|
On September 21 2012 23:37 rikter wrote:Show nested quote +On September 21 2012 23:20 Covariance wrote: S4: Protoss - 616155(36.4%), Terran - 617375(36.4%), Zerg - 460327(27.2%) S5: Protoss - 94502(35.3%), Terran - 89046(33.3%), Zerg - 84001(31.4%)
Good stuff Canikizu! Again, it's been a while since I tested significance for categorical data. I think in this case it's supposed to be a contingency table http://en.wikipedia.org/wiki/Contingency_table. At any rate, someone who's got some time on their hands and a little less rusty with this technique can check to see whether or not the discrepency is statistically significant! @ Rikter - Now that we have the base rates (and we'll assume they're accurate enough) we still have to see whether or not the discrepency is statistically significant, otherwise it could be due to chance. Although given the large sample size and the large discrepency, I have a good feeling about power. I won't discuss this 'crowd computing' thing since I'll be honest with the fact that I have no idea what it means. What I do know is what can and cannot be derived from descriptive stats alone. Canikizu provided some base rates, though, which is cool as that should allow someone to calculate the appropriate inferential statistics. And if someone runs the appropriate test (I suggested one earlier in my post) and finds the discrepency to be statistically significant, then hey, I'll be sold! Best example of crowd computing I could give you is the TV show, who wants to be a millionaire. Ever see it? Or one of its spin offs? One of your options, if you cant answer the question, is to poll the audience. Not everyone in the audience gets it right, but as a group they are over 90% accurate. The bigger the group, the better the results. Thats crowd computing. Basically, if a lot of people are doing something, there is probably some merit to it. I think the two most useful pieces of data are 1) the population of GM, by race, for each season. That way you can take the averages and use em to compute the standard deviation. Then compare that to this season to see how many deviations off we are. 2) Id want a graph of population vs. patch dates. Itd be interesting to see what kind of effects the various buffs and nerfs have had on the population. No correlation would tend to support the idea that aesthetics is the biggest factor in which race you play, while some correlation would support the idea that people are trending towards whatever is most powerful. not only that, results of past year of eu/na pro terrans in tourneys have been really bad. So lets put 2+2. terran in ladder underperforming + terran in tourney underperforming = something is wrong with terran ?
|
On September 21 2012 23:20 Covariance wrote:Show nested quote + S4: Protoss - 616155(36.4%), Terran - 617375(36.4%), Zerg - 460327(27.2%) S5: Protoss - 94502(35.3%), Terran - 89046(33.3%), Zerg - 84001(31.4%)
Good stuff Canikizu! Again, it's been a while since I tested significance for categorical data. I think in this case it's supposed to be a contingency table http://en.wikipedia.org/wiki/Contingency_table. At any rate, someone who's got some time on their hands and a little less rusty with this technique can check to see whether or not the discrepency is statistically significant! @ Rikter - Now that we have the base rates (and we'll assume they're accurate enough) we still have to see whether or not the discrepency is statistically significant, otherwise it could be due to chance. Although given the large sample size and the large discrepency, I have a good feeling about power. I won't discuss this 'crowd computing' thing since I'll be honest with the fact that I have no idea what it means. What I do know is what can and cannot be derived from descriptive stats alone. Canikizu provided some base rates, though, which is cool as that should allow someone to calculate the appropriate inferential statistics. And if someone runs the appropriate test (I suggested one earlier in my post) and finds the discrepency to be statistically significant, then hey, I'll be sold! Sure. Do you need medical history, DNA, criminal records of each players too? I think Terran is UP because of medical reason, Zerg is OP because they have good gene.
|
I find this bw terran comparison argument to be not so good when you talk about SC2. SC2 mechanics just don't matter very much at a certain point. In BW if you werent at a certain mechanical level ( and basically no foreigeners or non kespa pros were) you couldnt even take games off of Kespa Pros. Now in SC2, foreigners can easily win Bo3s, and masters players can easily win games off of any top level player. you even see nonpro NA Gmasters players taking games off Koreans sometimes. So because that mechanical wall doesnt exist in SC2, you can't use that as an advantage with ANY race. so saying "Terran is harder, deal with it" should not be a valid argument. If it was true that even if terran was harder you could win with greater mechanics, I would agree with this argument. But because even that won't help you vs Massive aoe/instant unit production machines of Z/P in later game stages, and everything to punish the races early has been nerfed to the ground through both Blizzard directly and the maps getting gigantic and granting free 3rds, something should be done to change either the design or balance of the game, and this is definitely the main reason people have left terran in droves. Its just too hard, and because 25 terran players in the world are capable of playing it to the max does not mean its fun or manageable for everyone else.
|
|
=p
I know it seems fastidious. And the discrepency is definitely vast. It just annoys me when people misuse stats.
|
On September 21 2012 11:14 xrapture wrote: It's pretty simple, really. Terran has been nerfed every "balance" patch since the games inception. Even when a stretch passed where a foreign Terran hadn't won a major live even in 1.5 years, Blizzard continued to nerf Terran because of a few Top level Koreans. None of us have the slightest chance of ever even being 10% as good as Taeja or MVP-- it's like saying: "well it will balance out once you reach Michael Jordan's skill and talent level."
I guess Zergs just magically learned to play like Fruitdealer when he won GSL and they were struggling, right? Nope. Buff after buff, maps got bigger, and Terran got gutted. Now, in HOTS, Terran is receiving nothing interesting-- a shitty spidermine and firebat.
And, as a high masters Terran, I DO find Zerg easier to play. When I have a advantage, I don't have to worry about losing my entire army in .5 seconds to 1 storm, fungal, or surround. I can just play and comfortably know I'll win.
Zerg players are just inherently better players.
Ok, trolling over.
|
Are we supposed to take anything from this?
For like all of 2011, by far the majority of top players (I guess I'm mostly talking about Koreans) were terrans. Now it's a bit more diverse, but either way, there "being less" terrans doesn't mean they're weak. Other things may indicate it, but not the simple fact that there's less of them.
|
I will note that (as the OP noted), even going down to lower levels, the percentage of Terrans is still low. Were Terrans equally popular as the other races but notably weaker, I would expect an equal number of Terrans as the other races by platinum or so.
They could just be unpopular. Which, of course, would beg the question of why they're unpopular. Are they for some reason un-fun to play as? If this is the case, that's a significantly bigger issue facing Blizzard than a simple balance issue, which can be fixed with numbers changes. Making something fun requires a lot more innovation than that.
|
I think part of the problem is in the past Blizzard patched issues that they thought were prominent instead of seeing if the metagame would change. Now they're waiting for the metagame to change when it simply wont. If Terran was so OP back in the day, then why did fruitdealer win GSL S1? Nestea won three GSLs, MC, won two. It's ridiculous that back then all the issues that were apparent were fixed with patches instead of just waiting to see if players changed the way they play.
|
All of 2011 had the original queens though. They arent really playing the same game now that they were then. I can agree that we could be more sure. Blizz gave a bunch of buffs, and the game evolved. It seems like Terran is being left behind. And in this case, the actual player numbers are the key pieces of information, so the fact that there is less of them means something, whether its being underpowered, too difficult to play at full power, too boring, whatever. Thats kind of the debate I guess.
Edit to add: I dont think playing as Terran isnt fun as a general thing. Some people want to be the humans fighting the aliens, after-all. It just seems like you have to work harder for less.
|
It would be interesting to see an overall statistic that shows how much people are playing terran,toss and zerg. That what you just showed is like saying asia is imbalanced to the rest of the world because of its population. Furtthermore if you look at the top GM spots at every region its clearly that the racial distribution is more even there. But you are toatly right that people get bored by the methodical play style you have to play as terran because T all ins got much weaker. Also you have to invest much more time in playin equal games as you could as protoss because you mostly get in and long eco game as terran due the defensive mechanics the race has to offer.
|
Terran just has a really steep learning curve. To go from platinum to masters with Zerg is fairly easy. To go from platinum to masters with Protoss is fair. To go from platinum to masters with Terran is really hard. You have to learn mechanics AND build orders/strategy where as Zerg only takes mechanics and Protoss takes only build orders usually.
|
On September 21 2012 23:59 Kluey wrote: Terran just has a really steep learning curve. To go from platinum to masters with Zerg is fairly easy. To go from platinum to masters with Protoss is fair. To go from platinum to masters with Terran is really hard. You have to learn mechanics AND build orders/strategy where as Zerg only takes mechanics and Protoss takes only build orders usually.
Please stop. You clearly have no clue.
|
On September 21 2012 10:45 Kfcnoob wrote: Terran is definitely unforgiving and requires alot of skill since refined timings and micro are critical for successful play. As you've said, the pro players have no problem with this sort of requirement. However, until high masters most players will not achieve this level of comfort. Is it imbalance? I'm not sure, but its definitely annoying to see protoss in GM league with 70 apm when no terran under 170 could compete at that level.
This is so true. My thought is that this is imbalance when the skill required is that different between the races.
|
I switched from terran to zerg not because of balance, but because terran is not that fun to play and also tvt sucks because every game takes 45 minutes. There is no balance issues just look at korea...
|
On September 21 2012 10:40 sekritzzz wrote: Its a bit hilarious how all these people are throwing random conclusions such as "terran is a hard race to play, thus less people play it". Its as if sc2 came out a month ago. For people who didn't know terran was quite popular back in the day.... and zerg was played at a 20%~ rate. who cares
On September 21 2012 10:41 FakeDeath wrote: You should used Korean ladder as a better reference. Since Korean ladder is the highest level among all continents ladder. That would be a better judgement. we are talking about lower leagues, not GM Korea where 10 hours a day play evens everything out
|
On September 21 2012 23:44 Barrin wrote: So there is less than 33% terrans in the bracket with the most terrans?
Just so you know this is literal proof that there are less terrans than zerg or protoss. ^ this. I think this is why we look at winrates instead of racial presence when we are discussing balance.
|
On September 21 2012 14:50 AnomalySC2 wrote:Show nested quote +On September 21 2012 14:33 Tachion wrote: Looking more and more like BW, where Terrans complain about how hard they have it while the most dominant and accomplished players in the world are Terran. Terran was the hardest in BW as well. Mechanically speaking anyways. I don't think terran is really all that much harder to play than the other races in sc2, I just think most people prefer to play a passive style and you just can't do that with terran. You need to be a manly man and bring the heat if you want to play terran well in sc2.
Ahaha I love this post. I think Terran just requires more crisp timings and intentionality in playstyle because the flexibility (not variability) of its macro style is much lower than the other races. Micro yes, but it's not the only race that requires it. I can/have gotten to masters with zerg/protoss without any practiced build orders or timings (with good micro and macro), but that really doesn't work with terran.
|
On September 21 2012 11:04 Positronic wrote:Show nested quote +On September 21 2012 10:57 MateShade wrote:On September 21 2012 10:55 dOraWa wrote: It IS a more difficult race to play. When I switched from Zerg, I was mid masters and promptly dropped to low master. It took about 3 months to get back to mid master as Terran, but I'm now considering switching to Protoss because the prospects for Terran look grim and I play the game for fun anyways, not to fight an uphill battle in 2 of the 3 TvX matchups. I have a crazy theory about this. When you began off racing as terran, it wasnt as good as your main race??! and it took time for it to be as good as your main race???? no way!! lmao you are grasping at so many straws dude, it's obvious you don't play terran so why don't you switch to find out exactly how hard it is, or just pipe down with your stupid passive-aggressive bullshit about why people should consider T balanced (even though players better than you will say it has underpowered T3 units compared to the other races, and IS harder than the other races) you just have to look at the statistics and if you aren't smart enough to do that, gtfo
terran is harder. in fact if i switched from zerg to terran id prolly drop from high master to low diamond. but of course your offrace is going to be way worse than a race u never play. unless ur an RTS genius or u play ur offrace quite frequently.
btw to OP, if ur gonna include stats for masters, why not include diamond through bronze. all the players on NA ladder might as well be bronze players compared to the top players.
|
On September 22 2012 00:04 xgtx wrote:Show nested quote +On September 21 2012 10:40 sekritzzz wrote: Its a bit hilarious how all these people are throwing random conclusions such as "terran is a hard race to play, thus less people play it". Its as if sc2 came out a month ago. For people who didn't know terran was quite popular back in the day.... and zerg was played at a 20%~ rate. who cares Show nested quote +On September 21 2012 10:41 FakeDeath wrote: You should used Korean ladder as a better reference. Since Korean ladder is the highest level among all continents ladder. That would be a better judgement. we are talking about lower leagues, not GM Korea where 10 hours a day play evens everything out
who cares bout lower leagues
|
|
|
|