EPTRax buildtime --> 3minutes
Designated Balance Discussion Thread - Page 1221
| Forum Index > SC2 General |
|
Clonester
Germany2808 Posts
Rax buildtime --> 3minutes | ||
|
SC2Toastie
Netherlands5725 Posts
On January 22 2015 05:13 Dwayn wrote: So blizzard is finally nerfing the raven? Awesome, I've been saying for ages PDD is broken. It took too long, but I'm glad that they finally realized it. Next they really need to tackle the 2-rax issue. We shall follow thei comments, oh prodigy! What has thei in mind for fixing that dreaded two rax strategie, oh great seer? 2rax is fine, l2p/l2scout | ||
|
Deleted User 137586
7859 Posts
![[image loading]](http://i.imgur.com/VI4BnQJ.png) As you can see, there a lot of variation but that's because the end of the year had fewer games. For example, the biggest swing happens at list 125 (end of November, beginning of December) which had only 1600 games, compared to the 4500 games of list 128 (that finished on January 21st). Due to the massive volatility, it's difficult to see what's going on, but one could probably conjecture that the MUs are all close to balanced as there's no persistent trend. The closest is the slight advantage T has over P, there are swings but the swings don't take P above 50% on more than one occasion. There's a reversal of fortunes in PvZ, where P was suffering and then ended up swinging above 50%. But that MU is very even in terms of winrates. TvZ is mostly volatile with swings to the advantage and disadvantage of both T and Z. Not much can be deduced from this. Regarding populations, I took the number of mirror matchups as indications of tournament populations. To give a quick indication of the proportions I summed up all the mirror MUs and then took the percentage of each mirror from that. ![[image loading]](http://i.imgur.com/y8is2IG.png) One would expect an even distribution of 33% of each mirror. But what the graph shows is that there are way more ZvZ than TvT, and there are roughly the number of PvP that you'd expect, with volatility above 33% rather than below. For TvT, there was a slight trend towards 33% until list 126, but in 2015 that trend has been reversed. Interestingly, ZvZ has very little volatility, and it was mostly that TvT seems to have replaced PvP, with only a slight effect on ZvZ from more or less TvT. Some conclusions: From the old wisdom that there are less T players playing in tournaments, it looks like it's still the case that the very best of T are playing a wider range of Z and P. Unlike before, they are doing quite well (or around 50%, with swings either way) in both MUs. It seems that Z is the limiting factor on T, as T seems to be having an easier time against P at the moment. The T numbers seem to keep P weaker, which is supporting Z numbers, as Z seems to have a weaker MU against P. Here follow the old lists. On October 02 2014 06:56 Ghanburighan wrote: Aligulac list 120. ![[image loading]](http://i.imgur.com/Xhy2haa.png) We continue to have fewer games than usually, so the numbers below are more volatile. But we're at 2,5k, down from >3k games, so we're closer to a representative number of games than before. Winrate analysis Let's look at ZvT first, unlike every list since the patch, Z has overtaken T with a winrate of 51%. One could concede that list 117 broke the mold as well with 49.85%, so basically even 50%. But one could also make an argument for a trend which favours Z. List 118 was at 44%, list 119 was at 46% and now it's at 51%. So Z is doing better with every list. But it's only 3 lists of 2 weeks each. More time (and more games) are needed to assess the situation. PvT seems to be in a volatile spot. 118 was at 47%, 119 at 51% and 120 is now at 45%. There is no trend, but P took a beating these last two weeks. This might just be a symptom of the lists including fewer games than usually. PvZ is still Z favoured. List 120 was at 47%, list 119 at 49% and list 118 at 48%. So Z has a consistent but very slight advantage over P. There is no real trend. Population analysis The population numbers seem to have stabilized to list 119 levels. P has 2/3 of Z mirror matches. While T lags behind with around 1/3 of Z mirror matches. So the T population remains lower than the P population, and much lower than the Z population. On a related note, I wish they removed SEA games from aligulac... | ||
|
Tuczniak
1561 Posts
It's interesting how much Aligulac balance statistics are useless. | ||
|
Thieving Magpie
United States6752 Posts
| ||
|
SC2Toastie
Netherlands5725 Posts
| ||
|
Pirfiktshon
United States1072 Posts
This suggestion = Terran being useless and just trying to survive more so than normal in both matchups LOL Production for terran is one of the greatest weaknesses it takes mining time and extra cost per building per reactor and such it would just collapse terran as a race all together LOL | ||
|
Pirfiktshon
United States1072 Posts
| ||
|
The_Red_Viper
19533 Posts
On January 23 2015 00:44 Pirfiktshon wrote: 2 rax is a scouting issue and as stated above who cares LOTV is going to destroy this notion anyway Yeah it will become 3 or 4 rax instead | ||
|
Thieving Magpie
United States6752 Posts
On January 23 2015 00:44 Pirfiktshon wrote: 2 rax is a scouting issue and as stated above who cares LOTV is going to destroy this notion anyway Whatever the fastest rush of each race is, that rush will always be vilified. Doesn't matter if it's proxy-gate, 2rax, or 3hatch-no gas. The problem is not the game, the problem is gamers in the internet. | ||
|
igscubanex
5 Posts
| ||
|
y0su
Finland7871 Posts
On January 23 2015 13:41 igscubanex wrote: Guys what do you think of inbase naturals ??. As a Zerg i think that they are a bit imbalanced in favor of P/T What aspect? Scouting, punishment, meta/bo gamble or other map feature (hard to take 3rd etc)? | ||
|
Svizcy
Slovenia300 Posts
On January 23 2015 13:41 igscubanex wrote: Guys what do you think of inbase naturals ??. As a Zerg i think that they are a bit imbalanced in favor of P/T I don't really see problem with them, since Nexus first or CC first is easily scoutable, and you can just take extreamly fast 3rd then as a zerg. Why do you see it as being imba? good day, svizcy | ||
|
SC2Toastie
Netherlands5725 Posts
On January 23 2015 18:16 Svizcy wrote: I don't really see problem with them, since Nexus first or CC first is easily scoutable, and you can just take extreamly fast 3rd then as a zerg. Why do you see it as being imba? good day, svizcy The problem is more in scouting past the initial units are out. Scouting a natural is a pretty important thing. That being said, Zerg has the least reason to complain as they've got flying, unkillable supply depots that can get pretty perfect scouting on the natural. If you want to know everything; drone scout. You're not Soulkey, you don't need the tiny edge, stop being that greedy and whine about it ^_^ | ||
|
Big J
Austria16289 Posts
On January 23 2015 18:38 SC2Toastie wrote: The problem is more in scouting past the initial units are out. Scouting a natural is a pretty important thing. That being said, Zerg has the least reason to complain as they've got flying, unkillable supply depots that can get pretty perfect scouting on the natural. If you want to know everything; drone scout. You're not Soulkey, you don't need the tiny edge, stop being that greedy and whine about it ^_^ The problem is Zerg aggression and counterattacking, or rather the lack of it. It has nothing to do with CC first or nexus first. The midgame and lategame is very hard for zerg on these maps and forces you into very predictive turtle-builds. Though I think Vaani vs Terran isn't bad, just very boring. | ||
|
igscubanex
5 Posts
| ||
|
Deleted User 137586
7859 Posts
On January 23 2015 21:51 igscubanex wrote: More like lets said that the P tries to do a 2 base all in, and then i make a lot of units and he backs, now i cant counter atack a 1 ForceField Ramp. And it also limits a lot of 2 base agression from the zerg. I think the last thing we need is map limiting our builds. Don't we? I'd say it's generally a good thing to have different maps in the pool, such that not every game plays out the same. But to keep people from playing their favourite or the `optimal' build, you need maps to act in a limiting fashion, making the build less optimal due to the map architecture. I don't think we should focus on such maps, but having one in the map pool seems to shake up the games nicely. Another question is whether these maps are imbalanced, as you say. Without committing either way, I'd say that it's not too impactful as long as there are a large number of different maps in the pool, each providing advantages to different races and different builds. P.S. I wrote a summary of Aligulac Lists a few posts back, but there was almost no response. Is the lack of response due to the fact that it's Aligulac or because it was decidedly pointing towards a lack of balance issues. | ||
|
Big J
Austria16289 Posts
On January 24 2015 00:16 Ghanburighan wrote: P.S. I wrote a summary of Aligulac Lists a few posts back, but there was almost no response. Is the lack of response due to the fact that it's Aligulac or because it was decidedly pointing towards a lack of balance issues. You do a good job with those. The only thing that I could point out (again) is that aligulac is not completely coherent with the Korean, European and American prolevel. E.g. In Korea the amount of mirrors and race distributions currently in proleague, NSSL and GSL is T=P>Z, so the exact opposite of aligulac. In EU and AM Challenger it is roughly equal for all 3races, so still not coherent with the "more Zerg and Protoss mirrors" from aligulac. I've been arguing this now for about a year or so and therefore watched the data for a long period. The aligulac data isn't more than a balance indicator. We have seen times with only barely over 50% winrate for Protoss against Terran, but in Korea blink allins wrecked the Terrans and led to 16Protoss and 3Terrans in GSL. We have recently seen times in which Protoss was looking like it was not playable against zerg for short time-durations on aligulac, while in proleague and multiple challenger/qualifiers PvZ was at around 75-80% winratio and struggling in not a single league. I know I sound like a broken record, but that is why I didn't reply to the data in the first place. Aligulac looks nice currently. More intersting is that all the Premier and Challenger Leagues look balanced, with some neglectable distribution disadvantages for Zerg in Korea. | ||
|
w3c.TruE
Czech Republic1055 Posts
On January 23 2015 21:51 igscubanex wrote: More like lets said that the P tries to do a 2 base all in, and then i make a lot of units and he backs, now i cant counter atack a 1 ForceField Ramp. And it also limits a lot of 2 base agression from the zerg. I think the last thing we need is map limiting our builds. If P tries to do a 2base all-in and fails, you don't have to counterattack. You should be on 3 bases obviously, so you just contain him on 2 bases, tech up and take forth eventually. Also 2 base aggresion from Zerg is terrible regardless of the map... | ||
|
HellHound
Bulgaria5962 Posts
On January 24 2015 02:25 w3c.TruE wrote: You can't make it easier to defend the all in by forcing warp ins at home or slowing down the push by posturing your units agressivly. That's a really big deal.If P tries to do a 2base all-in and fails, you don't have to counterattack. You should be on 3 bases obviously, so you just contain him on 2 bases, tech up and take forth eventually. Also 2 base aggresion from Zerg is terrible regardless of the map... | ||
| ||
![[image loading]](http://i.imgur.com/IaF4Rp3.png)
![[image loading]](http://i.imgur.com/mpO5wqJ.png)
![[image loading]](http://i.imgur.com/9r04q4l.png)
![[image loading]](http://i.imgur.com/mqKrKFk.png)
![[image loading]](http://i.imgur.com/6zp9WrK.png)
![[image loading]](http://i.imgur.com/280eKxW.png)
![[image loading]](http://i.imgur.com/Oc9x1bJ.png)
![[image loading]](http://i.imgur.com/pKEYuFe.png)
![[image loading]](http://i.imgur.com/wem39XJ.png)
