Alright, so I've been having this problem recently where Im doing my build, going about my own business and then all of a sudden all my drones are dead because my spines weren't in time. Obviously that is something that can be avoided, so I decided to start thinking. There are two obvious solutions:
1. Get a build which is safe against any and all hellion timings
2. Scout better and react to any hellion timings
The first sounds good initially, but you'd have to build the spines pretty fast, and if they happen to be putting their cc down fast, then you could be just setting yourself behind. There are options like doing a delayed roach push off 2base, like I've seen sheth and DRG do in the past. Only thing about that is I don't believe its a build that will always work, as if they know you're going to do it, they build a marauder or two and a bunker, which means zero damage. You get map control and push the hellions back, but it does cost a bit to invest in the push. So option two seems like the better option in this case. The question is, how can it be achieved?
Well, for this, we'll assume the drone scout is at 13 supply (Because thats usually when I scout) and will be scouting the terran's gas around 2:30-3:00. The reason I say such a huge timing window is because I like to set my drone to attack the first worker, which may take a while or may be denied right away, depending on if the terran is asleep or not. So then, we can see if the gas is up or not. The problem here being that we don't know how long its been up for, if its completed. So, this brings us to this technique which I heard about in lastshadow's video blogs on strategy.
Clicking the gas to check how much has been mined! Ok, so now we have a number. How does this help us predict when hellions will hit us? Well, if they have three workers on gas, it turns out it takes 1:20 to mine 150 gas. Why 150 gas? Because reactor = 50 gas, factory = 100. So, if we can click on the gas and then predict when 150 gas will be mined by, we can find:
- When hellions will pop out of factory
- When hellions will hit our base assuming they aren't delayed
The second figure being the more important here.
A factory takes 60 seconds to build
A hellion takes 30 seconds to build
The switchoff on reactor takes 8-10 seconds (In the replay I checked with, it took 10 but I think it could be done slightly faster than how he did it)
So, thats approximately 1:10 between the start of the factory and the hellion popping out. The rush distance on metalopolis turns out to be about 30 seconds for hellions, which is obviously not enough time for spines to be started and complete (They take 50 seconds).
So, anyway, what Im trying to get to is this:
Check the gas
Estimate the time that he'll have 150 gas mined by
Add 2 minutes to that time (less or more based on rush distance)
-> That is when hellions should hit your base
The two minutes is the factory construction time + hellion build time + rush distance. Why did I leave out the switchoff time? Because it would REALLY suck to have done all this calculation and have the hellions runby at the last second because the calculation was a bit too exact. I'd rather have it about 10 seconds in advance.
So, the hardest part of all of this is checking the gas and estimating the time, but hopefully this shows some zergs how you can be a bit more creative in scouting. I'm going to attempt to use this method, so please point out any mistakes I may have made. All the timings are taken from a replay I played yesterday vs a high masters terran, who was on top of all his timings. I don't really want to release the replay just for this, because I dont want to make it look like a brag replay, or make the guy feel bad. If anyone finds timings that are drastically different in their replays, I'll have a look at them and sort it out though.
Although, overlords cost 200 minerals, so Im sure I should factor that in too
Edit: Although this seems like a lot of work, if I do it enough times, I will just know when the hellions will be hitting me off of memory and wont need to calculate it anymore.
Dude how do you make such simple things sound so complex?
Haha was nice to read (thought processes are fun) but seems overcomplicated and I don't quite get what you're saying:
"it turns out it takes 1:20 to mine 150 gas. So, if we can click on the gas and then predict when 150 gas will be mined by, we can find"
So... if you click on the gas and see, for instance, 24 gas mined, you'll do some mathematical magic and get a time? And you will try to time your Spine (or whatever response) to finish at the ideal time based on that?
I never knew the calculations had to be so precise, I just assumed early/later hellions was a matter of gas-rax or rax-gas.
And won't Drone-scouting at 13 mean that if you go in the wrong direction he would have his second depot up? (Or even if in the right direction, some Terrans like to get that early depot to deny your 'later-than-9' drone scout.)
Idk. Sometimes I don't even drone scout, I just send like 6-8 drones when my nat is abt to complete while building 4 ling. This helps me push off 2 raxes, and if there wasn't an attempt of a 2 rax bunker rush, send those lungs two seee if they have a reactor or tech lab on their rax/fact.
If its reactored hellion opening, hellions will generally come to you when ur at 28 supplyish and abt 32 supplish if he throws down a CC before making first 2 hellions. In any case, idk why ur making it so complicated lol, jus get ur spine up before 32 supply and u shod be fine. 2 hellions can't do alot of DMg against 2 queens and 4 lings.
___________________________________
Cheese is Art.
I found this article very interesting. The thought process as much as the result. I think thinking that way make you more precise than just relying on experience in the long run and at the very top level precision can certainly make a important difference.
To be honest, the same result could probably be achieved just by clicking the gas and just seeing how long its been up for, to narrow it to general timings. Like, seeing if it was gas before rax, a later gas meaning probably a command centre, just roughly. To avoid overthinking everything, I might try to simplify it down to that.
Perhaps I should have better stated the problem; there are some hellion timings where the gas is put down before the rax, which makes it extremely hard to minimise losses if you don't know the timing of it. There could be other options as well like keeping the drone around to check if a command centre is dropped down or not, and also see when the factory is made to ensure that builds like that aren't a huge issue though.
If it looks like I put too much effort into a seemingly small part of the game, its because I am. I don't want to die to hellions because I didn't expect them to be that early, essentially autolosing.
Edit: Also, I dont really feel as if I should be asking what the most optimal way of doing things like this is, I prefer to see what I can come up with out of watching replays and thinking about it. I don't have to reinvent the wheel, but being able to think for myself is an important part of the game imo
Even the smallest donations help keep sc2sea running! All donations go towards helping our site run including our monthly server hosting fees and sc2sea sponsored community tournaments we host. Find out more here.