|
Ehmmm, hmmm.. how to write it... you know, there is a MU called TvP, where mine drops from early stages of HotS caused a lost game early on. Basically if it wasn't hellbat drop it was a mine drop. There were some DH(?) games, where this kind of drop basically ended the game with 10+ probes dead. It's like the opposite of oracle but just drop, burrow and forget... which is wrong. Buffing early/midgame T in TvZ means buffing it in TvP where midgame isn't problem(IMO), problem is scouting in early game and playing lategame.
Yeh for the first week of HOTS. Then people figured it out, and it was somewhat rarely used post Summer 2013, stop rewriting history please.
Further, this change is completley irrelevant for TvP as extra damage vs splash can be reduced in order to maintain Widow Mines strenght vs protoss while making it better vs zerg. And yes, that was what David Kim implied with his post. He isn't gonna buff the Mine vs protoss.
|
Why do these pro opinions sound like most of the usual drivel you'd read in the balance threads here?
|
Probably because it's been a long time and the problems are very easy to identify
|
Maybe a drop pods mechanics would work? It gives Terran a similar resupply ability as Zerg and Protoss. Could be an upgrade from the Fusion Core (for 100/200/90s) and requires a tech lab on the barracks to use, dropping units in a visible area instantly (and then barracks goes on cool down like a warp gate). Putting it on fusion core and a long research time forces it as as late game upgrade.
|
All the progamers want is more safety, not more aggression potential.
|
On July 08 2014 21:34 havok55 wrote:
Why do these pro opinions sound like most of the usual drivel you'd read in the balance threads here?
Because _everything_ is "drivel" when it comes to balance, everybody has their own opinion and everything else is just so ridiculously stupid.
|
On July 08 2014 19:34 deacon.frost wrote:
Ehmmm, hmmm.. how to write it... you know, there is a MU called TvP, where mine drops from early stages of HotS caused a lost game early on. Basically if it wasn't hellbat drop it was a mine drop. There were some DH(?) games, where this kind of drop basically ended the game with 10+ probes dead. It's like the opposite of oracle but just drop, burrow and forget... which is wrong. Buffing early/midgame T in TvZ means buffing it in TvP where midgame isn't problem(IMO), problem is scouting in early game and playing lategame.
________
Weren't there some games where an oracle or a DT ended a games with 10+ workers kills too ?
What's the difference ?
|
United States7483 Posts
On July 09 2014 04:46 Dandy_Moustachu wrote:Show nested quote +On July 08 2014 19:34 deacon.frost wrote:
Ehmmm, hmmm.. how to write it... you know, there is a MU called TvP, where mine drops from early stages of HotS caused a lost game early on. Basically if it wasn't hellbat drop it was a mine drop. There were some DH(?) games, where this kind of drop basically ended the game with 10+ probes dead. It's like the opposite of oracle but just drop, burrow and forget... which is wrong. Buffing early/midgame T in TvZ means buffing it in TvP where midgame isn't problem(IMO), problem is scouting in early game and playing lategame.
________
Weren't there some games where an oracle or a DT ended a games with 10+ workers kills too ? What's the difference ?
There isn't a big one, it's a similar concept. The only real difference is the amount of time it takes the units to kill the amount of workers. The mine gets more kills a lot faster, then sits for a cooldown, whereas the oracle and DT take a fair amount of time to get the kills, and are slightly easier to escape from. That said, I don't see that there's any real difference here, and I don't think mine drops are 'imba', even if buffed a little.
|
They are too focused on keeping each race in an archetype while trying to make things "exciting". No one wants each race to play the same or be mirrors of each other, but they should ALL have the same tools, or near enough, and if not the same tools they should have a way to counter the things they don't have.
IIt's sad that they can sit and quote tournament rankings from games played by people who make a living playing the game and say things like "we want to see more exciting games/use of this Unit.". I don't want to see an exciting game. It's "exciting" watching a Terran struggle against swarm hosts, meta clouds, and tempest, yes, or watching a widow mine connect. But it's not because it's FUN to watch, it's because I'm sitting here wondering how effective their defense or aggression will REALLY be. Is it enough to stave off defeat? Will they pull through?
You can call a mugging, a car jacking, and stealing something "exciting", but that doesn't mean it's GOOD.
|
stop forcing to make widdowmines work! as for the medivacs nice thought but boosting the unload speed would make earlygame medivac herras vs impossible to hande. making their combat usefullness or reducing their cost slightly would be better i think, or even the movementspeed by a little bit but not the unload speed. I can see it happen already: one moment you are just macroing as a happy zerg with your units securing your mapcontrol, the next moment terran boosts in with his medivacs drops all his units INSTANT in your main, stims/sieges, kills your entire main lifts up and boosts away before speedlings even arrive and while you think about trying to stay in game your see your base on the other side of the map being overrun by the same units that just killed your main. I agree terran needs a buff, but not the speed+unload speed buff as that would make the game just more annoying and in no way more fun than a normal terran buff.
|
Snute asked for ""the correlation between Mutalisk count and win rate", so I ran some numbers.
![[image loading]](http://i.imgur.com/N6kZWjp.png)
The data comes from approximately 2100 ZvT games from roughly April 2013 to present on Spawning Tool. The vast majority of replays are from tournament replay packs and should represent high-level play, though other games are not filtered out.
The frequency of counts tails off, and I truncated the data around 50ish since past that, there were fewer than 10 games at each point. The raw data is available here.
Looking at the data myself, I'm not seeing a lot. There may be a bump around 16 Mutas, but otherwise, it's hovering around 50% with some variance due to small sample sizes. What do you guys think?
|
On July 09 2014 23:44 DeathSoror wrote:Snute asked for ""the correlation between Mutalisk count and win rate", so I ran some numbers. The data comes from approximately 2100 ZvT games from roughly April 2013 to present on Spawning Tool. The vast majority of replays are from tournament replay packs and should represent high-level play, though other games are not filtered out. The frequency of counts tails off, and I truncated the data around 50ish since past that, there were fewer than 10 games at each point. The raw data is available here. Looking at the data myself, I'm not seeing a lot. There may be a bump around 16 Mutas, but otherwise, it's hovering around 50% with some variance due to small sample sizes. What do you guys think?
Needs to be weighted for how many games are at each point. If 80% of games are around 30 mutas then then it will look much different than if 80% of games were only 10 mutas.
Example:
Say the sample size is 100 games. 50 of those games were reached 30 mutas and the rest are divide evenly. According to this data then, 50% of games have an 80% win rate in favor for Zerg.
Ultimately though, this confirms there is a "sweet spot" or "critical mass" for mutas and that if you make mutas, aim for around 30 because that will indicate you'd have a maximum advantage.
Edit: This also shows the early pressure from Terran. When you rush mutas, youre vulnerable to a timing attack by Terran, and at that time you'll have 3-4 mutas but end up losing becuase you don't have anything else. You can see that on the win rates that Zerg loses often when they fail to defend a common Terran timing attack at the 3-4 muta mark.
Edit2: It also shows the 16 muta point which is also an ideal point where 16 muta secondary bounce deal around 50 damage which kills Marines and SCVs a whole attack cycle faster. Its also around this point you can one shot medivacs with mutas stopping drops. When you get to 30 the bounces can kill medivacs and you can one shot turrets and have enough DPS to kill beefy units and have enough DPS to negate small packs of marines.
|
"I talked to bbyong and he said that TvP lategame is hard mostly because of feedback which is devastating to packs of ghosts."
How can a single-target spell can be devastating to a pack of units, whereas one spell from the said pack can shutdown a huge pack of templars ? Usually by the time you can feedback every ghost you'll take enough EMP to have no energy for anything.
|
On July 10 2014 00:04 Sakray wrote:
"I talked to bbyong and he said that TvP lategame is hard mostly because of feedback which is devastating to packs of ghosts."
How can a single-target spell can be devastating to a pack of units, whereas one spell from the said pack can shutdown a huge pack of templars ? Usually by the time you can feedback every ghost you'll take enough EMP to have no energy for anything.
I'd love to know this, too. Even if you use a stray templar to feedback, you've essentially emp'd your templar, anyways. Is the plan at a high level to snipe obs then use mass cloaked ghosts?
|
On July 10 2014 00:04 Sakray wrote:
"I talked to bbyong and he said that TvP lategame is hard mostly because of feedback which is devastating to packs of ghosts."
How can a single-target spell can be devastating to a pack of units, whereas one spell from the said pack can shutdown a huge pack of templars ? Usually by the time you can feedback every ghost you'll take enough EMP to have no energy for anything.
Feedback is instant. Feedback will hit before EMP and then EMP will never occur. You can spam Feeback, its cheap and devastating unlike EMP where missing one can cost the game.
EMP may have a 1 range advantage, but if you're not on point and aim the burst (which requires no unit to target) and you just happen to get in range of the Feedback (which can be instant and queued) then you lose a ghost. Even if you EMP a unit, it doesn't actually kill them while feedback will.
It comes down to the fact that mistakes with ghosts are easier to make an more costly than missing a feedback is for templar.
|
On July 10 2014 00:04 Sakray wrote:
"I talked to bbyong and he said that TvP lategame is hard mostly because of feedback which is devastating to packs of ghosts."
How can a single-target spell can be devastating to a pack of units, whereas one spell from the said pack can shutdown a huge pack of templars ?
High level Protoss hide Templars to flank and can quickly shift Feedback 4 Ghosts, which can result in up to 4 instant casualties. Trading 1 HT vs 2-4 Ghosts is of course very favourable.
|
On July 09 2014 23:44 DeathSoror wrote:Snute asked for ""the correlation between Mutalisk count and win rate", so I ran some numbers. The data comes from approximately 2100 ZvT games from roughly April 2013 to present on Spawning Tool. The vast majority of replays are from tournament replay packs and should represent high-level play, though other games are not filtered out. The frequency of counts tails off, and I truncated the data around 50ish since past that, there were fewer than 10 games at each point. The raw data is available here. Looking at the data myself, I'm not seeing a lot. There may be a bump around 16 Mutas, but otherwise, it's hovering around 50% with some variance due to small sample sizes. What do you guys think?
Seems like in Korea, zergs are favoring more early banelings, (especially True) over getting a big muta flock up quickly. Killing Terrans with overwhelming Baneling numbers, with just a small flock of muta to harass, from majority of the Zerg Proleague games and Code S in Korea.
Seems like greedy baneling styles are giving Terran more issues than mass muta in Korea. Watch more games decided on early gas in upgrades and banelings than mutas.
Ever since the Mine nerf, Banelings have been the issue, not really mutas, with Terrans not having a effective way to kill mass banelings.
|
United States7483 Posts
On July 09 2014 23:44 DeathSoror wrote:Snute asked for ""the correlation between Mutalisk count and win rate", so I ran some numbers. The data comes from approximately 2100 ZvT games from roughly April 2013 to present on Spawning Tool. The vast majority of replays are from tournament replay packs and should represent high-level play, though other games are not filtered out. The frequency of counts tails off, and I truncated the data around 50ish since past that, there were fewer than 10 games at each point. The raw data is available here. Looking at the data myself, I'm not seeing a lot. There may be a bump around 16 Mutas, but otherwise, it's hovering around 50% with some variance due to small sample sizes. What do you guys think?
Dat 29 mutas.
|
On July 09 2014 23:44 DeathSoror wrote:Snute asked for ""the correlation between Mutalisk count and win rate", so I ran some numbers. The data comes from approximately 2100 ZvT games from roughly April 2013 to present on Spawning Tool. The vast majority of replays are from tournament replay packs and should represent high-level play, though other games are not filtered out. The frequency of counts tails off, and I truncated the data around 50ish since past that, there were fewer than 10 games at each point. The raw data is available here. Looking at the data myself, I'm not seeing a lot. There may be a bump around 16 Mutas, but otherwise, it's hovering around 50% with some variance due to small sample sizes. What do you guys think?
Over all it seems to be pretty noisy.
But I think I can see certain general trends:
0-16; WR <50% : Games were Zerg dies of while transitioning to mutas.
17-32; WR >50% : Games were Zerg builds a Mutaball and does not replace any.
33+ WR =50% : Games were Zerg either trades Mutas and rebuilds them or were Zerg masses them.
|
On July 09 2014 23:44 DeathSoror wrote:Snute asked for ""the correlation between Mutalisk count and win rate", so I ran some numbers. The data comes from approximately 2100 ZvT games from roughly April 2013 to present on Spawning Tool. The vast majority of replays are from tournament replay packs and should represent high-level play, though other games are not filtered out. The frequency of counts tails off, and I truncated the data around 50ish since past that, there were fewer than 10 games at each point. The raw data is available here. Looking at the data myself, I'm not seeing a lot. There may be a bump around 16 Mutas, but otherwise, it's hovering around 50% with some variance due to small sample sizes. What do you guys think?
I'm not sure if the X axis label is just a typo or if you actually did the analysis against "Mutas Made".
I don't think that will solve for the root of what Snute was asking if so.. I think you'd have to use Mutas Made - Mutas Lost.
A player who is sloppy and has to spend all his gas to keep his Muta count high is much less likely to win..
Am I misunderstanding the plot?
LOL this shit still makes me so sick too
[image loading] Acer.Nerchio: Yes, i think drops are too powerful and zerg is too weak outside of creep while being too strong on creep.
Medes are too strong.. Mutas are fine.. but Medes are too strong.. same cost - slower - cant attack ground or air - have 0% chance of entering an area with adequate static and don't regen.. but too strong..Mutas tho.. NP
|
|
|
|
|
|
EPT
