|
On March 16 2012 20:01 neoghaleon55 wrote:Show nested quote +On March 16 2012 19:52 Fubi wrote:
You STILL have to prove, with math/stats, that this difference isn't simply due to random chance, because no matter how many samples you take, there will STILL be a chance that the difference is purely due to randomness.
You can NOT, simply say "I FEEL like x% is large enough to show that there is a difference".
I'm not trying to be a dick or anything, but you're making yourself look REALLY bad to the people that actually understands your stats. Thank god they don't let people work in any serious jobs simply from one year of college education.
You're going to have to try harder than that. No, I don't need to prove about "difference due to random chance" Are there more things contributing to a win/loss rather than balance and skills? absolutely! There's also luck, but it's not easy to quantify luck. And if you're so caught up with confidence, why don't you do it?
Exactly, luck (or chance) is involved. And you CAN quantify luck, because luck = chance = variance, therefore, that is why standard deviation, confidence interval, margin of errors, etc, exists in statistics.
Here, I'll use YOUR graphs to help you understand
look at your own graphs that you just made: even at n = 32, there is still a fairly large chance that you can get 14/32 (=~44%) heads or 18/32 (=~56%) heads (+/- 2 from normal), simply due to random chance.
So, say we flip 32 times, and I get 12 heads, and you get 20 heads, that is a ~12% difference. BUT, you can NOT prove whether or not this ~12% difference is due to our skills at flipping, or simply due to chance. You have to do variance analysis with a confidence interval to prove the significance of this statistical test
*edit: and btw, you made that statistical analysis, and based on that, you made a statistical claim, therefore, the onus of proof is on you, not on me, hence why you need to be the one to do it if you want to prove that your stats has any statistical significance, not me.
|
On March 16 2012 20:04 Fubi wrote:Show nested quote +On March 16 2012 20:01 neoghaleon55 wrote:On March 16 2012 19:52 Fubi wrote:
You STILL have to prove, with math/stats, that this difference isn't simply due to random chance, because no matter how many samples you take, there will STILL be a chance that the difference is purely due to randomness.
You can NOT, simply say "I FEEL like x% is large enough to show that there is a difference".
I'm not trying to be a dick or anything, but you're making yourself look REALLY bad to the people that actually understands your stats. Thank god they don't let people work in any serious jobs simply from one year of college education.
I'm not trying to be a dick or anything You're going to have to try harder than that. No, I don't need to prove about "difference due to random chance" Are there more things contributing to a win/loss rather than balance and skills? absolutely! There's also luck, but it's not easy to quantify luck. And if you're so caught up with confidence, why don't you do it? Exactly, luck (or chance) is involved. And you CAN quantify luck, because luck = chance = variance, therefore, that is why standard deviation, confidence interval, margin of errors, etc, exists in statistics. Here, I'll use YOUR graphs to help you understand look at your own graphs that you just made: even at n = 32, there is still a fairly large chance that you can get 14/32 (~44%) heads or 18/32 (~56%) heads (+/- 2 from normal). So, say we flip 32 times, and I get 12 heads, and you get 20 heads, that is a ~12% difference. BUT, you can NOT prove whether or not this ~12% difference is due to our skills at flipping, or simply due to chance. You have to do variance analysis with a confidence interval to prove the significance of this statistical test *edit: and btw, you made that statistical analysis, and based on that, you made a statistical claim, therefore, the onus of proof is on you, not on me, hence why you need to be the one to do it if you want to prove that your stats has any statistical significance, not me.
Hey Fubi.
Thank you for your input.
|
On March 16 2012 20:06 neoghaleon55 wrote:Show nested quote +On March 16 2012 20:04 Fubi wrote:On March 16 2012 20:01 neoghaleon55 wrote:On March 16 2012 19:52 Fubi wrote:
You STILL have to prove, with math/stats, that this difference isn't simply due to random chance, because no matter how many samples you take, there will STILL be a chance that the difference is purely due to randomness.
You can NOT, simply say "I FEEL like x% is large enough to show that there is a difference".
I'm not trying to be a dick or anything, but you're making yourself look REALLY bad to the people that actually understands your stats. Thank god they don't let people work in any serious jobs simply from one year of college education.
I'm not trying to be a dick or anything You're going to have to try harder than that. No, I don't need to prove about "difference due to random chance" Are there more things contributing to a win/loss rather than balance and skills? absolutely! There's also luck, but it's not easy to quantify luck. And if you're so caught up with confidence, why don't you do it? Exactly, luck (or chance) is involved. And you CAN quantify luck, because luck = chance = variance, therefore, that is why standard deviation, confidence interval, margin of errors, etc, exists in statistics. Here, I'll use YOUR graphs to help you understand look at your own graphs that you just made: even at n = 32, there is still a fairly large chance that you can get 14/32 (~44%) heads or 18/32 (~56%) heads (+/- 2 from normal). So, say we flip 32 times, and I get 12 heads, and you get 20 heads, that is a ~12% difference. BUT, you can NOT prove whether or not this ~12% difference is due to our skills at flipping, or simply due to chance. You have to do variance analysis with a confidence interval to prove the significance of this statistical test *edit: and btw, you made that statistical analysis, and based on that, you made a statistical claim, therefore, the onus of proof is on you, not on me, hence why you need to be the one to do it if you want to prove that your stats has any statistical significance, not me. Hey Fubi. Thank you for your input.
Thanks, glad you understood your errors
|
First we have a terran whine thread based on TvP and terran ladder participation.
Now we have a zerg whine thread about ZvT stats at the highest levels.
It's pretty obvious that both OP's of these threads have a objective to balance whine by veiling it in misinterpreted stats.
For the OP:
What if terran just has a couple more really high end players compared to zerg?
What's the win rates the last 6 months or so for zerg? Are they better or worse? (I'd guess they're better)
There's been huge balance patches and map changes that really affected ZvT, doesn't those make the win/loss statistics pretty uninteresting and skewed for us now in march 2012? (marine+scv all in has provided MKP with at least 5+ wins that would never happen today, close position has also given terrans a couple wins each which is also moot now)
How would the fact that the terran race is the most popular amongst koreans affect the talent pool and finally the top player distribution?
There's just so much more than just posting 6 terran and 6 zerg players win/loss and deducting all sorts of stuff from it.
|
This is the kind of shoddy statistical analysis that allows pseudoscience like parapsychology thrive.
And Fubi is the James Randi of esports
|
On March 16 2012 20:09 karpo wrote:
First we have a terran whine thread based on TvP and terran ladder participation.
Now we have a zerg whine thread about ZvT stats at the highest levels.
It's pretty obvious that both OP's of these threads have a objective to balance whine by veiling it in misinterpreted stats.
What if terran just has a couple more really high end players compared to zerg?
There's a confusion here
The top players are not all Terrans.
BUT the top terrans are all good against Zerg.
There is in fact no Terrans in the top 50 ELO that have a lower than 50% winrate in TvZ.
What's the win rates the last 6 months or so for zerg? Are they better or worse?
According to these charts
http://imgur.com/a/1aAfu
It has not improved...in fact it's getting worse.
There's been huge balance patches and map changes that really affected ZvT, doesn't those make the win/loss statistics pretty uninteresting or us now in march 2012?
We don't know. There's no data. More time is needed.
How would the fact that the terran race is the most popular amongst koreans affect the talent pool and finally the top player distribution?
This point of yours is hard to argue.
Did the top players pick Terran because Terran is the best race?
Or is Terran the best race because the top players pick it.
|
Damn, these sample sizes are so big, everything you say must have some merit.
|
On March 16 2012 20:07 Fubi wrote:Show nested quote +On March 16 2012 20:06 neoghaleon55 wrote:On March 16 2012 20:04 Fubi wrote:On March 16 2012 20:01 neoghaleon55 wrote:On March 16 2012 19:52 Fubi wrote:
You STILL have to prove, with math/stats, that this difference isn't simply due to random chance, because no matter how many samples you take, there will STILL be a chance that the difference is purely due to randomness.
You can NOT, simply say "I FEEL like x% is large enough to show that there is a difference".
I'm not trying to be a dick or anything, but you're making yourself look REALLY bad to the people that actually understands your stats. Thank god they don't let people work in any serious jobs simply from one year of college education.
I'm not trying to be a dick or anything You're going to have to try harder than that. No, I don't need to prove about "difference due to random chance" Are there more things contributing to a win/loss rather than balance and skills? absolutely! There's also luck, but it's not easy to quantify luck. And if you're so caught up with confidence, why don't you do it? Exactly, luck (or chance) is involved. And you CAN quantify luck, because luck = chance = variance, therefore, that is why standard deviation, confidence interval, margin of errors, etc, exists in statistics. Here, I'll use YOUR graphs to help you understand look at your own graphs that you just made: even at n = 32, there is still a fairly large chance that you can get 14/32 (~44%) heads or 18/32 (~56%) heads (+/- 2 from normal). So, say we flip 32 times, and I get 12 heads, and you get 20 heads, that is a ~12% difference. BUT, you can NOT prove whether or not this ~12% difference is due to our skills at flipping, or simply due to chance. You have to do variance analysis with a confidence interval to prove the significance of this statistical test *edit: and btw, you made that statistical analysis, and based on that, you made a statistical claim, therefore, the onus of proof is on you, not on me, hence why you need to be the one to do it if you want to prove that your stats has any statistical significance, not me. Hey Fubi. Thank you for your input. Thanks, glad you understood your errors
I appreciate your understanding of statistics. I wonder though, if it annoys you when people INSIST that you need a sample size of 1000s of games for any analysis to be relevant when that isn't how stats works at all. There is a lot more than just looking at the averages of huge sample sizes.
|
On March 16 2012 20:19 Sideburn wrote:Show nested quote +On March 16 2012 20:07 Fubi wrote:On March 16 2012 20:06 neoghaleon55 wrote:On March 16 2012 20:04 Fubi wrote:On March 16 2012 20:01 neoghaleon55 wrote:On March 16 2012 19:52 Fubi wrote:
You STILL have to prove, with math/stats, that this difference isn't simply due to random chance, because no matter how many samples you take, there will STILL be a chance that the difference is purely due to randomness.
You can NOT, simply say "I FEEL like x% is large enough to show that there is a difference".
I'm not trying to be a dick or anything, but you're making yourself look REALLY bad to the people that actually understands your stats. Thank god they don't let people work in any serious jobs simply from one year of college education.
I'm not trying to be a dick or anything You're going to have to try harder than that. No, I don't need to prove about "difference due to random chance" Are there more things contributing to a win/loss rather than balance and skills? absolutely! There's also luck, but it's not easy to quantify luck. And if you're so caught up with confidence, why don't you do it? Exactly, luck (or chance) is involved. And you CAN quantify luck, because luck = chance = variance, therefore, that is why standard deviation, confidence interval, margin of errors, etc, exists in statistics. Here, I'll use YOUR graphs to help you understand look at your own graphs that you just made: even at n = 32, there is still a fairly large chance that you can get 14/32 (~44%) heads or 18/32 (~56%) heads (+/- 2 from normal). So, say we flip 32 times, and I get 12 heads, and you get 20 heads, that is a ~12% difference. BUT, you can NOT prove whether or not this ~12% difference is due to our skills at flipping, or simply due to chance. You have to do variance analysis with a confidence interval to prove the significance of this statistical test *edit: and btw, you made that statistical analysis, and based on that, you made a statistical claim, therefore, the onus of proof is on you, not on me, hence why you need to be the one to do it if you want to prove that your stats has any statistical significance, not me. Hey Fubi. Thank you for your input. Thanks, glad you understood your errors I appreciate your understanding of statistics. I wonder though, if it annoys you when people INSIST that you need a sample size of 1000s of games for any analysis to be relevant when that isn't how stats works at all. There is a lot more than just looking at the averages of huge sample sizes.
Not saying he needs more games, I'm saying he is missing the other half of what makes statistics what it is. Even if he gets up to sample size of 1000 games (or flips 1000 coins), there is still a chance that the difference in the results are due to pure randomness rather than actual significant results.
The whole point of statistics is to use math to prove that the result you have is not simply due to chance. I'm guessing the OP is in the middle of his first semester in Statistics class, and since the semester isn't over, he hasn't been taught the other half of what makes stats stats.
|
What I wonder is, do you have a conclusion you draw from these stats, or did you only want to show some facts for the statistically interested (like me).
I mean, do you want to argue balance? Which is what most people assume from threads like this.
Do you want to argue skill?
Do you want to argue Zerg weaknesses/Terran strengths? (slash want people to argue about why this winrate disparity exists)
|
On March 16 2012 20:25 Big J wrote:
What I wonder is, do you have a conclusion you draw from these stats, or did you only want to show some facts for the statistically interested (like me).
I mean, do you want to argue balance? Which is what most people assume from threads like this.
Do you want to argue skill?
Do you want to argue Zerg weaknesses/Terran strengths? (slash want people to argue about why this winrate disparity exists)
I'm arguing that nobody has any idea what they're doing in ZvT.
I mean, when the best ZvT player is absolutely Terrible at ZvT compared to his other matchups...you can't help but feel something's wrong.
In contrast, everyone knows how to play TvZ...if you're GSL caliber, your TvZ tends to not fall below 50% winrate.
I have no idea why this is or how to fix it.
|
On March 16 2012 20:27 neoghaleon55 wrote:Show nested quote +On March 16 2012 20:25 Big J wrote:
What I wonder is, do you have a conclusion you draw from these stats, or did you only want to show some facts for the statistically interested (like me).
I mean, do you want to argue balance? Which is what most people assume from threads like this.
Do you want to argue skill?
Do you want to argue Zerg weaknesses/Terran strengths? (slash want people to argue about why this winrate disparity exists) I'm arguing that nobody has any idea what they're doing in ZvT. I mean, when the best ZvT player is absolutely Terrible at ZvT compared to his other matchups...you can't help but feel something's wrong. In contrast, everyone knows how to play TvZ...if you're GSL caliber, your TvZ tends to not fall below 50% winrate.
I'm sorry but 64% win rate is considered REALLY REALLY good in Starcraft. Have you thought of the possibility that it would make more sense that it's the OTHER 2 matchups that are unbalanced while possibly, TvZ is the more balanced one?
|
On March 16 2012 20:18 neoghaleon55 wrote:Show nested quote +On March 16 2012 20:09 karpo wrote:
First we have a terran whine thread based on TvP and terran ladder participation.
Now we have a zerg whine thread about ZvT stats at the highest levels.
It's pretty obvious that both OP's of these threads have a objective to balance whine by veiling it in misinterpreted stats. What if terran just has a couple more really high end players compared to zerg?There's a confusion here The top players are not all Terrans. BUT the top terrans are all good against Zerg. There is in fact no Terrans in the top 50 ELO that have a lower than 50% winrate in TvZ. What's the win rates the last 6 months or so for zerg? Are they better or worse?According to these charts http://imgur.com/a/1aAfuIt has not improved...in fact it's getting worse. There's been huge balance patches and map changes that really affected ZvT, doesn't those make the win/loss statistics pretty uninteresting or us now in march 2012?We don't know. There's no data. More time is needed. How would the fact that the terran race is the most popular amongst koreans affect the talent pool and finally the top player distribution? This point of yours is hard to argue. Did the top players pick Terran because Terran is the best race? Or is Terran the best race because the top players pick it.
Shouldn't everyone top 50 elo have a 50% + winrate?... and metagame shifts so the winrates change and people generally choose a race based on what they like and theres no way to tell what is the best race and no one plays perfect so its only difference in skill level and what thing might be easier to execute than to defend and shit like that balance and imba discussions are fucking pointless there are only things that you need to overcome thats the way they did it in broodwar now theres all this imbalance whining about fucking everything
|
On March 16 2012 20:30 Fubi wrote:Show nested quote +On March 16 2012 20:27 neoghaleon55 wrote:On March 16 2012 20:25 Big J wrote:
What I wonder is, do you have a conclusion you draw from these stats, or did you only want to show some facts for the statistically interested (like me).
I mean, do you want to argue balance? Which is what most people assume from threads like this.
Do you want to argue skill?
Do you want to argue Zerg weaknesses/Terran strengths? (slash want people to argue about why this winrate disparity exists) I'm arguing that nobody has any idea what they're doing in ZvT. I mean, when the best ZvT player is absolutely Terrible at ZvT compared to his other matchups...you can't help but feel something's wrong. In contrast, everyone knows how to play TvZ...if you're GSL caliber, your TvZ tends to not fall below 50% winrate. I'm sorry but 64% win rate is considered REALLY REALLY good in Starcraft. Have you thought of the possibility that it would make more sense that it's the OTHER 2 matchups that are unbalanced while possibly, TvZ is the more balanced one?
Not when you take into account that any Terran worth his salt is an absolute beast at TvZ. Top 50 ELO Terrans have winrates greater than 50% where as Zerg have people like Lucky, BBBB and Zenio...who you can argue are terrible at ZvT but they are still top tier. I mean, if you scroll all the way to the bottom of the pile of Terrans...you get people like Yoda and theBest...who still have greater than 50% TvZ!
|
On March 16 2012 20:27 neoghaleon55 wrote:Show nested quote +On March 16 2012 20:25 Big J wrote:
What I wonder is, do you have a conclusion you draw from these stats, or did you only want to show some facts for the statistically interested (like me).
I mean, do you want to argue balance? Which is what most people assume from threads like this.
Do you want to argue skill?
Do you want to argue Zerg weaknesses/Terran strengths? (slash want people to argue about why this winrate disparity exists) I'm arguing that nobody has any idea what they're doing in ZvT. I mean, when the best ZvT player is absolutely Terrible at ZvT compared to his other matchups...you can't help but feel something's wrong. In contrast, everyone knows how to play TvZ...if you're GSL caliber, your TvZ tends to not fall below 50% winrate. I have no idea why this is or how to fix it.
Yet DRG has about the same win percentage against terran as 5 out of 6 top ELO terrans have versus zerg, all but MMA (who has a retardedly high winrate). And the other top ELO zergs have a few percent lower win rates versus terran than the top terrans have versus zerg. Yet this turns into "nobody has any idea what they're doing in ZvT".
DRG has high ZvZ win rate which Nestea also has. DRG is great against P and a slightly worse against T, yet there's close to no terrans that do better against zerg either so where's the huge problem?
It's balance whine veiled in statistics and "analysis".
On March 16 2012 20:36 neoghaleon55 wrote:Show nested quote +On March 16 2012 20:30 Fubi wrote:On March 16 2012 20:27 neoghaleon55 wrote:On March 16 2012 20:25 Big J wrote:
What I wonder is, do you have a conclusion you draw from these stats, or did you only want to show some facts for the statistically interested (like me).
I mean, do you want to argue balance? Which is what most people assume from threads like this.
Do you want to argue skill?
Do you want to argue Zerg weaknesses/Terran strengths? (slash want people to argue about why this winrate disparity exists) I'm arguing that nobody has any idea what they're doing in ZvT. I mean, when the best ZvT player is absolutely Terrible at ZvT compared to his other matchups...you can't help but feel something's wrong. In contrast, everyone knows how to play TvZ...if you're GSL caliber, your TvZ tends to not fall below 50% winrate. I'm sorry but 64% win rate is considered REALLY REALLY good in Starcraft. Have you thought of the possibility that it would make more sense that it's the OTHER 2 matchups that are unbalanced while possibly, TvZ is the more balanced one? Not when you take into account that any Terran worth his salt is an absolute beast at TvZ. Top 50 ELO Terrans have winrates greater than 50% where as Zerg have people like Lucky, BBBB and Zenio...who you can argue are terrible at ZvT but they are still top tier. I mean, if you scroll all the way to the bottom of the pile of Terrans... you get people like Yoda and theBest...who still have greater than 50% TvZ!
And do you know why theBest has a above 50% win rate against zerg? Because he abused cheeses and all-ins that now do next to nothing to good zergs. Strip away the wins terran got from marine+svc all in, close position smaller maps, blue flame hellions before nerf, ghost before nerf, and other outdated stuff you've probably got completely different statistics.
|
On March 16 2012 11:14 neoghaleon55 wrote:
Recently, during the GSTL, Khaldor said something that struck me as pretty damn odd.
He said "DongRaeGu's weakest matchup is against Terran."
I thought, "wait a minute....isn't DRG suppose to be some Terran killer god?! Why is his weakest matchup ZvT?" It turns out Khaldor was right...
DongRaeGu
All: 67-30 (69.07%)
vT: 32-18 64.00%
vZ: 8-2 80.00%
vP: 27-10 72.97%
64% looks pretty terrible against the backdrop of 80% and 72%. So I shrugged it off...but it's been eating at me for over a week...I decided to investigate more...
The question was: Who has the best ZvT%? Is it possible to have a good ZvT percentage compared to TvZ?
Here are some findings
Top 6 Zergs by ELO using TLPD. I used the top 6 cutoff because nobody was really worth mentioning after Curious and Symbol...and then you get down to the top 20-ish on ELO...which by itself is no longer top tier play.
DRG vT: 32-18 64.00%
Nestea vT: 52-39 57.14%
Leenock vT: 62-40 60.78%
July vT: 29-19 60.42%
Curious vT: 25-19 56.82%
Symbol vT: 13-8 61.90%
Clearly DRG's ZvT outclasses everyone by at least 3-5%, DongRaeGu is indeed the best ZvT player in the world....statistic wise.
So...what about TvZ
Top 6 Terrans by ELO using TLPD
MMA vZ: 21-5 80.77%
MVP vZ: 43-21 67.19%
aLive vZ: 28-14 66.67%
MKP vZ: 39-21 65.00%
Taeja vZ: 58-29 66.67%
Gumiho vZ: 30-15 66.67%
....the discrepency is HUUUUUUGEE!!
Top Korean Terrans really can play well against Zerg...by a margin of greater than 5% win/loss ratio. How can this be? I am not sure it's a balance issue or design issue or just the meta game, but ZvT does not compare well to TvZ statistically in korea.
I come to the conclusion that....unless you're DRG, nobody's really any good at ZvT.
But then the stats themselves clearly are saying...DRG's not actually good at ZvT either...he's just good in general and it bleeds over to his ZvT.
It's pretty simple. The individual terran players you listed carry the win rates of the race. There are simply more top tier terrans than there are zergs, so they get better winrates overall at the very top.
|
On March 16 2012 20:35 KAmaKAsa wrote:
Shouldn't everyone top 50 elo have a 50% + winrate?...
That's not the case
Go talk to Zenio/Lucky and BBBB why their ZvT statistics is completely abysmal.
|
On March 16 2012 20:36 neoghaleon55 wrote:Show nested quote +On March 16 2012 20:30 Fubi wrote:On March 16 2012 20:27 neoghaleon55 wrote:On March 16 2012 20:25 Big J wrote:
What I wonder is, do you have a conclusion you draw from these stats, or did you only want to show some facts for the statistically interested (like me).
I mean, do you want to argue balance? Which is what most people assume from threads like this.
Do you want to argue skill?
Do you want to argue Zerg weaknesses/Terran strengths? (slash want people to argue about why this winrate disparity exists) I'm arguing that nobody has any idea what they're doing in ZvT. I mean, when the best ZvT player is absolutely Terrible at ZvT compared to his other matchups...you can't help but feel something's wrong. In contrast, everyone knows how to play TvZ...if you're GSL caliber, your TvZ tends to not fall below 50% winrate. I'm sorry but 64% win rate is considered REALLY REALLY good in Starcraft. Have you thought of the possibility that it would make more sense that it's the OTHER 2 matchups that are unbalanced while possibly, TvZ is the more balanced one? Not when you take into account that any Terran worth his salt is an absolute beast at TvZ. Top 50 ELO Terrans have winrates greater than 50% where as Zerg have people like Lucky, BBBB and Zenio...who you can argue are terrible at ZvT but they are still top tier. I mean, if you scroll all the way to the bottom of the pile of Terrans...you get people like Yoda and theBest...who still have greater than 50% TvZ!
So... you found 3 out of FIFTY zergs that has less than 50% win rate vsT... and now you're making a conclusion based on that? That is like even worst than your opening post from a statistical point of view.
|
On March 16 2012 11:48 neoghaleon55 wrote:Show nested quote +On March 16 2012 11:33 WeaponX.7 wrote:
I find it very unlikely that these stats are all current patch... and therfore pretty useless. In that regards...the entire TLPD is useless and we should just discard it. I disagree.
TLPD is useless in the way you're trying to use it, yes.
There's definitely been a T>Z trend for most of SC2's lifespan, that's however completely irrelevant in terms of arguing the state of TvZ right now.
I would argue that Zergs have vented plenty over those abusive years and that there's no reason to assume the matchup has suddenly taken a turn for the worse. Again, the fact that TvZ was so Terran favored in the early stages of SC2 would explain why Terrans have more dominating percentages in the MU overall, doesn't mean that if we started anew the same trend would continue.
|
|
|
|
|
|
|
|
EPT
