Get the Desktop App for Battle.net Now
- All your games in 1 place
- Log in once
- Automatic game updates
Alrighty, I've done several guides on several things. Now I want to go in-depth on some things. If I go over the same thing twice I'm sure you'll forgive me so lets get into this.
Q: What is a transition? A transition is when you are doing one thing in your build to get to another thing. Sounds simple right? Lets get to an example that you are all familiar with.
Example: 5 rax reaper. Remember this? People made a ton of reapers to force Z to make combat units and to deal damage to Z. When Z went to lair the transition for 5 rax reaper was marine/marauder. See how well that flows? Also because reapers were very gas intensive but not cash intensive you could expand fairly quickly due to your pressure on Z and your influx of money. Wow that build works well.
Q: Ok, I understand transitions, why does it matter? Transitions matter for the same reason that having a plan matters. Your objective is to kill your enemy, having a plan puts you on a path to do that. If you say you want to have 3 base and keep your opponent to 2, then figure out a build that transitions into being able to do this well.
Example: 5 rax reaper again, the point was to make zerg and you on equal~ terms as far as econ goes. If you went ahead and transitioned from 5 rax reaper to say... banshee how does that make sense? You now have to build more infrastructure, banshees won't handle well against lair, and it's going to cost you alot to get there. It was a transition to bio because the transition worked well. You already had the barracks.
Q: Ok, so transitions are important, how do I do this in my build? From the same above example, if you're going for bio don't all of the sudden decide to mix in battle cruisers. Plan your build out. With planning and foresight you can figure out your transitions.
Q: I'm having all sorts of problems defending, I can't stop drops, muta harass, protoss warp ins. Is there even a way to defend against this? Yes, every race has a way to detect and defend against these things. Before I answer how to do it lets look at the other 2 races. Protoss has observers, they are invisible scouts. If you make several observers you can gain a great deal of map vision and thus ensure your ability to detect these things. Zerg has overlords, even better. They can/do spread their ovie's out all around their base and other places. Zerg has the best map vision. Terran has sensor tower. The reason this thing was put into the game was to give us some form of early detection. I am NOT saying to build a bunch of sensor towers every game. I AM saying that if your opponent is dropping on you or muta harassing, build a sensor tower or two to give you early warning.
Q: Ok great, now I can defend my airspace from harass. Every time I move out though I get zerglings running by me to attack my mineral line in my natural. Protoss waprs in zealots on a hidden pylon. Terran runs some hellions by. How do I deal with this? Two ways, number one is to wall off your mineral line. If you have to build supply depots, why not use those depots to make some form of wall? Second is by taking a very small control group/your reinforcements (depending on how big of a harass it is) and killing it. If you're attacking and it's just a small harass don't turn around.
Q: Wow that wall off works great! What if it's a full counter attack with his whole army? This is where you make a split second decision to either base trade or defend. Do NOT hesitate. If you're going to base trade A move your army to his base and then start lifting and running your workers. If you're going to defend then run your workers back and move to engage his army. Indecision will kill you here.
Scouting: I know I've done scouting before, but I think it needs to be re-emphasized.
Q: How do I scout? Taking terran as our example you have several options. You can use buildings, you can use your workers/marines, you can use an air unit, you can scan. All of them have their advantages and disadvantages.
Q: Ok, well how do I know what to scout with? Much better question. Unless you have to don't use scan. Use a worker or marine to scout their unit composition/front base. When I say scout I mean I want you to watch that unit run in, die, and see what it dies to. Use a building or air unit to scout their base to see what their tech is like. Use marine/SCV to scout for expos. Use scans late game to scout for your army when you're attacking. Scans > mules late game.
Q: Whats the best way to ensure no hidden expos? When should I scout for these? I like to set an SCV to patrol a group of 3 expos, then just scout other expos as needed. I do this around 10-15 mins into the game. I want to know when my opponent expands, what his timing is, and what the saturation looks like. Remember expansions are money sinks until you pay for both the expansion and the workers working it.
Q: I saw your How To: Attacking thread and I still am not quite sure I understand timing. Timing is a window in which you have a upgrade/key unit finishing and you can use that to attack. The point isn't to upgrade something and then sit around and not take advantage of it.
Example: Stim is about to finish (10 seconds) I am going to move my army now so that when it finishes I am in position to attack. Thus I have created a timing attack.
Q: That makes sense, I also hear day9 talk about timing all the time. What is he talking about? He is talking about the probable times in which something will likely happen. It is near impossible to do it before, it is unlikely to happen if it comes after.
Example: Mutalisks. We all hate them. For a zerg generally a lair starts at between 5:30-7 minutes. Mutas usually do not come out anytime before 9 minutes. Thus I can infer that if I do not have missle turrets up at the 9:30 mark I have a big chance of losing workers to muta harass. The timing for mutas is 9-10 minutes. You will almost never see them before 9 and it is unlikely to see them after 10 minutes.
Q: Oh ok, cool, how do I figure out timings? The best way is to test it by going into a game vs. an AI and just running the same build over and over again. Then you can see the optimal timing of a build. I usually add about 30 seconds for imperfections/adjustments and then you have a timing window between that time and say 1-2 minutes after that time. Now you know when you need to be afraid of something.
Edited by Beammeup on 2/11/2011 8:24 AM PST
Punishing Your Opponent:
Q: What does this mean? I've heard it before. Punishing your opponent is a timing attack in which you have the ability to make your opponent pay for something he has done.
Example: Your opponent expands early. You have two options, punish him or expand. He has sunk 300-400 minerals on an expansion and he is missing that much in either infrastructure, tech, or army value. Sometimes it's more then just the minerals, he can also cut gas to do it. Think of the older 1 rax expand.
Q: Why not just expand? He can't attack if he has expanded already? Well that is an option, it isn't even a bad option. However, the reasoning behind punishing him is usually to take an advantage and expand shortly after.
Example: Protoss expands early. I decide to drop in his base. I kill many probes and army value and start my expansion. I have now decimated protoss and I am nearly on even terms as far as expansions.
Replay: This is a replay that is a perfect example of this.
Q: Can I punish my opponent in more ways then just an expansion? I like the idea of punishing him. Well... you probably have some sort of sick fantasy... ANYWAYS... yes you can. His army is out of position? Punish him. He leaves a collosus alone? Punish him. There are several ways to punish him. Even taking a ninja expo is punishment for not scouting.
Q: I think I got everything you said. How do I put it to practice? You win with it. At some point it isn't about practicing and instead it's about winning. Eventually you're going to reach a limitation of how well YOU can play. That limitation is pro level, but if you don't have the time to dedicate to practice then you're going to hit a place where you can't advance. Now you just win.
Q: What do you mean? If you have to proxy racks in their base to win every game then do it. It's not about if you're cheesing, if you have everything I've said down as best as you can then just win. Explore your options, have fun, win. Figure out what works best for you. My way of playing isn't your way. My strats aren't your strats. My thought process isn't your thought process. If you decide T isn't your main way to win, then play P or Z. As long as you have fun and your winning... well thats the point right?
I'd like to finish by saying no amount of practice can make someone prepared for certain things. If you find a new way to all-in (cheese) then you can win games with it. I am by no means advocating cheese, I am saying that when you reach your potential for grasping and improving on the concepts of gameplay then however you win is how you should play.
Even the pros get cheesed. Even the best players lose to that. If I was in a tournament with players like idra and jinro I'd cheese every game because thats my best chance to win. There is no shame in it.
I hope it was an interesting read, if you guys have any other suggestions on something I should make a thread of let me know.
This is Beammeup, peace.
Edited by Beammeup on 2/11/2011 8:39 AM PST
Beam, although a little off topic you should definately consider doing a How To: Builds describing how one should go about constructing a build, mastering a build etc. This might be able to help progress the currently changing meta game and help an inflow of thought provocations and reduce QQ.
On topic great post. Very helpful in specifics. A lot of people really dont know how important transitions really are, especially with MMM in PvT.
Regarding scouting for tech, the best option depends on the map and your BO. If its close by air position (ie. Scrap Yard) then a floating barracks works really well. If its a far position you're better off using a reaper or viking/banshee as per your build order. Sometimes base cliffs aren't that vulnerable to reapers though (Steepes), in which case its viking/banshee or scan.
Versus T and P, I always keep one marine at my opponents natural to see when he expands. Versus T you can often kill the building SCV to delay it a bit. For me anyways, its really important to know who expanded first. It determines whether I'm going to macro up and do a 2 base push, or if I'm going to harrass/push to put us on even ground.
Winterborn was livestreaming himself practicing a build to beat the new tankless tvt strat. Unfortunately his livestream crashed and his 46 minute stream was lost. It was very informative. Watching his thought process on doing a thor/ marauder build. Watching timings and etc.
Threats of violence. We take these seriously and will alert the proper authorities.
Posts containing personal information about other players. This includes physical addresses, e-mail addresses, phone numbers, and inappropriate photos and/or videos.
Harassing or discriminatory language. This will not be tolerated.