|
As you mentioned what people do often is completely tunnel visioning on microing their units and forgetting to macro at that point, mechanically that's one thing you probably need to be watching for and trying to train your brain to be aware about multitasking in any small breaks.
On September 04 2026 22:30 WGT-Baal wrote:
Another question for you though to keep the thread moving: what do you do when you watch a replay and genuinely do not know why the opponent is so ahead at a given time? try to find similar openings on the same map in a pro VOD?
Yeah, when I can't figure it out I tend to binge watch pro videos and find similar games, I feel super stumped quite often.
Ultimately I'm noticing that the devil is generally in the details, so if you see the pro guy doing a similar thing overall to what you're doing, you have to deep dive into the replay to notice things you are doing wrong, usually bad economic management, in my experience. In fact, I rarely see something crazy different to what I was trying, or else there are several options and they all work if well executed.
It's kind of crazy when the game seems so mapped out how often I lose to something and feel stumped, but often those are the best things to analyze as it's a way to find proper gaps in your understanding of the game.
I had an issue with goliath mech terrans in ZvT, I just had to learn to be way more greedy as they have no early pressure, but also not then over-greed and prepare for the timing push.
What was happening is I was over-defending like against a fictional bio terran, and then over-defending a ~9 goliath push which can kill you sometimes, and then keep making units as well from there and droning way too late
You just need to barely defend the low number of goliaths (vision is important, burrow/overlords/lings etc) and mostly greed to defend the stronger 10-11 minute push with 24 goliaths and 3 tanks.
But I saw in CWAL replays many other Bish rank zergs, while playing similarly, manage to beat the same builds playing a bit more aggressively using mutas to trade their hp for SCV kills (goliaths and turrets kill mutas slowly without focus fire) and using more sunkens to delay, for example, so several styles work (muta/hydra with mostly hydra vs mass muta and sunken first)
|
Great, tell me all of the mistakes I made in this game. I think I only made one: I didn't switch to hydra in time. I don't see any other glaring errors when I look at the replay, so maybe you replay analysts can tell me what I did wrong other than the ONE mistake I pointed out (late hydra transition/late hatchery add for the hydra transition)
https://repmastered.icza.net/game/1wu65LoDH-Zufgxfco2L9T-SrEhXSBfYwEx7afEfSCo
of course, I should have countered into his base knowing that his army will kill my army, but I didn't know that at the time, I thought I would win, so now that I know I would play different
|
On September 05 2026 14:58 iopq wrote:Great, tell me all of the mistakes I made in this game. I think I only made one: I didn't switch to hydra in time. I don't see any other glaring errors when I look at the replay, so maybe you replay analysts can tell me what I did wrong other than the ONE mistake I pointed out (late hydra transition/late hatchery add for the hydra transition) https://repmastered.icza.net/game/1wu65LoDH-Zufgxfco2L9T-SrEhXSBfYwEx7afEfSCoof course, I should have countered into his base knowing that his army will kill my army, but I didn't know that at the time, I thought I would win, so now that I know I would play different
Here's my opinion, obviously a better player could probably provide a more accurate analysis
+ Show Spoiler [game review] +-You get full scout of CC first, not many marines produced and even goliath mech confirm with the 2 ling poke, but you are not really diverging your build at all, to take advantage of Terran having almost no early pressure aside from a couple of vultures -You get ling speed at a normal time, but you don't really produce many lings. It kind of worked out because you managed to kill the vulture with your speed lings, but I feel like this is unreliable. With the scout I would be looking to skip or delay ling speed for an earlier 3rd hatch @ 3rd base or as a wall off -At spire finish you immediately hatch 5 mutas and keep constant production, check JDs game vs CC first mech by flash here as an alternative option + Show Spoiler [ASL21 game] +https://www.youtube.com/watch?v=gnRgC33s7YA&pp=ygUUamQgdnMgZ3JvdXAgb2YgZGVhdGg%3D Basically you can just build 1 muta to fend off vulture and put down two hatcheries and drone up instead. -You have 10 drones at main and only 6 drones at natural (7 patch base) for most of the game, if you are playing this low eco at least rebalance them, you are losing a bunch of minerals for no reason. I think it's probably viable to play with full muta production on 2 hatch if you harass well but I doubt its optimal. Going drones until the last minute makes more sense with a fixed amount of mutas needed to harass before he gets goliath range -Terran has basically no map pressure from 5 to 8 minutes, yet you are producing almost all mutalisks and very few drones and hatcheries, so your eco is shot. You are working with 3 hatch production when you should have 5. Your 4th hatch is really late compared to a VS BIO build where you need less eco as you rely on defilers and muta micro to survive. -Your muta harass is pretty wasteful. The zergs I see do well get a lot of SCV kills at 7-8 mutas (8 mutas to have option to 2 shot goliath) while goliath range is not finished, you can tank a bunch of shots and move from main to nat sniping SCVs, ignore the missile turrets and try to find spots only 1 can hit you, goliaths are immobile and bad moving between bases. -You had a big window where terran got vulture speed instead of goliath range (a mistake?) to harass SCV that you didn't use. -You mainly look to harass like you are playing vs a bio player, taking out turrets so you can pressure SCVs, but golaiths with range mean you basically can only harass when goliaths are not there, so you use the harass more to make sure he stays in base, and not looking to lose mutas for free, or trade them for turrets. Terran will eventually be GAS constrained and not mineral constrained so taking out turrets is not going to slow down his timing push. -You didn't take burrow, this is an interesting option against mech to guarantee map vision -Terran doesn't move out until 11 minutes, and you have a really low drone count, only 4 drones at natural, 4 hatcheries total to produce, vs 5 fact terran. You had opportunity to drone up to 5-6 hatch with additional 10-15 drones by 7-8 minutes on the clock and be already mass producing mutas and hydras. -Why are mutas idling outside his natural close to goliath range? That's a recipe for baiting a bad fight which is what happened, you can get away with it more vs bio due to the range -The rally point also cost you some mutas, but that's a mechanical thing to remember to change it -You picked a fight close to his base, most zergs I watch are stalling to the last minute, you have about 6 mutas in eggs, and are fighting around 20 goliaths +3 tanks with +1/0 with 22 1/1 mutas but the engage favours him as you have to move into range. You obviously lose this fight and then you manage to put up another fight again randomly in the middle of the map? You should be in desperation stalling mode 1. You have 850 minerals, you can pull drones from gas and minerals in natural to put down 6 sunkens or more to buy time 2. You can use the 2 groups of mutas to cut off reinforcements or try to dive the factories to force him all in 3. The sunkens will allow you to keep massing another 6 mutas in eggs + 6~ mutas while he has to siege/unsiege while you yourself cut off reinforcements 4. Despite all these in my opinion mistakes where you lost many many mutas, and took two bad fights, the game was still close -I'll also point out you have no defense in your main against vulture run-bys at all, you are relying on mutas. Now, if you are going full mutas he probably won't produce many vulture rounds, but It feels risky as a terran with good multi-task can run by with 3-4 vulturesand you lose every drone -As to your comment about not knowing if his army will kill yours or not, unless you are 100% sure the safe play is to assume it will beat yours, and delay the extra 60s building sunkens while denying reinforcements, if you STILL can't beat his army you can try go for base trade and put sunkens at your 3rd base and a spare spire. Once his factories are dead he can't produce more. At your gas mining rate you can get 9 mutas per minute plus anything in eggs so denying reinforcements and delaying with sunkens has the option to increase your army count by around ~50% this early in the game! The sunkens delaying also gives you something to lock down the terran and give you vision, so when you do engage with mutas your timing and angle will be much more favourable than running into him when he has more vision and range
To not hide behind my forum ID - here are my last 3 games against goliath mech, with no selection bias as you can see from the game i got smashed by just 9 goliaths (where I had a terrible early game defense vs vults)
https://cwal.gg/viewer?match=MM-5E260C5E-A7AD-11F1-866E-FA167B5461B6
https://cwal.gg/viewer?match=MM-0DF13202-A647-11F1-866E-FA167B5461B6
https://cwal.gg/viewer?match=MM-1A7E39FE-A622-11F1-866E-FA167B5461B6
and here is a selected game which I think was actually a decent game although I lost I felt the necessary adjustments were in sight to have a good game
https://cwal.gg/viewer?match=MM-8348390A-A319-11F1-866E-FA167B5461B6
These games are played at an around B rank mmr but I think my approach is ok or at least much better than it was before
|
In terms of ling speed, it's going to be useful almost no matter what he does, except for mass vulture where you need like muta basically
6 drones on nat is a mistake, for sure, but you're suggesting different strategic approaches that work better sometimes and worse in other cases
But yeah, like I said the biggest mistake is thinking my bigger army would win, but at 1-1 goliaths have an advantage and I need to switch to hydra at like 9 minutes or counter into his base.
Skipping speed for burrow won't make a big difference if my lings are sitting in front of his base either way
Transition to hydra early is a different strategy altogether and might get shut down by an early goliath pressure against the third, it's a different style altogether that some people favor
I appreciate the review!
|
Burrow, ling speed, and so on are minor details and up for debate, the 10-6 drone split is also a minor point, just a loss of ~30 minerals a minute, where you had enough.
Regardless of if you want to get hydra earlier or not, the economy management I think is the main part which could be better, i.e. having more units, drones, and total production at the same point in the game without sacrificing a whole lot. It seems to me you macro'd two hatcheries with mutas and drones from the third without really ever thinking about the optimal timing of drones versus mutas versus saving up for extra hatcheries.
The only reason I mentioned ling speed was to ask if you even thought about delaying it or skipping it to get a faster hatchery. Actually delaying/skipping it makes more sense against factory expand where they make 4 vultures with speed and lings are useless. But the main point is there are many decisions being made automatically and any automatically made wrong decision can lead to losses where it feels like you made few mistakes. Usually losses come from accumulation of minor mistakes or disadvantages (build order counters)
The way you approach taking the fight is the other critical point, I don't think there's ever a reason to fight the mech army in the middle of the map - actually it was closer to his natural, you have many advantages to stalling at your base (time, reinforcement cut off, vision so you can time your 1a2a3a, creating the option of a base trade)
To even have the option of countering the main you need to get units out of goliath range, and goliaths away from his base, so idling everything at his natural just helps him pick off mutas for free, you really want him to have to choose between killing your base and saving his, putting him all in
|
On September 10 2026 03:49 Slayer91 wrote:
Burrow, ling speed, and so on are minor details and up for debate, the 10-6 drone split is also a minor point, just a loss of ~30 minerals a minute, where you had enough.
Regardless of if you want to get hydra earlier or not, the economy management I think is the main part which could be better, i.e. having more units, drones, and total production at the same point in the game without sacrificing a whole lot. It seems to me you macro'd two hatcheries with mutas and drones from the third without really ever thinking about the optimal timing of drones versus mutas versus saving up for extra hatcheries.
The only reason I mentioned ling speed was to ask if you even thought about delaying it or skipping it to get a faster hatchery. Actually delaying/skipping it makes more sense against factory expand where they make 4 vultures with speed and lings are useless. But the main point is there are many decisions being made automatically and any automatically made wrong decision can lead to losses where it feels like you made few mistakes. Usually losses come from accumulation of minor mistakes or disadvantages (build order counters)
The way you approach taking the fight is the other critical point, I don't think there's ever a reason to fight the mech army in the middle of the map - actually it was closer to his natural, you have many advantages to stalling at your base (time, reinforcement cut off, vision so you can time your 1a2a3a, creating the option of a base trade)
To even have the option of countering the main you need to get units out of goliath range, and goliaths away from his base, so idling everything at his natural just helps him pick off mutas for free, you really want him to have to choose between killing your base and saving his, putting him all in
6 drones on nat is a mistake, for sure, but you're suggesting different strategic approaches that work better sometimes and worse in other cases
You have 12-16 larva per minute out of 3-4 hatcheries that game you have to decide which unit to make, at which point does making a drone instead of a muta become a different strategic approach? There's a sliding scale toward "all-in aggression" versus "total greed" that is somewhat stylistic but I think making more than 11 mutas when they can't reasonably do much damage to the terran when you have the opportunity to get additional drones and hatcheries FIRST to have even more mutas later, is most probably just a mistake, not a strategic approach. Being over-greedy of course can be a mistake too.
Yeah, going heavy drones directly into hydra maybe you can call it a different strategy. But delaying a few mutas to get a 5th hatchery and a few more drones, and then re-spending the gas allows you either transition to 5 hatch hydra OR even just 4 gas mutas (so you have 12 mutas per minute instead of 9, while terran is capped at 12 goliaths per minute if he spends every bit of his 2 gasses on just goliaths. 12 goliaths still wins but you can try to kill his base and then engage later)
Of course, you can argue the terran can move out earlier in the exact window the drones haven't paid for themselves (<1 min after you made extra drones), but terran has to wait for goliath range, for siege tanks and siege, and moving with pure goliaths on 3 facts is kind of unattractive, as your ability to reinforce with 9 mutas/minute and also place sunkens makes it a dilemma for him to decide to punish or stay in base
Additionally, he has no way to know if you are droning or making pure muta, making it even harder for him to decide to move out.
Even typing this creates so much uncertainty in my head where the right answer is, what constitutes stylistic choice vs mistake, that I would never look at a game like that thinking I made just "one mistake"
I would approximate that the earliest a terran can reasonably move out with both goliaths and tanks I would say is around 9 minutes from my experience, so you can make a lot of drones up to about 8 minutes into the game, at which point there is a judgement call about how to balance production between 100% mutas or some drones also.
Any move out earlier than that is probably punishable even if you got greedy, by putting down sunkens, stopping reinforcements and waiting for eggs to hatch before taking the fight.
The main way it kills you is when you weren't aware they even moved out, so the new eggs are drones, thats when having burrow can be great but any vision at all helps
|
You can see that by sitting outside his natural I snipe 3-4 goliaths when he's moving his army and the rest is clumped inside his natural behind a wall. If after this I re-rally to my own natural and make a few sunkens I would be fine to take a fight a bit closer to my rally point
By "one mistake" I'm referring to my assumption that a bigger muta army can beat goliaths, but you clearly need to focus fire and moving shot better. I accidentally pressed Ctrl+E that brought up the surrender menu, which happens a lot when I try to hotkey mutas to 2 and 3
If I assume the Terran army is slightly stronger, then I don't take the bad fights and stall a bit more, take the 4th gas faster or make hydra earlier.
So far, I've seen about two most common timings, the 12 goliath push against the third, and the 10 minute goliath+siege push. So the idea would be to drone in between these timings, add hatcheries and have a huge army ready by ten minutes
|
So, my overall view is that you've misplayed the game by the time that engagement has happened. Meaning I think you should have more drones, an extra hatchery, and planning a defense at your base instead of an engagement at his.
I think the bad fight just highlights that the game is not going as well as you thought it was, rather than it being your main mistake.
Whatever about judging who's army is bigger, it's easier to fight closer to your rally point than his. He has to walk to your base, walking further from his rally, and closer to yours, meaning you gain 30 seconds of production for free while his army will take 30 seconds longer to rally. If you fight at his nat he's getting 5 extra goliaths to come in to the fight which can swing it in his favour. If you fight at his nat you have no vision to even accurately judge how big his army is. These are game fundamentals which you can improve instantly in my opinion.
This is a big weakness in replay analysis as both of us are likely to be wrong or inaccurate about most of our ideas about the game (unless you are one of the people who believe that mechanics is the main difference between pros and average) , but I think slow, careful thinking through replays and games is one way to get better at it.
Point by point - I had a quick look at the replay again
You can see that by sitting outside his natural I snipe 3-4 goliaths when he's moving his army and the rest is clumped inside his natural behind a wall. If after this I re-rally to my own natural and make a few sunkens I would be fine to take a fight a bit closer to my rally point
You both lost a couple units due to miscontrol. He sniped some mutas of yours, you sniped some goliaths of his. That happens. Just because he happened to accidentally have 2-3 goliaths wander out does not make it a good decision. In my opinion, those idle mutas are asking to be sniped when you aren't looking at the screen in favour of your base or main muta flock. That also makes it easy to get baited in a bad engagement near his base
By "one mistake" I'm referring to my assumption that a bigger muta army can beat goliaths, but you clearly need to focus fire and moving shot better. I accidentally pressed Ctrl+E that brought up the surrender menu, which happens a lot when I try to hotkey mutas to 2 and 3
You have a relatively similar unit count, about 2 groups of mutas versus 2 groups of mech, and your mutas are fairly damaged as well. Focus fire and moving shot would make no difference, the mutas already were targetting the front few goliaths and only a couple of hits happened before you realized you wouldn't win the fight.
There are several things working against you here
-He's moving out into the map without your direct vision, good luck getting your mutas clumped perfectly and doing moving shot on two groups of them while trying to get 3 (2 + straggler mutas) groups of mutas into the right range.
-Goliaths have longer range than mutas, this means you can never "poke", you always want a decisive fight, or else he gets free shots at the start and end. This means you shouldn't "guess" if he has more army or you do
-Goliaths are stronger than mutas, you want to make sure you have at least 20-30% more mutas than he has goliaths
-Your mutas have hp damage from doing harassment
-Microing the mutas becomes impractical at large numbers, as you'll just lose the clump and then it becomes impossible to efficiently moving shot
-He is next to his base and can rally in 5 goliaths when normally they have to walk across the map
If I assume the Terran army is slightly stronger, then I don't take the bad fights and stall a bit more, take the 4th gas faster or make hydra earlier.
It's already too late for all of this. You don't have enough drones to make tons of hydras, you have only 4 hatcheries, making hydras at this point would be wasteful because it's just giving his tanks something to hit. At this point in the game, with your economy setup, going all mutas is the correct play to stop this push (with proper stalling and reinforcement sniping) and it would have worked I think because you did kill off enough golaiths in the earlier game.
Think about it this way - your economy is 3 gasses muta on 4 hatchery, you have enough drones for that, with 4 hatcheries you can only produce 14 0/0 un upgraded hydras per minute when he has 3 siege tanks, the 9 1/1 mutas per minute sounds better than wasting money on the hydra upgrades this late.
Hydras are great when you are up to 5-6 hatcheries with enough drones to support it and have prepared the upgrades.
The logical next step for you is to get 4th gas and continue pure muta, which you were doing. Terran would probably need science vessels to counter that, as pure goliath is risky due to base trades.
I do wonder if you can get away with just going 4-5 gas all muta against this meching builds since they build 3 siege tanks and can you just kill off their factories and re-engage at your natural? Maybe the problem is when they get up to a 3rd gas and science vessels
So far, I've seen about two most common timings, the 12 goliath push against the third, and the 10 minute goliath+siege push. So the idea would be to drone in between these timings, add hatcheries and have a huge army ready by ten minutes
I disagree on this- you want to drone and add hatcheries as early as possible in the game - before he has the ability to punish you, if you drone in-between you are barely getting value out of the drones before his push arrives.
These timings are somewhat arbitrary, he can move out at 6,7,8,9,10,12,18,24 goliaths. The only significance of 10 minutes is it usually takes about that long to get 5 facts and siege mode up. He can produce about 12 gols a minute on 5 fact so the difference between 12 and 24 goliaths is just a minute. However, the timing gaps can be bigger as Terran often starts with 2-3 facts so you often have like 9 goliaths move out around pretty early.
In principle, you should be MUCH greedier before the first push, because as long as you have 4-5 hatcheries online, the stuff coming out of eggs is a much bigger % of overall army than later in the game. If he has just 10 gols getting 5-10 hydra out of eggs is enough to defend with muta while if he has 25 gols and 3 tanks getting a few more hydras out of eggs doesn't matter as much.
This does create a series of dilemmas which is the meat and potatoes of the struggle vs meching teraran
1. Just how greedy can you be to still have enough units to stop an early push with 7-12 goliaths, assuming you see exactly when he moves out? I think the answer is fairly greedy, with 8 mutas, you definitely have enough time to get 2 sunkens at a base and whatever else out of eggs.
2. How many mutas to make for harassment purposes as you can't make mutas to harass and drones both. If you make mutas for harassment, how much damage are you willing to take to snipe SCVs, knowing he could try to move out soon to punish.
3. Do you take a 3rd and 4th gas or else go more on minerals and get heavier hydras instead of mostly mutas?
4. What if Terran doesn't move out at 10 minutes, and slowly just secures his 3rd and gets science vessels and more mech and upgrades? How do you manage your drone production if he delays the push to 12-13 minutes. Is there a window to get queens in? How many units do you need before you should feel confident you can switch to queens and have time not to die waiting for energy?
5. Does it make sense to tech to hive for dark swarm as it's still useful on hydras vs goliath and tank balls
|
Of course it's too late at 4 hatch to make hydra
But basically there's no point in moving out with like 6 goliaths because the reinforcements will crush that even if I was droning before the move-out
12+ is basically the same thing without siege, you can make a few sunks and be fine
So you have to decide if you get way more muta, throw down sunkens and base trade or you stop making mutas and transition into hydra. I see those as both viable options vs. this
If the Terran tries to get a third base you can just go queens/hydras. Note that I saw he went 5 fact so he won't have starport
|
|
|
The reason to do 12 hatch is when you want to earlier scout otherwise you float larva, I'm guessing. (Since I've seen pros do both builds, scouting earlier with 12 hatch)
Here you got pretty lucky I would say
-You have overlord scout on top of his 8 rax
-Despite getting scouted, he doesn't start his bunker until way too late, I think he should be starting it and moving the SCV off when it get harassed, if this happens, his bunker would finish earlier and you lost too many drones (?)
-You only pulled 5 drones?
-You made an overlord lol
My question is, in ladder games whats the value of an earlier scout (~33% of the time it's a waste by default) versus getting the lings 5s faster.
For example, what if he does CC first, or BBS in center, in both cases do you scout in time to react? I don't know
Soma seems to go 11 hatch or overpool as standard so seems he decided thats the way to go but not everyone is doing it
Let's say I'm playing a tournament game, and want to Cross scout to check for BBS in the center, or be able to react to a CC first by playing greedier does that give me an edge versus having a cleaner game vs a scouted 8 rax
Also there is more game theory
-It's 33% for you to overlord scout his cliff and 66% for him to spawn elsewhere
-He can build 8 rax away from the cliff so you never scout it if you don't drone scout.
-I assume the game theory is that he wants to wall 8 rax if the thinks you might 4 pool and if not he always should hide the rax further out and float it back
I think you're oversimplifying, but I think 11 hatch is already the standard now? 12 hatch isn't used as much
|
In similar note, I got smashed by a nice timing push at 9 minutes with around 19 goliaths 2 tanks
https://cwal.gg/viewer?match=MM-21283012-AEA0-11F1-866E-FA167B5461B6
Trying to figure out where to improve
He did zero harass so I think I was late getting my production and drones online and also I tried to cut him off on the highground in center instead of stalling where I can get more hydras out
I also think I need to choose between getting 7-8 mutas early to snipe SCVs or powering hatcheries, on cross spawn probably makes more sense to try to just power up
I am always at a loss about how much to harass the goliath terran player.
If he invests in vultures I can afford to delay things but when he doesn't I probably should take my 4 bases and 6 hatcheries with 1 drone per patch and that should be enough to hold the 9 minute push
Pure mutas always seems hard to me when he has +1 armour goliaths x20 at 9 minutes. The hydras seemed to do well despite not getting range upgrade in time or +1 attack in time
|
I don't think he can get many seconds on the bunker because I kept the scouting scv low hp
I could have pulled all the drones back and lose 2 fewer, clean up anyway - I only pulled 5 because I didn't mean to fight with them too much, my pool was very close to done! The idea is to delay like a few seconds
as far as your replay, you're not doing boosting that well, overlord at 0:56, hatch at 1:36, 1:57 pool, 2:00 drone, gas 2:07
you give up and go for the 3:01 lair timing because you're not on pace for a faster one
that's 4 seconds delayed compared to me at the start, if you faced a bunker rush you would need to respond with a lot more drones
the lair being later means you didn't get a fast 7 muta timing
so Terran gets up to 40 scvs at 7 minutes on like 4 turrets and some goliaths and you didn't harass him - I harassed just a bit at this point killing one goliath, but this made my opponent make a siginificant investment into turrets
there's no point in having full health mutas, since when he comes out at 9 minutes you would have had 2 minutes to heal
you harass him at 9 minutes which is a bit late anyway, he's already attacking you at 95 supply while you only have 72 (although hydra cost more minerals per supply than muta and need more larva)
my opponent only has 81 supply and no siege at this point because he keeps spending his minerals on turrets, he's floating 300 gas because he's concerned with my harass and I start killing more scvs which makes him have less income
I think the mutalisk harass did a good job, it made my opponent's push out slower
If I would change my rally point away from his base at some point I wouldn't lose mutas for free
back to your game, you should 100% make 7 muta and 1 shot scvs until your mutas are hurt
having 5 muta did nothing except kill some random marines which are useless, just 200/200 more and you could have racked up 4-7 scv kills since opponent didn't turret very heavily
at 7:30 you made 36 drones but you didn't make a sixth hatchery which means you will have too much money and not enough production for hydra
to make the 6th hatchery you need to pull off second gas for two minutes, trade 600 gas for about 400 minerals (right after you decide to drone you should pull off the gas at 5:50)
then throw down three hatcheries right away starting from 6 minutes by floating 2 larva at each hatchery and mining only 1 gas, get back on gas when you get more drones and start den/evo in main
you should make hydra den and evo in the main base in case you base race and have to mass sunken on the high ground (two base mech all-in often ends in a base race)
you also forgot hydra range
at 9:00 instead of 183 minerals 939 gas would have 600 minerals 300 gas, but the 6th hatch would have been long done already so you could have already had 6 more hydras (only 450/150) and hydra range done
I did a single player game, at 9:00 I had 80 supply (I lost like 6 supply fighting the computer) and you had 72 with both having 5 mutas so I just had 4 more drones and 4 more hydras and I did start my range upgrade which would be done soon
I could just make fewer drones and have 8 more hydra which would be a significant amount more in that fight
I couldn't get the cwal.gg to get the replay up so you'll have to either download it here
https://repmastered.icza.net/game/6kh9mk2Bjo8A3Y-0gBkOYqnVTOFDHy8lvsqDctCa1FY
or just trust me you could have had a lot more
|
Thanks -
-The building placement into main for evo/den you are absolutely right about. Good point. Also creates the question about if one of the hatcheries should go into main highground location.
-The start, yes if you see I fucked up the split totally, I'm not good at boosting anyway especially on bad turn rates, I do try.
-Agree on getting 7-8 mutas OR just 1 into 7-8 instead of 5, 5 seems a bit useless versus mech. Maybe I should delay flyer carapace to rush 7 mutas as it's finishing early anyway? By the time I had enough mutas to harass he was already moving out, which was earlier than I expected.
-Our games were different, can't compare them directly, since my opponent optimised purely for fast goliaths without vultures, he had more stuff to defend but I had a comfortable time expanding and droning up. He had more goliaths, facts, and range by 7 minutes so harassing was more difficult to be fair. But the back of his main and back of his nat should be options.
-I over-made drones for 5 hatch, intending to put a 6th hatch later.. I wanted space to comfortably build my economy, seems I don't have a window to get 6 hatcheries. I don't want to get stuck on 30 drones if terran doesn't go for the timing push. Maybe I need to find a number like 24 hydras before expanding my drone count.
-Interesting idea, cutting 300 gas a minute for 114 minerals seems rough.. I kind of prefer the banked gas to give me option to switch into all mutas or queens if he transitions out of goliaths? I I was aware I was floating gas, but it's usually useful later when they get more siege tanks. Perhaps its worth it to allow me to get into 6 hatch hydra and 4 bases and take the third gas later to catch up. My feeling is I should commit to just 5 hatch production and making a couple less drones and a later 6th hatch.. trying to get more SCV damage with mutas and more stalling and reinforcement cutting. The single player game you played a bit greedier than me since you get 3rd hatch at 3rd base, no sunk, because you aren't walling off against a potential vulture threat, also I had a big supply block at 60 supply, and tons of floating resources, so I believe you about taking drones off gas but my supply should have been higher just with better mechanics.
I'm torn between 3 ideas
-Going more into mutas early on with 3 gas and switching to hydra late like the pros do, as the upgrades are expensive you can get harass and counterattack value out of mutas earlier. You still go up to 5 hatcheries but you are going over 11 mutas at the start
-Your idea to go straight to 6 hatch hydra (but with 4 bases if he's not getting vulture, then I need just 1 drone per patch, with a 6 patch mineral base
-What I did but getting 7-8 mutas first, then staying on 5 hatch hydra 2 gas, and just trying to hold the timing push and figuring it out after 10 minutes if it doesn't come, prevent him getting an easy third gas and deal with vessel/tank focused push.
The problem with taking drones off gas is you have no room to switch into queens/pure muta until you get 3rd gas online, its tricky. But what I did is kind of middle of the road, not good, I am too used to having until about 10 minutes before the push comes, against an optimized single player build that he did, I don't have that luxury.
|
https://cwal.gg/viewer?match=MM-EC3443D0-AEAB-11F1-866E-FA167B5461B6
Here's an interesting one, I played pretty terribly, mechanics wise, slow build, terrible/non existent harass, hydra range was basically missed until the end, engaged for no reason in mid when i should be stalling, using my whole army and building lurkers and spending all my resources
But what was interesting is he went like 12 goliaths into marine medic tank ball. I wonder what the play is against that if you don't instantly scout it to go hive?
So I just go for lurker/ling/hydra/muta (lol) army crush?
Should I try to just muta micro pick off tanks or something and pretend goliaths don't exist? It's a funny comp, in theory it sounds terrible, as he's getting all units and 1/1 upgrades on both bio and mech, I guess if you scout it you can play even greedier as his push will be delayed
I'm guessing that lurkers with some muta snipe micro to delay while teching to hive is the safest, maybe just adding lurkers gives you enough DPS as I had enough economy and supply it's just his army was way more efficient since hydras are ass against marine/medic/tanks
It's kind of like the valkyrie build where he trades off having early pressure + a big bio ball with having a marine/medic/tank push that is hard to snipe due to valkyries on top of it
With valkyries you usually want some kind of lurker ling or muta ling army crush to abuse less marine dps
|
The problem with just banking the gas is that your gas wasn't used in the first replay so you died gas rich
Hydra/lurker beats any comp that's not like half tank.
|
Fair, it's just hard to accept that big loss trade in resources as being optimal, it very well could be
My problem with the gas banking is, adding more mutas (after 8-11) instead of more hydras I don't think makes you fight better against mostly goliaths, but the swap to mutas or queens is incredibly strong if he goes mostly tanks after a first push, so ideally I want to use that gas later, if I can get away with it
Muta switch I guess makes it harder for him to take a 3rd as well as its harder for him to turtle up if he needs units in his main base as well and turrets against 11+ 1/1 mutas, well you need a lot of them
---------Second game
I don't feel like hydra/lurker is THAT strong but when my economy is that good and he has a weird mix of units you are probably right
Lurkers are probably not as weak as I think they are against mech units when you play a flanking/surround attack as well compared to just camping them where they are out ranged or even ranged and don't splash
I just realized guardians are an interesting option if you get to hive against that weird goliath/tank/marine/medic comp as he doesn't have many goliaths and no vessels if you have 11 mutas lying around I don't think he can deal with hydra/guardian at all
But lurkers you can actually get out in time
|
That's not a good way to think about economy management. You should be thinking of "how can I optimize the stage of getting to 6 hatch hydra?"
because if you get slowed down at any point Terran has more timings to straight up kill you
thinking about having broodling at like 14 minutes when you're at minute 6 is crazy, especially since you just die to timings at 10-12 minutes
there's another way to go about fighting a very mixed comp and that's just 4 gas muta with upgrades, make infinite sunkens and counter into terran base
|
I don't necessarily agree - if there's no other way to hold the timing sure, but I don't think that optimizing versus one exact timing is a robust way to play, suppose the terran decides to play for 1/1 or 2/1 or stall and take a base and attack with a later timing, with more tanks and vultures, and you are stuck on pure hydra, it might be a bad situation.
I did lose a game yesterday because I optimized against the last game and this guy just played more defensively, got vessels and mines, harassed with vultures and got tons of siege tanks and pushed out later with 2/1.
Now, I played completely terribly, but the problem was trying to optimize against what you lost to last game, you are just doing a local optimization, it doesn't mean you have the more robust way to play overall.
It should have been an easy win as I had lots of queens but I just didn't drone enough as I'm stuck in a mindset of just holding a push which might come in the next 30 seconds. This is also about reacting properly to information after scouting, which I did but not well.
What you really want is just enough production to survive without sacrificing your ability to scale later. Probably at some point once your production is online and you have some army you need to be actively scouting and trying to figure out how many more drones you can get constantly.
On re-watching the first game where I died at 9 minutes, I wonder if the main issue is actually me engaging too early and even after that, I lost a disgusting amount of hydras for free when I kept trying to take fights instead of waiting till the last second, if I used my first army to just cut reinforcements I can probably stall to 10 minutes and engage with enough to wipe him out.
That said, my production probably did start just a bit too late, that's part of the reason hydra range was missed because hydra den itself was late. (Or at least, late for what I was doing, seems more common to stay on mutas for longer)
I'm thinking that rather than rushing to 6 hatch getting just enough production for 5 hatch hydra, and getting hydra production started earlier, and starting a 6th hatch and mixing in 1-2 hatch drone production after that if he delays the push might be smarter, but then that's basically what you said at the start, that I made too many drones for 5 hatch.
If I have like 11 mutas and 12-16 hydras I can still harass and switch to all hydras if he moves out and drone up if he is teching or expanding.
There is something I don't do enough which is mixing unit/drone production, I get the tendency to either pure drone and then pure units, that leads to easily being too greedy or too poor. It always feels optimal to pure drone at the right timing but when facing someone on constant unit production it does make sense to match it at least somewhat.
I've also considered a decision tree for when to greed vs rush to 7-8 muta
VS 2-4 speed vults/dropship etc
-Get the mutas as you need them to secure expansion anyway
-Send one to scout while rest clear vultures to get expansion
-Harass ~at 7 mutas (scvs) or 8 (2 volley goliaths if they are not too many)
-Because he won't be able to get goliath range (due to getting speed and/or mines) and as many goliaths (due to investing in vultures) you should be able to get more damage done and potentially snipe goliaths without losing too much
VS goliath rush with no vultures or just 1 vulture without upgrades
-He will probably have goliath range done sooner and more goliaths to punish poke with a fast push
-Instead get 1 muta and expansion and the rest drones before switching into mutas and later hydras
-Vulture should be easy to kill as it's alone and not upgraded
-You need more mutas fast as his goliaths can move out from as early as 7 mins. Hydras might be late
-You can still harass but its more difficult if he stays in his base
thinking about having broodling at like 14 minutes when you're at minute 6 is crazy, especially since you just die to timings at 10-12 minutes
When playing ZvT vs bio you get a queens nest fast to tech to defilers at like 7 minutes so they are up in time to defend at 11 minutes, while at the same time Terran is trying to kill you or at least your third base with 4-5 rax medic marine. You wouldn't advise someone dying to medic marine busts on 5 rax not to tech to hive, you just have to figure out how to hold it while still teching to hive because he MIGHT just not bust you if you defend well and wait for 3 tanks and a 2 vessels and you are cooked
I don't think it's crazy to plan ahead, and the gas as I mentioned in the short term can be used to switch into more mutas which feels very strong if the goliath count ever drops off.
Queens are just the obvious next step against a terran who decides not to push based on his info and then wants to build up siege tanks, it's very hard to deal with otherwise once he gets enough, and going into defiler/ling/ultra without upgrades probably takes too long
It's interesting because when its bio->mech nobody ever would play hydra/queen. I think ling/ultra is just way stronger against mech but you don't have the time and upgrades against a mech start
I do find hydras are particular garbage against tank/vulture with minefields, you have no way of dealing with the mines, hydras will attack them and sometimes run into them while getting blasted by tanks. At least melee units run in and drag mines
|
there's another way to go about fighting a very mixed comp and that's just 4 gas muta with upgrades, make infinite sunkens and counter into terran base
This is true, I always found going mass mutas versus goliaths kind of counter-intuitive, but what it is good at is forcing the terran to stay on basically all goliaths. There are a lot of builds that go into vessel to deal with it, also goliaths scale better with upgrades and he's easier for him to get double upgrades.
The risky part with pure muta is that if you fall behind at all on muta you can never win the fight and are forced to try to base trade and it gets messy
At least muta/hydra can fight it cost efficiently it's just a tight window to scale up from 3 hatcheries.
There is a lot to be said about proper scouting too as it's too easy to fall into the trap of guessing or scouting just a couple times and optimizing into the last scary version of whatever you faced, only to find out the guy is doing something else
Greedy play tends to be very strong obviously if he doesn't elect for the natural timing attacks, as even if you scout late your economy is already better, but if you optimize against the timing attacks you won't be greedy
|
|
|
|
|
|
EPT![[image loading]](https://i.ibb.co/sJFDMNPF/Screenshot-20260912-162439.png)
