's just that the live event environment causes both players to play less than superb and since zerg doesn't punish mistakes as much (-cough- infestors are too flexible -cough-), then the zergs just end up winning.
tl;dr..... Skill ceiling with zerg is bad, randomness in professional games is ALWAYS bad, and live events can produce poor games.
On December 02 2012 16:05 Cascade wrote: Not saying that sc2 is balance or inbalanced, but this is a completely pointless thread. Like a lot of "statistics" threads in sc2.
1. Include errors snip..
TL:DR error analysis pl0x.
Ok, this post makes my brain bleed. While I don't find the OP particularly interesting or important, the poster calculated the actual win rates for this tournament. He took all the games that were played and just divided two numbers. There is no error.
Do you understand the concept of error in statistics?
You see, we're trying to ascertain whether or not the matchups are balanced, that would be the population parameter. These games are merely a sample of all possible games that can be played, giving us a sample statistic.
We can now use the sample statistic to estimate the population parameter.
So if ZvT in IPL 5 is 70% or something, that is the sample statistic. Now, to see whether the matchup is balanced or not, we have to include error. Let's just say error is 15%. Thus, the actual matchup balance (disregarding bias, incorrect sampling, other variables such as player skill, etc.) is between 55% and 85%. That is what error means.
On December 02 2012 16:05 Cascade wrote: Not saying that sc2 is balance or inbalanced, but this is a completely pointless thread. Like a lot of "statistics" threads in sc2.
1. Include errors snip..
TL:DR error analysis pl0x.
Ok, this post makes my brain bleed. While I don't find the OP particularly interesting or important, the poster calculated the actual win rates for this tournament. He took all the games that were played and just divided two numbers. There is no error.
Do you understand the concept of error in statistics?
You see, we're trying to ascertain whether or not the matchups are balanced, that would be the population parameter. These games are merely a sample of all possible games that can be played, giving us a sample statistic.
We can now use the sample statistic to estimate the population parameter.
So if ZvT in IPL 5 is 70% or something, that is the sample statistic. Now, to see whether the matchup is balanced or not, we have to include error. Let's just say error is 15%. Thus, the actual matchup balance (disregarding bias, incorrect sampling, other variables such as player skill, etc.) is between 55% and 85%. That is what error means.
Why would you need error though, I'm kinda confused on that. Isn't error usually used when you want to extrapolate a sample statistic onto a larger unknown population?
On December 02 2012 16:05 Cascade wrote: Not saying that sc2 is balance or inbalanced, but this is a completely pointless thread. Like a lot of "statistics" threads in sc2.
1. Include errors snip..
TL:DR error analysis pl0x.
Ok, this post makes my brain bleed. While I don't find the OP particularly interesting or important, the poster calculated the actual win rates for this tournament. He took all the games that were played and just divided two numbers. There is no error.
Do you understand the concept of error in statistics?
You see, we're trying to ascertain whether or not the matchups are balanced, that would be the population parameter. These games are merely a sample of all possible games that can be played, giving us a sample statistic.
We can now use the sample statistic to estimate the population parameter.
So if ZvT in IPL 5 is 70% or something, that is the sample statistic. Now, to see whether the matchup is balanced or not, we have to include error. Let's just say error is 15%. Thus, the actual matchup balance (disregarding bias, incorrect sampling, other variables such as player skill, etc.) is between 55% and 85%. That is what error means.
Why would you need error though, I'm kinda confused on that. Isn't error usually used when you want to extrapolate a sample statistic onto a larger unknown population?
Yes, exactly! I don't think anyone really cares about the IPL 5 statistics themselves, all they show is that Zerg did well at this particular tournament.
The larger unknown population we're interested in is the population of all possible games, the matchup itself... so we extrapolate IPL 5 TvZ statistics into the parameter of the TvZ matchup as a whole.
So, if we say something like Zergs win 70% of their games versus Terran, just look at IPL 5! That would be flat-out wrong because we didn't include the error... Now, with error, we can say something along the lines of "we are 95% confident that Zergs win between 55%-85% of their games against Terran based upon the data from IPL 5" (yes I'm making the numbers up, but we can calculate the actual numbers, I'm just too lazy to do so right now)
On December 03 2012 05:08 Jebediah wrote: The stats might not be spot on, because Bomber vs Polt distracted me while counting :D I looked through the Nov premier tournaments on Liquipedia and took the map wins, not the match wins. Mirrors weren't counted.
On December 03 2012 06:15 Zooper31 wrote: Am I not the only person who came into this thread thinking race had to do with actual races, Korean, American, Swedish, etc? lol
No, since if we talk about player's actual race, it's going to be like 95% Korean, 5% foreigner or something.
Aside from being interesting using data like this from a single tournament is pointless. Sample size way to small.
Sample size? He used the population. There can not physically be a bigger "sample".
As you say, the sample used is the entire population of games at IPL, so we can, based on the sample, conclusively say that Zerg has done better than Terran on average. However, when people say the sample size is too small, they mean it's too small to draw conclusions regarding the overall balance (or perhaps just the overall pro-level balance) of Starcraft 2. In this case, the tournament is only a small sample of all the games played at a professional level, and it's probably too small to draw any strong conclusions from. Combining the results from the IPL5 with those from other more recent tournaments does provide fairly strong evidence that Zerg is too strong at the moment, but then again, not many people are denying that anymore.
Regarding overall winrates, it's expected to be 33%/33%/33% if there wre equal numbers of all races. However, since there weren't an equal number of players from each race, this is not the case. A more useful statistic would be to look at how significantly each race has over or under-performed its expected rate.
I was going to be a little more blunt with the guy but thanks to you looks like I don't have to.
On December 02 2012 10:39 achan1058 wrote:
On December 02 2012 10:36 zEnVy wrote:
On December 02 2012 10:27 dcemuser wrote:
On December 02 2012 10:05 achan1058 wrote: Seriously, if this is such a problem, there's no need to wait for blizzard. The tournament organizers can throw back in Steppes and be done.
Wait, solving balance through maps? That sounds like blasphemy to me.
So, if Terran plays on Steppes they win, and if they don't play on steppes they lose?
How is that balance?
That's how BW was balanced.
Noooooooooooooooooooooooo.
lol
-_-
Look, there have been plenty of imbalanced maps in the past BW tournaments but we don't go about our business in this way. In fact, there are plenty of maps which have very short-lived runs because they are found out to be too imbalanced. Unfortunately you still have some tournaments like WCG who never get the memo who use old maps and don't change shit.
Gorky Island anyone?
It's not as simple as that.
Blizzard are terrible at designing maps and they'll say, "Oh we're just doing that because we want people to try different strategies." The reality is... they never really had a great map design team and in many cases that's because they just don't frigging understand the game yet till the pro's start doing their thing.
"Oh, so that's how the game will play out."
They will either leave it or patch it. They rarely look at the maps and it's up to our community to play Mr. Fix It because they're too busy elsewhere.
I did not suggest that all BW maps are balance, but rather that you can skew balance one way or another by putting in certain maps, and that's what I am suggesting SC2 tournaments do as well. Throw away all the bigger maps, and put in some that's really small, small enough to guarantee a Terran favorite. If Terrans are too good after that, switch a few maps back. Heck, even if we just throw away every single map other than Antiga, the game's balance would change already. Aside from balance, there are good reason to throw away some maps, namely that the game is getting stale because of the lack of map changes already. (of course, this only works for tournaments, and will suck for anyone on ladder)
1. I knew what you were saying.
2. We do not skew balance in that way by putting in certain maps. It's a silly suggestion and that's why I was rolling my eyes at you the first time.
3. We need better maps and we need tournaments to update their map pools regularly. That has nothing to do with skewing balance. We need maps that all players would be happy to play on.
On December 02 2012 16:05 Cascade wrote: Not saying that sc2 is balance or inbalanced, but this is a completely pointless thread. Like a lot of "statistics" threads in sc2.
1. Include errors snip..
TL:DR error analysis pl0x.
Ok, this post makes my brain bleed. While I don't find the OP particularly interesting or important, the poster calculated the actual win rates for this tournament. He took all the games that were played and just divided two numbers. There is no error.
Do you understand the concept of error in statistics?
You see, we're trying to ascertain whether or not the matchups are balanced, that would be the population parameter. These games are merely a sample of all possible games that can be played, giving us a sample statistic.
We can now use the sample statistic to estimate the population parameter.
So if ZvT in IPL 5 is 70% or something, that is the sample statistic. Now, to see whether the matchup is balanced or not, we have to include error. Let's just say error is 15%. Thus, the actual matchup balance (disregarding bias, incorrect sampling, other variables such as player skill, etc.) is between 55% and 85%. That is what error means.
Why would you need error though, I'm kinda confused on that. Isn't error usually used when you want to extrapolate a sample statistic onto a larger unknown population?
Yes, exactly! I don't think anyone really cares about the IPL 5 statistics themselves, all they show is that Zerg did well at this particular tournament.
The larger unknown population we're interested in is the population of all possible games, the matchup itself... so we extrapolate IPL 5 TvZ statistics into the parameter of the TvZ matchup as a whole.
So, if we say something like Zergs win 70% of their games versus Terran, just look at IPL 5! That would be flat-out wrong because we didn't include the error... Now, with error, we can say something along the lines of "we are 95% confident that Zergs win between 55%-85% of their games against Terran based upon the data from IPL 5" (yes I'm making the numbers up, but we can calculate the actual numbers, I'm just too lazy to do so right now)
Yes, exactly.
While the OP is not wrong in the litteral sense, it is increadible misleading for the majority of the readers, which is just as bad. All the imba-shouters that does not have both understanding in statistical errors and a critical mindset (which I'd estimate to at least 99% of them) will take these numbers as an argument that zerg is overpowered. As we have seen in the replies.
The truth is that the sample is too small to say much about balance on itself, which would be clear if the OP had included a little bit of error analysis. OP asks for no balance whine, while at the same time providing misleading (and overly sensational) information.
The sample size isn't too small. There is no 'error'. These statistics aren't attempting to conclusively prove game balance issues, they're simply the win ratios for each race from a major tournament. They're raw statistics, and these statistics are simply a data point for analysis. Now that we have this data point, we can weight it place it with other data points and begin to paint a picture of game balance. Not coincidentally, every data point (this one included) points to the same conclusion, but that's well-understood by now.
As a Zerg, I don't mind if they nerf infestors. I'm a ling/baneling/mutalisk player anyways xD. I only get infestors/broodlord if it guys to mega late game.
As imbalanced as z vs t probably is, terran players really have no one to blame but themselves. All of that whining over how "unfair" t vs p was, all the while having ridic stats in their favor. Guess what? The stats are still just as skewed as ever. How can anyone take a foreign terran player seriously? It's like the boy that cried wolf, whose main problem probably lies in the fact that they are so much worse than the Koreans that play their race. Should probably spend more time analyzing that imbalance.
If there's anything to be learned, it's probably to enjoy having a 55-56 win percentage instead of acting like a poor, helpless victim. You know your race had it good when it's even possible to moan over such a thing. Toss players are guilty, too. Everyone singles in on one or two things that seem to be glaring imbalances; meanwhile, they miss all of the subtle imbalances that add up, ending up with players that aren't even whining about the right mu that is most "unfair." It's a circus.
The OP is interesting from an "IPL this particular weekend" perspective, but has not very much to do with overall balance of the game. The sample size is too small to use for generalization. One should not make this a bigger deal than it is.
EPT
thanks! ***
