|
On July 11 2012 02:17 Chill wrote:Show nested quote +On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum?
The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players.
|
On July 11 2012 02:17 Chill wrote:Show nested quote +On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum?
As far as i know he only/mainly uses data from diamond, master, grandmaster. These are the only players whose MMR is accurate.
|
I read it I cant say I understood all of it but I have one question.
If the race you are playing is under powered and MMR is created to match you with an opponent which have on an average the same MMR as you, would you continiously be matched up with opponents that technically are worse than you but compensated by the more powerful race?
Put it this way (possibly a bit simplified, but I hope you understand where Im getting at)
MMR = Skill * race power
Winrate = 50% regardless of your MMR.
|
Data:
i have currently 106374 games. When i talk about my users i dont mean myself i mean:
http://www.teamliquid.net/forum/viewmessage.php?topic_id=334561
For this calculation i use diffrence users: i have 60.000 diffrent mmr values of ladder users worldwide.
Most of them i dont know the race because i added the race in a later version. So racedata is about 9000 sc2 accounts.
I get near 5000 games daily with ^ 1000 -2000 valid mmr results for new account.
Matchups:
I dont look on single game results. I look for the skill rating of the hole account and what main race this account playes.
So i can not say anything about specific matchups!
On July 11 2012 02:03 Mendelfist wrote:Show nested quote +On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters?
My data includes bronce to gm. But most of the accounts i track are diamond - gm.
With more data in future i could do this data for each league and server.
On July 11 2012 01:50 1st_Panzer_Div. wrote:
Whoah, rechecked that, you have 149,000 games of data. And you are claiming 4% of that is you as well?
of my users not of myself
And why did you run the random deviation tests than only running 1,000 games, and not at least equal to the 149,000. (You actually should run random monte carlo's for whatever the estimated current userbase is to get some mock battle.net ladders from a perfectly balanced game). I could easily pick 1,000 games out of your current data and show significant imbalance towards any of the three races.
Also what are the dates your data is from?
This is a cool idea... the deviation bit is just not nearly enough random games to be accurate. What program did you use to run these tests, and was it length of the test that prevented you from doing a few hundred thousand?
the program i use is selfwriten. Before running a test i have to analyse the mmr of all accounts. this alone need 1 hour.
Also it does not care how often i run the random values at all. It would be enough to do it 100 time.
. I could easily pick 1,000 games out of your current data and show significant imbalance towards any of the three races.
Yes you could, and if you check the result, you see that the chance that this happens randomly in my case is under 0.0001 %
should run random monte carlo's for whatever the estimated current userbase
..... ahhhhhhh .... monte carlos can be wrong . statistic can be wrong.... I dont see any other point why you mention them.
Also you dont say random monte carlo. You call this algorithem family monte carlo because they all work with random...
On July 11 2012 02:17 Chill wrote:Show nested quote +On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum?
Than ANY balance is dismissed in your eyes. This is not a question about my data this is a question about the definition of balance!
like i said in my post the data is very biased toward skill. The average of the ladder is gold. The average of my data is high diamond!
Also last time i check everyone on ladder cared for balance at his level ^^
On July 11 2012 02:23 Moonalisa wrote:
I read it I cant say I understood all of it but I have one question.
If the race you are playing is under powered and MMR is created to match you with an opponent which have on an average the same MMR as you, would you continiously be matched up with opponents that technically are worse than you but compensated by the more powerful race?
Put it this way (possibly a bit simplified, but I hope you understand where Im getting at)
MMR = Skill * race power
Winrate = 50% regardless of your MMR.
skillrating is your skill. Means if you are underpowered you have less skill. So you play less skill players. If they are from an overpowered race than they are race independent worse then you . Yes.
|
Calgary25976 Posts
On July 11 2012 02:19 OrbitalPlane wrote:Show nested quote +On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? As far as i know he only/mainly uses data from diamond, master, grandmaster. These are the only players whose MMR is accurate.
He said "mainly". But he doesn't say what ratio.
|
On July 11 2012 02:29 Chill wrote:Show nested quote +On July 11 2012 02:19 OrbitalPlane wrote:On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? As far as i know he only/mainly uses data from diamond, master, grandmaster. These are the only players whose MMR is accurate. He said "mainly". But he doesn't say what ratio.
"The average of the ladder is gold. The average of my data is high diamond!"
http://www.teamliquid.net/forum/viewmessage.php?topic_id=351786¤tpage=2#24
|
Calgary25976 Posts
On July 11 2012 02:18 Shiori wrote:Show nested quote +On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players.
We can't just dimiss it. Imagine, for whatever reason, that there is a strong bias for new players to automatically choose Terran. The remaining players try all the races and determine which of the three fit their styles, making them more likely to win. Because you can imagine a situation where Zerg and Protoss average win rates are higher than Terran it must be addressed.
|
Posts like these are what makes TL so exceptional. Very nice OP, most impressive analysis.
|
On July 11 2012 02:11 Huragius wrote:Show nested quote +On July 11 2012 02:08 Stiluz wrote:
It would be interesting if you could divide this by league too. Or... Divide by zero ?!
Let's not collapse the universe until HotS is out, at least.
|
Calgary25976 Posts
On July 11 2012 02:25 skeldark wrote:
Than ANY balance is dismissed in your eyes. This is not a question about my data this is a question about the definition of balance!
like i said in my post the data is very biased toward skill. The average of the ladder is gold. The average of my data is high diamond!
Also last time i check everyone on ladder cared for balance at his level ^^
I guess so. I'm just wondering what effects from these results you want. Common sense says that balance at the highest level can be different from the average balance. If this is just for information then it's good to know, but I think a more focused approach would be more useful.
|
This thread is an embarrassment. I think it's been adequately explained *why* it would be impossible to balance for all levels, but tbh as a low level player I couldn't really care less. Obviously MMR and matchmaking mean that I don't lose more than I win, but I'm not more mad if the game is imbalanced and I lose to a "less skilled" player, I don't see why it has any meaning.
A good analogy would perhaps be martial arts. A guy who's a lot bigger and stronger than me has an obvious advantage in a fight, he is "imba". But I know that if I improve my skills I can still beat him, and thats what I get the satisfaction from, not from ensuring I only fight guys my exact height and weight. Similarly in SC2 just winning more and knowing that I'm getting better is where the satisfaction comes from.
Besides, until high diamond macro is by far the most important reason people suck. Improve your macro and stop whining about balance.
|
On July 11 2012 02:31 Chill wrote:Show nested quote +On July 11 2012 02:18 Shiori wrote:On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players. We can't just dimiss it. Imagine, for whatever reason, that there is a strong bias for new players to automatically choose Terran. The remaining players try all the races and determine which of the three fit their styles, making them more likely to win. Because you can imagine a situation where Zerg and Protoss average win rates are higher than Terran it must be addressed.
That sounds incredibly unlikely in the sense that you're suggesting, though. When Blizzard says that more new players choose Terran, they don't say or even suggest that those people choose their race in a different way than the people who choose other races. It's entirely possible (and likely, I'd say) that most people pick Terran because they're the protagonists and because they're human beings. The people who pick Zerg/Protoss at the noob level don't know enough about their "styles" to make a choice that really affects their ability to win, because it's not actually very clear from the start what the styles of the races even are.
The people who pick P/Z are probably motivated by the same thing that motivates players to pick Terran: they think the race is cool. It just so happens that there are fewer of them because aliens are less appealing than humans.
Besides, even if there were a strong bias for new players to automatically choose Terran, talented RTS players are talented RTS players, and weak RTS players are weak. I don't believe that players have an inbuilt magical bias to one race that influences them so much that they'd be incapable of playing the other races at a high level. If Terran is just more appealing to human beings in general, then it's going to attract all sorts with random distribution, meaning that the average Terran player isn't going to be any better than the average P/Z player because it's just a larger sample but is still evenly distributed.
Until you can show that there's a race which has a baffling number of good players but almost no bad players, the point is moot.
|
On July 11 2012 02:18 Shiori wrote:Show nested quote +On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players.
I don't follow you here. The point is that new players choose terran because of the campaign, and some of them later in their career switch race. The switch is important, because that WILL cause an over-representation of terrans at lower MMR. It has been known for a long time, for example by looking at sc2ranks statistics. There is no easy way to remove these types of biases from any data that we have. We also don't know how far up the leagues this bias persists. Is it only in bronze/silver? I have no idea.
|
On July 11 2012 02:35 Chill wrote:Show nested quote +On July 11 2012 02:25 skeldark wrote:
Than ANY balance is dismissed in your eyes. This is not a question about my data this is a question about the definition of balance!
like i said in my post the data is very biased toward skill. The average of the ladder is gold. The average of my data is high diamond!
Also last time i check everyone on ladder cared for balance at his level ^^ I guess so. I'm just wondering what effects from these results you want. Common sense says that balance at the highest level can be different from the average balance. If this is just for information then it's good to know, but I think a more focused approach would be more useful.
Read my frist statement in the thread. I dont want anything. The last thing i want to do is to talk about how to blance or whine about units.
I got the data, i publish the result. Thats all. What to do with it can everyone decide for themself.
About balance on skilllevels. At the moment i get enough data that i double the data i used here in 1 week.
So in 2-3 weeks i should have enough to do it only for master+
On July 11 2012 02:40 Mendelfist wrote:Show nested quote +On July 11 2012 02:18 Shiori wrote:On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players. I don't follow you here. The point is that new players choose terran because of the campaign, and some of them later in their career switch race. The switch is important, because that WILL cause an over-representation of terrans at lower MMR. It has been known for a long time, for example by looking at sc2ranks statistics. There is no easy way to remove these types of biases from any data that we have. We also don't know how far up the leagues this bias persists. Is it only in bronze/silver? I have no idea.
There is one way to remove this type of biases. Its very complicated but i try to explain it.
you have different amount of group a and b but want to compare them.
You first add all the values of a and all values of b.
now comes the trick: divide them by the amount!
And suddenly it dont care how many there are in the group any more.
The miracle of average....
This have nothing to do with my data directly but about this hole blance and high skill.
Go on ladder and play 3 games. In at least one of them a guy crys over balance. And he means the balance of YOUR games not the pro game he saw yesterday! and you know who this players are ? most of them have an account on this site!
So dont act like tl only cares for pro balance... They say they do if they post but most of them only care for blance of their game.
And i understand that ( not the whiners on ladder tho... ) , in the end : Sc2 is our game . Not the game of the pro players.
I know some people are here to only watch pro players. Im here because i like the game. For myself!
But like i said thats off topic.
I agree its interesting to see the balance at different levels and if i have more data i will do this for each league and server.
Also the change that patches bring!
|
On July 11 2012 02:35 Chill wrote:Show nested quote +On July 11 2012 02:25 skeldark wrote:
Than ANY balance is dismissed in your eyes. This is not a question about my data this is a question about the definition of balance!
like i said in my post the data is very biased toward skill. The average of the ladder is gold. The average of my data is high diamond!
Also last time i check everyone on ladder cared for balance at his level ^^ I guess so. I'm just wondering what effects from these results you want. Common sense says that balance at the highest level can be different from the average balance. If this is just for information then it's good to know, but I think a more focused approach would be more useful.
The reason why balance at the highest level is emphasized in common sense is the (quite valid imo) assumption that these players make fewest mistakes and balace discrepencies will be more easily distinguished from player errors when looking at a win/loss scenario.
But in a random samples, such variables do get ironed out. Assuming perfect (or close) balance, a diamond Terran, on avg, will make no more mistakes than a diamond Zerg. Thus, making this result valid.
I'm open to any input if I've made an error here.
|
On July 11 2012 02:41 skeldark wrote:
About balance on skilllevels. At the moment i get enough data that i double the data i used here in 1 week.
So in 2-3 weeks i should have enough to do it only for master+
That would be very interesting. Please keep up the good work, and don't take our criticism too hard. This is how science is made, after all. :-)
|
On July 11 2012 02:40 Mendelfist wrote:Show nested quote +On July 11 2012 02:18 Shiori wrote:On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players. I don't follow you here. The point is that new players choose terran because of the campaign, and some of them later in their career switch race. The switch is important, because that WILL cause an over-representation of terrans at lower MMR. It has been known for a long time, for example by looking at sc2ranks statistics. There is no easy way to remove these types of biases from any data that we have. We also don't know how far up the leagues this bias persists. Is it only in bronze/silver? I have no idea.
The bias you're addressing is (presumably) independent of skill. The over representation of Terran at lower MMR has no bearing on an analysis correlating skill by race with MMR. He's not taking a snapshot of race at all skill levels and saying the Terran average MMR is lower, he's analyzing winrates between races but factoring in the "hidden" MMR rating. In other words, a 50% TvZ winrate may at face value appear balanced, but if the average MMR of the Terrans in that sample is statistically significantly higher than that of the Zergs, it suggests imbalance.
OP please correct me if I'm mistaken.
|
On July 11 2012 02:41 skeldark wrote:Show nested quote +On July 11 2012 02:35 Chill wrote:On July 11 2012 02:25 skeldark wrote:
Than ANY balance is dismissed in your eyes. This is not a question about my data this is a question about the definition of balance!
like i said in my post the data is very biased toward skill. The average of the ladder is gold. The average of my data is high diamond!
Also last time i check everyone on ladder cared for balance at his level ^^ I guess so. I'm just wondering what effects from these results you want. Common sense says that balance at the highest level can be different from the average balance. If this is just for information then it's good to know, but I think a more focused approach would be more useful. Read my frist statement in the thread. I dont want anything. The last thing i want to do is to talk about how to blance or whine about units. I got the data, i publish the result. Thats all. What to do with it can everyone decide for themself. About balance on skilllevels. At the moment i get enough data that i double the data i used here in 1 week. So in 2-3 weeks i should have enough to do it only for master+ Show nested quote +On July 11 2012 02:40 Mendelfist wrote:On July 11 2012 02:18 Shiori wrote:On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players. I don't follow you here. The point is that new players choose terran because of the campaign, and some of them later in their career switch race. The switch is important, because that WILL cause an over-representation of terrans at lower MMR. It has been known for a long time, for example by looking at sc2ranks statistics. There is no easy way to remove these types of biases from any data that we have. We also don't know how far up the leagues this bias persists. Is it only in bronze/silver? I have no idea. There is one way to remove this type of biases. Its very complicated but i try to explain it. you have different amount of group a and b but want to compare them. You first add all the values of a and all values of b. now comes the trick: divide them by the amount! And suddenly it dont care how many there are in the group any more. The miracle of average....
In that case I don't think you understood me. You ignored my sentence about race switching.
Take 100 coins. Place them on the table so that 70% are heads and 30% are tails. Then grab a random amount and throw them up in the air. You will now find that most likely there are less than 70% heads and more than 30% tails. This is what's happening on the ladder.
|
On July 11 2012 02:51 sevencck wrote:Show nested quote +On July 11 2012 02:40 Mendelfist wrote:On July 11 2012 02:18 Shiori wrote:On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players. I don't follow you here. The point is that new players choose terran because of the campaign, and some of them later in their career switch race. The switch is important, because that WILL cause an over-representation of terrans at lower MMR. It has been known for a long time, for example by looking at sc2ranks statistics. There is no easy way to remove these types of biases from any data that we have. We also don't know how far up the leagues this bias persists. Is it only in bronze/silver? I have no idea. The bias you're addressing is (presumably) independent of skill.
No it's not, because the bias is dependent on time (because in a certain time period there always a chance that a player will switch race), and MMR is also time dependent. Therefore you will find a correlation between MMR and race.
|
On July 11 2012 02:51 sevencck wrote:Show nested quote +On July 11 2012 02:40 Mendelfist wrote:On July 11 2012 02:18 Shiori wrote:On July 11 2012 02:17 Chill wrote:On July 11 2012 02:03 Mendelfist wrote:On July 11 2012 01:34 skeldark wrote:
The chance that -race independent- stronger players pick a specific race is near 0.
That's an unsupported statement. I don't know where I've heard it, but I'm pretty sure some Blizzard representative, maybe Josh, has explicitly stated that there is a preference for low level players to choose terran. Does it look the same if you exclude for example everyone below masters? Agreed. The whole basis for this project is defeated by one realistic (in my eyes) claim that is dismissed. Also, why do we care about average balance? If Zerg is easier than Terran from Bronze - Masters, does it really matter to the members of this forum? The fact that low level players might have a preference to choose Terran is not the same as saying that low level talented players have a tendency to choose Terran. If it's just low level players in general choosing Terran, then the average will be sustained by the fact that more untalented and more talented players will be choosing Terran. So the OP's claim is correct, because he qualified it by saying "stronger" players aren't more likely to choose Terran in the sense that they're no more likely to choose it than weak players. I don't follow you here. The point is that new players choose terran because of the campaign, and some of them later in their career switch race. The switch is important, because that WILL cause an over-representation of terrans at lower MMR. It has been known for a long time, for example by looking at sc2ranks statistics. There is no easy way to remove these types of biases from any data that we have. We also don't know how far up the leagues this bias persists. Is it only in bronze/silver? I have no idea. The bias you're addressing is (presumably) independent of skill. The over representation of Terran at lower MMR has no bearing on an analysis correlating skill by race with MMR. He's not taking a snapshot of race at all skill levels and saying the Terran average MMR is lower (that would be easy to do, and wouldn't prove anything since the ladder is biased for 50% anyway), he's analyzing winrates between races but factoring in the "hidden" MMR rating. In other words, a 50% TvZ winrate may at face value appear balanced, but if the average MMR of the Terrans in that sample is statistically significantly higher than that of the Zergs, it suggests imbalance. OP please correct me if I'm mistaken.
No. thats basic it.
Mendelfist i dont throw this coins.
What i do is:
take the avg weight of the 70% head coins and the avg weight of the the 30% tail coins .
And than i say : the coins that show head are 10g heavier than the coins that show tail.
I don't care how many of each are on the table.
I saw some post about winrates. The advantage of this method is: I ignore winrates! I use the skill function of blizzard, the same method they use to get ladder balance data!
Another method to get Pro trends is to watch the MMR - CHANGE of the highest players on ladder compare to their race.
|
|
|
|
|
|
EPT
