Search| Features \ \ Strategical basics | ||||||||||||||||||||||||||||||||
In today's installment, we take a look at one of the fundamentals of Starcraft at any decent level of play: Scouting and Countering. Inside I explain why, and also give solid advice on how to learn to execute this part of your game properly.1.1a - Necessity of Scouting in GeneralScouting and Countering are of utmost importance to a Protoss player. If you want to have a decent chance of coming ahead after the early/mid game, you will need to scout and react to what the opponent is doing. As previously stated in the introduction, Protoss is the countering race. When facing a Terran, he will have a few strong openings against you, especially on Lost Temple, Jim Raynor’s Memory, and other maps with high ground overlooking your natural that requires a shuttle to reach. When facing a Zerg, he will have even more strong openings to use against you; especially on maps in which he can secure his natural more easily than you (almost all land maps unfortunately). So why do you have to respond to their threat, and not the other way around? Well, this is where the inherent strengths of the different races come into play. Protoss has a variety of openings as well, however, unlike Terran and Zerg openings, Protoss’ openings are only good in special circumstances. The basic reason for this is the mid game threat period. Protoss’ mid game threats simply come later than their Terran and Zerg counterparts’ do. In addition, Protoss’ mid game threats lack the same kind of bite that Terran and Zerg have. Protoss’ best chance does not lie in a sudden unexpected strike, but rather in the gradual overwhelming buildup that is easier for us to perform than our counterparts. Here are some brief examples, though of course this will be gone into in much more detail on the race specific sections. 1.1b - Necessity of Scouting vs TerranSay your Terran opponent is going early dropships on you. If you had similarly decided to go for reavers, you will be at a distinct disadvantage. Not only will he be ready to attack first, requiring fewer tech structures, tanks building faster than reavers, and dropships being produced at a separate structure (as opposed to a robotics having to produce a shuttle and then a reaver, the Terran can produce tanks and dropships simultaneously), but his tanks will be superior to your reaver in a fight.
1.1c - Necessity of Scouting vs ZergIf your opponent is Zerg, the news is even worse. Again an example: suppose your Zerg opponent is going for a 2 hatch lurker rush, and you're going for a hard timed zealot attack. You’re planning to catch your opponent off guard with a big force of 12 or more zealots, but right when you get to his base lurkers are hatching. He can easily hold you off, chase you to your base, and possibly even climb your ramp.
Zerg is just never going to be in a position where he’s randomly choosing his build order because he will always know, or have a good idea (or at least think he does) of what you’re up to, and he will certainly act based on that information. However, YOU still will have to guess or at most make a well educated guess about Zerg’s intentions. And even if you ARE right, Zerg can tech switch on you so fast that it may hardly matter. Protoss’ best chance to beat Zerg is with a single, sudden, crushing strike (however, don’t limit yourself to thinking that means an attack, seizing and holding a critical expansion can accomplish the same effect). Wars of attrition will generally favor the swarm, as they can continually shift their tactics, will constantly be on the attack, and will always be reacting to and countering you more quickly than you can do the same to them. This is because of a variety of factors. In the early game, sunkens build so quickly that even if Zerg makes a minor misjudgment of your army, he can construct them in time to save himself. If he has an overlord or zergling scouting the entrance to your main, and you attack with slow zealots, he can start building sunkens and have them finish before your zealots get to his main on most starting positions of most land maps. This makes Zerg safe to do as he will, knowing he can always bail himself out if necessary. Cannons warp 10 seconds slower and Zerg units travel much faster than slow zealots, so Protoss has no such similar luxury. Zerg’s advantage only grows through the mid and late game, as Zerg has centralized production facilities: he doesn’t need to worry about whether and how many gateways, stargates, and robotics facilities to produce—all he needs are hatcheries, no matter what units he wants to make. Finally, there is the ultimate mobility advantage afforded to Zerg by having an unlimited, free supply of transports available at all times—Zerg will always have enough overlords on hand to transport any amount of troops anywhere at anytime. Zerg only reasonably has to defend 2 places on most land maps: his natural entrance and the natural entrance of another main he has taken. However, Protoss has to defend every expansion, because every expansion is almost equally easy for Zerg to threaten. This situation is not equalized unless and until Protoss can produce either Arbiters or 5 or more speedy shuttles—a rare proposition at best when Protoss’ ground troops are already outgunned and, unlike Zerg, Protoss must commit extra resources towards this mobility, while Zerg is going to need overlords for supply anyways. The lesson here is that not only are scouting and countering going to increase your odds, as in Protoss vs Terran, but in fact scouting and countering Zerg very hard is probably your only real chance at winning unless you are just a much better player mechanically. 1.1d - Necessity of Scouting vs ProtossAgainst Protoss it should be elementary that the player who successfully scouts and counters is almost always going to win. When the units are exactly even, only 2 things matter: who out-thinks who and who out-plays who. Protoss vs Protoss is possibly the single most important matchup for this. The counters in the build orders in PvP are harder than in any other matchup. For this reason, Protoss vs Protoss is probably the most random of all matchups. Almost anyone can beat almost anyone in PvP with a lucky set of builds. No matter who you are, if you get DT dropped and you have no detection, you’re going to lose.
1.1e - RecapAll of the previous was really more advanced than I wanted to get into right away, but the first thing that usually happens when people read guides of any sort is that they are skeptical, resistant to change, and will probably not heed many of the lessons in a guide. Most people play the way they do for a reason—sometimes just habit, but also because of subconscious inclinations built up over hundreds or perhaps thousands of games that add up in your mind to a comprehensive set of instincts for how to conduct yourself in a game of Starcraft. Obviously, whatever you are doing and thinking about how to play Starcraft now is the best that you know. When you do anything in Starcraft, you don’t do it because you think you could be doing something else better, you do it because you think it’s the best thing you can do. When you have an idea about Starcraft, it’s probably something you put some thought into based on your own experiences with playing the game, and you probably have some pride and ego attached to that idea. Competing ideas are more than just that, a competing idea; they are also an attack on your ego. Therefore, changing the way you play and the way you think is very difficult. Not only do you have to consciously change your mind about your ideas of playing Starcraft, which is hard enough on it’s own when your ego is on the line, but you also have to change your subconscious habits, built up over hundreds or more games. It takes a great deal of effort to overcome these subconscious inclinations and habits and replace them with new, consciously considered choices. Therefore, whenever I say something like ‘You should always scout and attempt to counter rather than try to seize the initiative’ I will always back it up with solid, logical reasons for why. Otherwise you’re more than likely to either ignore it and go on as you did before or nod your head and then forget about it in an actual game. The key to improving quickly in Starcraft is being able to accept new theories and paradigms where they are superior (and simultaneously reject the inferior) and also replace your subconscious habits with consciously considered, superior paradigms or habits. Theorycrafting aside, here are some solid tips. 1.1f - Initial ScoutingYour first scout should pretty much always be your 7th probe, (8/9 supply), who should, after building a pylon, head to the other mains. If your opponent is Zerg, go the second closest main first, because you will know if he's in the closest main by his scouting overlord coming into your base by the time your probe gets to the second closest main. At which point you can easily redirect your hot keyed probe to his base to scout. If he's Terran, always go to the closest main first. (note that if you are at 9 or 6 on Lost Temple, the closest mains would actually be 12 and 3, respectively) Now if you know for a fact your Terran opponent always walls in, you're in good scouting shape because it means you only have to climb the ramp of a main to know whether he is there or not. That can save you 3 or 4 seconds for every main that you have to scout. Otherwise make sure your probe goes all the way to the enemy mineral line, because rather then wallin, some Terran players build 3 or 4 marines to block their ramp instead. This is less popular then walling in though, because it makes a drop ship build rather worse because the minerals spent on those four marines will be wasted (though they can still be used in a bunker to early push or fast expand). This means that it’s extremely unlikely that you’ll be dropped if he did not wallin, giving you the leeway to safely choose from among build orders that would normally be countered hard by a fast dropship strategy (especially on Lost Temple, JRM, or similar maps with an inaccessible cliff) such as 2 gateway robo into early expansion.
1.1g - Initial Scouting vs TerranAgainst Terran players, tell it to attack an scv that's building something—usually a barracks. Soon enough another scv will come along to chase your probe away. Lead it on its merry way throughout the Terran’s base, never letting the enemy scv touch it at all. As you get better, you can evade this scv and still attack the scv building the barracks, because probes have a bit of range to their attack, and because you can regenerate your shields. If you do succeed in damaging the scv down to the red, keep track of it. If you chase it to the mineral line right away, usually the Terran will bring scvs off of mining to block you and repair the other scv, preventing you from finishing it. So wait a few seconds, then find it again by clicking through his scvs and searching for the hurt one, then go finish it off. You can get to the patch that the hurt scv is mining by right clicking on it, letting your scv pass through his scvs, then attack it and finish it off, then right click your own main minerals to escape clean, passing through any scvs that are trying to block you. Some people think they can kill scvs chasing them with probes by hitting and running, then hitting again after the shields regenerate. While this is technically possible, you shouldn’t count on it unless you have progamer level handspeed. Scvs are far more efficient attackers and almost always get the best of a probe, meaning you will likely be hurt in hp as well as shields before being able to run away again. Also any decent Terran will simply replace his scv with a fresh one if it looks like your probe is actually going to have a chance at killing it. You want that probe to stay alive; staying alive to see what the Terran does is infinitely more valuable than killing an scv. At any rate, most Terrans will probably build a marine or 2 to finish off your probe. Evading marines is more difficult then evading scvs, but definitely possible. Just keep in mind 2 things about their movement: First off, marines have perfect acceleration but slower top speed. So the more you can keep your probe moving in a straight line, the more it will outpace the enemy marine. Do this by running around in the widest possible circles. Also, probes can move through units easily with their mining ai, while marines do not mine minerals. Since the minerals are always at one end of a base it’s possible for the marine to trap your probe there, but by clicking on a mineral at the far end of the patch, your probe effortlessly moves through the mining scvs, while the marine, trying to chase, gets stuck behind the scvs and goes much slower. If he has not built a refinery you can do this trick by clicking the geyser, but once his refinery has started construction, it no longer works. That can buy you enough time to get away easily. Also, occasionally your probe will get stuck behind the minerals, trapped between 2 scvs or an scv and a marine. In that case, just right-click a mineral behind the scv and your probe will pass through him. Anytime your probe gets trapped by a unit anywhere, it’s possible to free it by right clicking on minerals or empty geysers you have vision of. Don’t forget that if your probe gets trapped against a cliff you can free it by right clicking your own main’s minerals. However, workers cannot pass through buildings, so do not run behind minerals if the enemy is making a depot or probe at the other end of them. It's guaranteed that the marine will be able to fire the odd shot at your probe, but if you did not let it engage an scv it will probably have enough hit points to survive for a long time. Long enough to sometimes get some idea of the Terran strategy, or at least slow him down for a while by preventing him from committing to either a command center, starport, or second factory. Beware of the possibility of the Terran showing you something, then canceling after your probe is dead. For example he could make a second factory, showing you an early push build, then cancel it and make a starport and drop you while your forces are waiting at his choke for the early push. You basically just have to get a sense of the Terran’s timing and know that if he hasn't left his base after a certain point to push you, he probably isn't planning an early push at all. 1.1h - Initial Scouting vs ZergAgainst Zerg players, once you find the base, first make sure as to whether he fast expanded or not. A second hatch morphing in the main would indicate not. A hatch morphing at his natural would obviously indicate the inverse. Most Zergs these days do a second hatch in main third hatch in nat strategy, though some do a first hatch expansion and some do a one base lair tech game, or even more rarely a 1 base 2 hatch hydra or three hatch ling build. After your probe has determined Zerg’s opening build order (either pool first, hatch at nat first, or hatch in main first) it's time to harass some drones. Unlike Terran, probes can and occasionally do kill drones, and it is worth your effort to try. The best way to do it is to station a probe at the edge of the mineral patch (either the top, bottom, leftmost or rightmost mineral patch) and wait for a drone to come mine it. The probe automatically attacks the drone. Some Zerg players ignore the probe, giving you a chance to severely damage the drone and maybe later on finish it off. Others counter attack with the drone, but since your probe got the first hit it will win the fight. Beware though, the second it looks like he is bringing in a second drone run away immediately. It may be that he is fighting a fight he knows he will lose just to weaken your probe, only at the last second to attack with several other drones killing your own weakened probe before it can get away while moving his own injured drone to safety. Most often the Zerg will simply redirect his drone to another patch. This is good, because it means you're taking a whole patch away from him and making him play the first few minutes of the game as if his base has only 7 mineral patches, while your own has 8. Eventually his lings will pop out and chase after your probe. Lings are harder to evade then a single marine because they can block your avenues of escape, but if you pay attention you can get your probe to safety or stick around to scout a little longer (which you want to do if he hasn't gotten an extractor yet; you want to know for as long as possible if he's doing without). If you lose that probe (and you probably will) try to select another probe (or 2 if he has a lot of lings) at home and click on the minerals inside the Zerg’s base while your probe is revealing them. Because of the mining ai, that second probe will be able to pass through any enemy units on its way to that patch, but it only works if you click on the minerals inside the Zerg base while they are revealed by the first probe. This will give your second probe a chance to maybe see if the Zerg has expanded with a third hatch (in case it did a second hatch in it's main) and maybe even see whether the Zerg is upgrading to a lair, massing sunkens, massing zerglings, or building a hydra den or evolution chamber (an evo chamber would be a strong indication that he is not going mutas, since you can't upgrade mutas at an evo chamber and you can kill harassing sairs with scourge, there's no need to get one until later). 1.1i - Initial Scouting vs ProtossAgainst a fellow Protoss user scouting is perhaps the most important of all three races. For this reason, I rarely risk my probe at all trying to kill one his probes. Even though it's just as possible as it is against Zerg, the info you glean is just too critical to risk. For that reason I leave it sitting within sight of the enemy's geyser, to see when he will build an assimilator. Once he does, put it within site of his pylon(s), waiting to see what he will build beside it(them). You can't evade a dragoon for long because it's faster and does more damage then an enemy marine, so I try to run home with my probe when the enemy dragoon completes. Sending 3 probes from your main when his goon comes out by clicking on his minerals is often a good idea, as scouting is so important. However, only bother to do this if he has no zealots and only 1 gate, otherwise he will kill your probes before they can see anything unless you send so many that you’re committing suicide anyways. 1.1j - Followup ScoutingAfter these opening scouting procedures there's a lot to be done. First is confirmation, one way or another. You should start your build conservatively countering what your opening scout saw, but never forgetting that the enemy could be deceiving you and changing strats after your scout is gone. Therefore it is necessary to send another scout later on to attempt to confirm your suspicions. This should be done not less then 2 minutes from when your first scout died, or immediately when your first scout is dying in the case of Zerg. This second scout will probably be limited to a view of the enemy's natural, but by partly climbing the ramp you can get an idea of his troop count and makeup (though don't forget he could be hiding other troops out of sight). If it's a Zerg player and he has no sunkens, there is a chance that you can lure his zerglings down the ramp to chase your probe and then sneak up it for a look at his tech, though this is quite unlikely against a good player.
1.1k - Scouting Tips vs ZergA couple good tips for scouting a Zerg player with early ling dominance (for example if he 9 pooled or you did a fast teching build). You don't want to let the lings into your base because that would be a definite disaster. Slow zealots are not equipped to defend against speedlings in the open. But if you want to scout him you would have to move your zealot wall aside to let your probe through. That could risk his lings running up your ramp at the wrong time and getting into your base. To avoid that, take your probe and run it up to the edge of your high ground so that either your natural or mineral only natural minerals are in view of the probe. Then you can simply right click on those minerals and your probe will obligingly run right through your blocking zealots without them having to move at all. Good Zerg players going for a build with early ling dominance will not allow your probe to get to his base easily. With a well positioned overlord or zergling he will know when your probe heads down the ramp and use his zerglings to try to cut it off as best he can. If they have speed, your probe will almost certainly not make it to the enemy base. In that case you have a couple options. First, you can try sending out your zealots to protect your probe part of the way and maybe give it enough cover to get to the enemy base unmolested. But that is extremely risky, because the Zerg player could run past your zealots and into your main (unless you have it blocked by more zealots, which can be a good ploy) or more dangerously, he might have enough lings to simply surround and annihilate your whole zealot army, which is far weaker in the open then it would be holding your ramp. That is a first rate catastrophe to be sure. So a better ploy is to send a second probe a little after the first one. The first one can lead the zerglings on a merry chase while the second, undetected, gets through. While certainly there are Zerg’s clever and aware enough to spot the second probe and attempt to stop it too, he will still be forced to split his forces in two which gives one of your probes better odds of getting through. Whatever you do, don't give up on scouting Zerg. If he is going to a lot of trouble to keep you from scouting, it's probably because he has something to hide. It’s very difficult to win any game where Zerg successfully prevents you from scouting with your probe at all. If Zerg is not containing your scouting efforts with zerglings, you can be sure that he will at least have sunkens at his natural. In this case, you should send out a single zealot to scout. Sunkens kill probes really quickly, but a zealot can take enough hits to get some info. Aside from the obvious, namely seeing a tech building or actual units, either mutas or lurkers, you can get some critical information from how many drones he has. If he has few drones (how many is few? Play 500 games to find out... =p) you know he’s either teching or massing troops. It should probably be obvious if he is massing troops, from his having a lot of troops. Its very dangerous for him to keep massed troops hidden inside his main since if you had been massing troops as well and attacked, he would be at a large tactical disadvantage, so that’s quite unlikely. Therefore, if he has few drones and few troops, the safest assumption is that he is rushing for lair tech, either mutas or lurkers. You should immediately put down some cannons at both your ramp and main minerals, because you’ll need them whether he’s going mutas or lurkers. Ramp cannons will be necessary to both hold against lurkers climbing the ramp and to help you build low ground cannons to secure your expansion against mutas. Cannons at your mineral line will be necessary to hold either mutas or lurker drops. It’s impossible to give a single solid number of how many cannons is right, but typically I get 3 cannons at my mineral line and 2 at my ramp. If you get to Zerg’s base and he has few troops but many drones, he is powering very hard. In this case, you should rush to obtain psi storm, get more gates, and go easy on the cannons. When he hits you, it will be with huge numbers of things, against which cannons matter little but psi storm is ideal. You can adapt more specifically after you see what he hits you with, either way. 1.1l - Midgame ScoutingThroughout the early game you want to have a probe or something sitting in your enemy's natural until he expands there. You want to be aware the minute he expands and the minute he's leaving his base with an attack force. If you lose the probe, quickly look at what's there. Its vision will remain for a second after it’s dead, and that second can tell you a lot. If you see a single drone, probe, or floating cc, it almost certainly means expansion (obviously it always means expansion for a floating cc). If you see a large number of troops that keep moving towards your base without stopping or turning around, send another probe right away to see if he really is attacking. I often like to fake attacks by killing the enemy scout and moving all the way out of my natural before going home and expanding, especially against Zerg, but by the same token the enemy can do that to you just as easily. At some point you're going to have observers. No high level Protoss user goes through to the mid game without them, no matter what the situation. Against enemy Protoss and Terran users your build order is going to call for observers quite early on as a necessity to deal with dark templars and mines. Against Zerg they are not always necessarily needed right away, but no Protoss would seriously consider expanding to their natural without getting observers quite shortly after. The thing is, observers serve 2 roles. Their primary purpose is so that you can survive the enemy mid game threats; dts from Protoss, mines from Terran, and lurkers from Zerg. Their secondary purpose is at least as obvious, but for some reason often ignored. Observers make pretty decent scouts. Once I get an observatory up I usually build non stop observers out of my robotics unless I have a pressing need for a shuttle or reaver. The point is that in no game where you expand to your natural should you go with fewer then 4 total observers. You need at least 2 for your armies (since it's too easy to lose track of one or have one killed -- against Zerg I'd keep at least 3 since a favorite ploy of lurker users is to kill observers in the area preventing the detection of lurkers) and several more to scout. One over top of the enemy's natural to monitor troop movements and defense. 2 more, 1 patrolling each main (you can patrol in such a way as to cover a main, it's natural, and an island with 1 observer on Lost Temple).
Some people wonder about whether upgrading observers is worth it. I'd recommend getting speed against protoss and zerg if they are making use of dts and lurkers past the early midgame. Speedy observers will also be useful to escort a carrier fleet against a terran, in case he uses cloaked wraiths to counter. In addition, though this comes up very rarely, you'd want speedy observers to help against an enemy protoss with arbiters. I also recommend getting observer vision range against terran. He will probably have turrets up all over the place, severely limiting what your observers can see, but with the vision range upgrade you can hover just out of range of the turrets and still see a fair amount past them. Especially useful for hovering over a cliff wall overlooking terran's natural. You should also always get observer vision range vs a zerg lurker contain. This will dramatically reduce your observer's attrition rate, as they will be able to see lurkers without having to expose themselves to too much danger. 1.1m - CounterintelligenceCounterintelligence is another often underrated part of Starcraft (but not among the top players!). Just as your scouts are providing you with invaluable information, so are his scouts providing him. Therefore, killing them is worth your time and effort. When I build my first observer in PvP, my goal with it is always to kill the enemy observer right away. I scout his base with my second observer. In addition that means that if he's dt rushing me my observer will be handy to stop him. Another aspect of counterintelligence is purposefully providing the enemy with mis-information.
For example, let the enemy see you making a robotics, then cancel it and dt rush instead. Because he sees your robotics, he'll play his standard tvp build, giving your DT's a better chance of doing some damage. When playing a Zerg player and you have some kind of secret tech like a stargate, build a dragoon and place it in a spot where it can intercept any overlord that might be heading towards your stargate looking to uncover your scheme. Against Terran there is little you can do about comsats except spreading your different tech buildings around in unlikely spots. But don't spread out buildings in the same tech path! If you’re going for a dt rush, for God’s sake don't put your citadel on opposite ends of your base as your templar archives; put them right beside each other. If he scans either he'll know your game right away, so you might as well not double the chances of either being scanned (by the way, if he does scan you that early, it’s a sure sign he’s going for an M&M rush). Same for if you’re going for a surprise carrier fork, put your fleet beacon right with your stargates, hiding the whole kit and caboodle well away from any minerals and gas that are likely to be scanned in the Terran’s routine sweeps. 1.1n - RecapWhen you go out and start playing again, make sure you practice these things with an eye to improvement. Don't do what you've always been comfortable with--try to do more. Constantly push yourself. If you typically just sit in zerg's base until zerglings come out then run home, instead try to evade them and stick around for as long as possible. If you normally just get in Terran's base and run around, try to harass him, kill some scvs, block his factory or machine shop, wall his scvs in with a pylon, do whatever, and still try to keep up with your normal building at home. Don't be satisfied, don't play it safe, play some practice games and just go wild. Who cares if you lose, that's what practice games are for. Make a smurf name and play pubs if you have to, but don't play to win, play to improve. 1.1o - Examples of harassing Terran with your first probe
Putting a pylon here prevents the terran from completing his wallin. This forces him to construct a bunker or risk losing several scvs to a zealot rush.
The Mannar Pyron, invented by  Smuft, walls in the enemy workers and prevents them from mining. This is effective against either Terran or Protoss on any set of minerals in which it is possible to make an enclosed space. Terran can do the same to you with Depots, though it's much harder because Depots are larger and more difficult to place. If you successfully trap workers, you will come out ahead in minerals, though by how much depends on how badly the enemy panics. If he attacks the structure with all his workers you can cancel it at the last second and come out way ahead. If he just uses a few, you'll come out maybe 50 or 60 minerals. If he attacks it with only the trapped SCVs he'll be in better shape, and if he uses the mining/stacking trip to force extra workers free (by sending them all to mine in the same spot, then pressing S, they will revert from mining ai to movement ai, causing them to try to spread out; some of them will spread out overtop of the minerals/pylon and escape) until there's only 1 trapped, you'll barely break even but it will still be worth it for wasting his time with a delicate micro maneuvre.
This is extremely effective if Terran declined to make any marines, but very difficult, if still possible, if he elected to produce a marine or two. First make a pylon where his machine shop would go as his factory is under construction. Couple this with a hard ranged goon rush. He will be forced to either lift the factory or produce vultures only. If he lifts the factory, use your probe to block it from landing for as long as possible. Your goons should be arriving shortly, while his machine shop will not even be under construction yet. You will get a long time to bang on his wall and will most likely score several scv kills, making the cancelled pylon pay rich dividends. This is one reason almost all terrans build at least 1 marine and often more. Make him pay if he does not! ***Author's Note:*** Big thanks go out to Raistlin, RayOfLight, and Waxangel for help with the pictures
| ||||||||||||||||||||||||||||||||
|
| ||||||||||||||||||||||||||||||||
|
|
|
|
| |
