|
On February 03 2012 01:17 wo1fwood wrote:Show nested quote +On February 03 2012 01:07 fritfrat wrote: People can't read graphs at all, this is embarassing. The Korean PvZ presents like it's not even statistically different than the international PvZ! Just because the korean graph goes up to 80 instead of 60 and makes it look more even everyone is reaching big conclusions how it's better in korea than internationally. There's a %7.4 difference in PvZ win rates between Korea and International, that's not different? Edit: oh, I didn't look at the sample size, interesting. From what I see, there is a 3.7% difference not 7.4%. Unless I'm missing something.
|
Holy shit that ZvP sticks out like a sore thumb. WTH happened there lol.
|
I'd like to point out that if you assume equal distribution between the matchups for the Korean winrates, you come out with under 150 games per matchup. Not even close to being statistically useful. It's worth noting that the international data includes the Korean data, though.
|
Interesting:
International: All of the terran QQ vs protoss is unwarranted PvT is balanced. TvZ is balanced Muta QQ is completely warranted. The protoss were right all along.
Korea: PvT is balanced. TvZ is balanced ZvP... lololol.. zerg QQ.. ya... the protosses were right
|
On February 03 2012 01:48 Gevna wrote:Show nested quote +On February 03 2012 01:17 wo1fwood wrote:On February 03 2012 01:07 fritfrat wrote: People can't read graphs at all, this is embarassing. The Korean PvZ presents like it's not even statistically different than the international PvZ! Just because the korean graph goes up to 80 instead of 60 and makes it look more even everyone is reaching big conclusions how it's better in korea than internationally. There's a %7.4 difference in PvZ win rates between Korea and International, that's not different? Edit: oh, I didn't look at the sample size, interesting. From what I see, there is a 3.7% difference not 7.4%. Unless I'm missing something.
Z is winning 3.7% more, P is losing 3.7% more. So there's a 7.4% difference.
It's like if you have a 50/50 match-up and give 10% more to 1 side it becomes 60/40.
|
Wow zenio was right when he said that foreign tosses are nowhere near the level of korean toss. Instead of saying "op op" maybe it would be better to say "hmm what am I doing wrong?".
|
Going after statistics is retarded nuff said. Don't use them as a point for balance arguing. You have no idea if the protoss in those games went nexus first or held a 6pool or went for the same 6gate retarded push they've been doing for 6 months now or if they just can't play properly.
|
I heard zergs were having a though time in ZvP, haha
|
On February 02 2012 18:06 thezanursic wrote: Fuck yeah this fucking proves two things zerg is the easiest race to play and the game is pretty balance.
Yeah proof.... Are you retarded? One month of results shows nothing about either of those things, if you can't see that then you do not understand how this game works.
|
haha nice tears from the zergies, even tho they win 60% of their ZvPs. but keep crying, do it the idra way!
TvP and TvZ got very nice numbers! now change ZvP without affecting the other zergie matchups, and david kim has done an amazing job!
|
What do you guys think is more of a problem in ZvP?
Mutas or Brood lord / Infestor?
how bad do you think it would be if they got rid of the stupidest ability in the game: archon toilet?
|
On February 03 2012 01:54 Roxy wrote: Interesting:
International: All of the terran QQ vs protoss is unwarranted PvT is balanced. TvZ is balanced Muta QQ is completely warranted. The protoss were right all along.
Korea: PvT is balanced. TvZ is balanced ZvP... lololol.. zerg QQ.. ya... the protosses were right
I think if you only look at tvp macro games, you will see that it's very p favoured. Atm the graph is more close because of terran allins
|
When blizzard said they are okay with a 5% disparity, does that mean they want:
47.5% vs 52.5% winrate? or 45% vs 55% winrate?
|
I think it's also worth noting that the trend for the PvZs, it's getting evened out in Korea while getting wider for the International. Ignoring the trend line which is based on last 3 months, it seems like the International Protosses really did badly over the course of January.
|
On February 03 2012 02:11 Snowbear wrote:Show nested quote +On February 03 2012 01:54 Roxy wrote: Interesting:
International: All of the terran QQ vs protoss is unwarranted PvT is balanced. TvZ is balanced Muta QQ is completely warranted. The protoss were right all along.
Korea: PvT is balanced. TvZ is balanced ZvP... lololol.. zerg QQ.. ya... the protosses were right
I think if you only look at tvp macro games, you will see that it's very p favoured. Atm the graph is more close because of terran allins 
if they could give us an interactive chart that lets us know winrates of different games at different times.. even separated by expansion timings.. i would mess myself
i dont know that i have experienced such dominance in the late game as you have, but i would surely be willing to trade some late game strength in exchange for some early game strength. (or vice versa for terran)
|
On February 03 2012 01:54 Roxy wrote: Interesting:
International: All of the terran QQ vs protoss is unwarranted PvT is balanced. TvZ is balanced Muta QQ is completely warranted. The protoss were right all along.
Korea: PvT is balanced. TvZ is balanced ZvP... lololol.. zerg QQ.. ya... the protosses were right
what? so you can say that "the problem" is the mutalisk, by looking at overall winrates? The facts are: Protoss winrate in PvZ has been going down since forever, having it's worst point around the time zergs started to open 3hatch. In October Zergs started using mutalisks and the PvZ winrates went up since that month until this month.
|
On February 03 2012 02:11 Roxy wrote: When blizzard said they are okay with a 5% disparity, does that mean they want:
47.5% vs 52.5% winrate? or 45% vs 55% winrate?
Prolly the first one
|
On February 03 2012 02:18 Jakkerr wrote:Show nested quote +On February 03 2012 02:11 Roxy wrote: When blizzard said they are okay with a 5% disparity, does that mean they want:
47.5% vs 52.5% winrate? or 45% vs 55% winrate? Prolly the first one
I'd hope the first one.
|
On February 03 2012 02:11 Roxy wrote: When blizzard said they are okay with a 5% disparity, does that mean they want:
47.5% vs 52.5% winrate? or 45% vs 55% winrate?
47.5 to 52.5
|
On February 03 2012 01:54 Skwid1g wrote:Show nested quote +On February 03 2012 01:48 Gevna wrote:On February 03 2012 01:17 wo1fwood wrote:On February 03 2012 01:07 fritfrat wrote: People can't read graphs at all, this is embarassing. The Korean PvZ presents like it's not even statistically different than the international PvZ! Just because the korean graph goes up to 80 instead of 60 and makes it look more even everyone is reaching big conclusions how it's better in korea than internationally. There's a %7.4 difference in PvZ win rates between Korea and International, that's not different? Edit: oh, I didn't look at the sample size, interesting. From what I see, there is a 3.7% difference not 7.4%. Unless I'm missing something. Z is winning 3.7% more, P is losing 3.7% more. So there's a 7.4% difference. It's like if you have a 50/50 match-up and give 10% more to 1 side it becomes 60/40. This is true. However, these values should ALWAYS been viewed in reference to the point of ideal, which is 50%. To say that Zerg wins 7.4% more than Protoss is entirely true, but it makes the imbalance appear twice as large as it actually is, because both are only 3.7% from the ideal.
I'm not saying this to you in particular, but merely to remind *everyone* to consider these data in reference to the hypothetical ideal, rather than directly to each other. This data would be better if it presented match-up performances as a single bar representing deviation from 50% instead of representing match ups as two bars, one for each race. The latter makes differences appear larger than they actually are.
On February 03 2012 02:18 Jakkerr wrote:Show nested quote +On February 03 2012 02:11 Roxy wrote: When blizzard said they are okay with a 5% disparity, does that mean they want:
47.5% vs 52.5% winrate? or 45% vs 55% winrate? Prolly the first one
There sure are a lot of idealists. I am fairly confident that Blizzard would be thrilled if all match-ups were within 45/55 across all regions and would consider their game balanced. Unless, of course, the same race was 45% in both it's non mirror match ups.
EDIT: also, WTF with the differences in scale between the Korean and International graphs? That is shady statistical deception bullshit like none other. Fix that shit immediately.
|
|
|
|