Get updates via e-mail:

« Stanford Seminar on Gaming | Main | Malcolm Gladwell and Playing to Win »

UC Berkeley StarCraft Class, Week 14 (Final Week) 

In the last class of the semester for UC Berkeley's StarCraft class, several students gave their presentations.

Zergling Rush

The first student's topic was the Zergling Rush. The point of this strategy is to take advantage of Zerg's window of dominance at the very beginning of the game. Zerglings are weak, but you can get them out so early that your opponents might very well have little or nothing to defend with.

There are two main varieties of the zergling rush: the 4-pool and the 6-pool. 4-pool means that you only have four drones (resource collectors) when you build your spawning pool (the building that lets you produce zerglings). So you start the game by gathering resources for a bit, not building any drones, and going straight to spawning pool and zerglings. This is an "all in" strategy though. If you don't win with it, you are so severely behind economically, that you will probably lose. The reason to choose this strategy over a slightly more conservative 6-pool rush is if you think getting the zerglings to the opponent a few seconds faster will allow you to win the game on the spot.

The 6-pool strategy is still risky (putting you behind economically if you fail), but you can recover from it if things don't go as planned. The build order is to first build two drones (getting you to six total), then mine a bit, then build a spawning pool as soon as you can, then build another drone to replace the one that you just turned into a spawning pool, then build an overlord. If you do this correctly, these things will happen simultaneously: 1) your overlord finishes building, 2) you have all three larva ready to build units, 3) you have 150 minerals, the exact cost of building 6 zerglings from your 3 larva.

The student said the first goal of a zergling rush is to end the game immediately if you can. As soon as you arrive at their base, you probably have a good idea if this is even possible. If it's not, your next goal is to disrupt their economy. Remember that your economy is already disrupted by you doing a zergling rush (rather than building up your own economy) so you'll have to disrupt theirs just to stay even. If you can't disrupt their economy much, the next priority would be to at least force them to change their strategy or force them into a certain strategy. For example, they might have to build sunken colonies or bunkers right away to survive. And failing all that, at the very least you should harass with your zerglings and force the enemy to spend a lot of clicks to deal with you. The student also pointed out that there is some value in getting the opponent angry or flustered here, too. Zergling rushes are considered "cheesy" by some, so that's great to get those players mad. But even more enlightened players who don't care about that can still be thrown off their game having to deal with pressure that comes that early.

So the advantages of doing a zergling rush have to do with speed and pressure. But the disadvantages are the sacrifice to your own economy, and the ease of the opponent countering you if he scouts you or predicts this strategy. It's high risk, but high reward.

The main things to consider when contemplating a zergling rush are the size of the map (bigger map means more travel time, reducing the effectiveness), the number and placement of starting positions on the map, and whether you get scouted.

The class administrators pointed out that on a map with only two starting positions, it is viable to try a zergling rush against another Zerg player, but it's a bad idea against a Protoss player. At first glance you might think it's smart against Protoss because they are the mostsusceptibleto losing if you catch them with early zerglings, but you just won't "catch" them on a map with only two starting positions. Zerg players usually use an overlord to scout, which is slow and gives you some chance of surprising the enemy, but a Protoss player will pretty much always scout with a probe (faster than overlord), discover your intention to rush, then counter it.

The student showed a few examples of zergling rushes succeeding and failing, and we saw how important it was on a map with 4 starting positions to make an educated guess about where your opponent's base is, if you don't get lucky and discover it with your first overlord. You can make this guess based on the timing and direction of travel of your opponent's scouting unit, if you happen to cross paths with it.

Jaedong vs. Fantasy

The next student's presentation was about a series of games between legendary players Jaedong and Fantasy. Jaedong is sometimes called the "elite zerg" and is considered the best Zerg player in the world, and possibly the best player in the world, period. (Remember, I'm reporting what was said, not giving my own opinion.) He's also known for his mutlaisk micromanagement skills.

Fantasy is a relative newcomer, but he's dominated lately, including beating Bisu. He's known for his "Fantasy build" which is also the butt of many jokes. It involves using a dropship with vultures for early harass, and incredible vulture micromanagement. He's also known for his anti-air tactics, using both goliaths and even valkyries.

The first two games surprised many spectators because Fantasy destroyed the great Jaedong. Game three was a turning point though, and Jaedong changed his strategy. He pressures hard with mutalisks and follows that up with even harder mutalisk pressure. He then mixes it up with more mutalisk pressure after that. There's like 20 straight minutes of mutalisks killing marines+medics. At one point, Jaedong has TWO packs of mutalisks stacked (using the bug) on top of each other, so you can't even tell how many are there at first glance. This is enough that he can one-shot-kill Terran towers! Fantasy has five towers right next to each other and Jaedong destroys all of them by flying in, one shot kill one, fly back (waiting for cooldown on the shot), flying back in, one shot the next one, repeat. You don't see that every day, and even Professor Feng said "what??"

The most talked-about moment in the series was when Jaedong somehow perfectly thwarted a crazy trick. Fantasy sent one SCV and one vulture to Jaedong's back ramp that starts off blocked by a single mineral patch. He then used a difficult-to-execute trick (easier if he had more SCVs, but he only had one here) to cause the vulture and SCV to pass through the minerals. This allowed him to slip past Jaedong's defenses in the front and potentially wreak havoc. Somehow, the moment he went up that ramp, Jaedong had a couple zerglings and a drone right there. This was the perfect positioning at the perfect time, and looked incredibly random that his units would be there right then. He had no map-vision of this area, so he had no apparent way of knowing about this. Jaedong used the drone to block, and the zerglings killed the vulture immediately, before it could even do anything.

Some StarCraft community members think that this was purely luck and had to do with a pathing algorithms on that map. They think Jaedong sent those units somewhere else on the map, maybe his expansion, and the bad pathing happens to make them run the long way to the place where he encountered the vulture. Professor Feng did not believe this explanation though. He said it's plausible for the zerglings but it would not explain why the drone was there, but its pathing would certainly not have caused it to be there. Lore, an assistant to Feng, didn't believe that explanation either. He said that when Jaedong, best player in the world does something, we should probably give him the benefit of the doubt that he meant it. I suggested that maybe the most likely explanation is that he simply predicted this, that it's a known trick at that exact place on that exact map. Lore said yes, but it's still a bit far fetched for it to be pure prediction at that exact moment. He said that a few seconds earlier, it looks like one of Jaedong's zerglings just barely had vision of the vulture or SCV on its way there and if so, that's what triggered his timely defense

Anyway, Jaedong triumphed. Here's the first match:

Strategy to Fit the Player

The next student's presentation was about tailoring your strategy to yourself and your opponent. As Sun Tzu said:

It is said that if you know your enemies and know yourself, you will not be imperiled in a hundred battles; if you do not know your enemies but do know yourself, you will win one and lose one; if you do not know your enemies or yourself, you will be imperiled in every single battle.

He emphasized that you must know your own strengths and try to play in a way that takes advantage of them. For example, if you are good at micromanagement, use a strategy that involves many quick battles so you can use that skill a lot. You must also know your weaknesses. For example if you are bad at macromanagement, then you don't want the game to go long. A longer game means macromanagement becomes more important, so you'd want to try for strategies that end the game sooner. And of course you must know the opponent. If you can predict what he will do, your job is much easier. Best of all is if you can trick him into doing things he shouldn't.

He showed two replays of his own games. This first was on the map Destination and he said "all Terrans go mech on Destination." He means all Terrans go for a mech build with some combinations of vultures, tanks, and goliaths. He played Zerg against a Terran opponent and said he expected a mech build from the start. The Terran sent early vultures to harass and this (along with other clues) confirmed that yes, the opponent is doing a standard mech build. The early vulture harass did damage the student's economy, but not too badly.

The student's strategy was to research the burrow ability as soon as possible, emphasize building an army over getting expansions, then burrow most of his army in the middle of the map while leaving the rest of his army out in the open as a decoy. As the Terran moves his large mech force toward the decoy units, the Zerg player suddenly unburrowed and devastated the entire Terran force. Most of the students laughed out loud at how perfectly executed this was. The Terran's tanks were caught unsieged and far apart. The goliaths were also strewn about, not at all prepared for a battle at that moment. Meanwhile the Zerg units were bunched together, able to attack with high flux (meaning they could all attack simultaneously, right away). This attack was so devastating that the Terran couldn't recover, even though he had more expansions.

The next replay demonstrated what the student presented as a new build order he devised. It's a Terran build order only for use on the map Medusa. The build emphasizes the student's strengths of micromanagement and deception, hopefully avoiding his weaknesses of poor macromanagement skills.

The build starts with a barracks and fast gas. On purpose, it looks like a standard mech build at the point in time that the opponent would usually first scout it. That's what we assume his opponent though in the replay, as the opponent (Zerg) went for zerglings. But just after that, the student builds an academy (not a mech build). He uses this to go medics + firebats, which is a pretty unusual thing to do. This is crazily powerful against zerglings, so much so that it forced his opponent to build three sunken colonies at the choke point of his main base. That's exactly what the student hoped would happen.

The student sent his firebats + medics toward the opponent's base, but this is just deception. It's designed to get the Zerg player worried and make him waste resources on static defenses. The real play here is going to the opponent's back door. On the map Medusa, the main bases have a back door that's blocked by a neutral building (a temple) that apparently has lot of hit points. Firebats are well-suited to break down this barrier very early in the game, and that's exactly what he does. Remember, the opponent doesn't get a warning of this neutral building being attacked, so if he's not specifically scouting right there, he has no way of knowing this is even happening. The student did break through the barrier, he did go in the back door, and he did completely rock the Zerg player's main base. Those sunken colonies the Zerg player build at the choke point were completely useless. The student said he has used this same strategy a lot on this map, and it's been very successful overall.

There were other presentations, but because too many students waited until this last day to present, they went over the usual time allotted for the class. I was not able to stay through all of them, so that's all I have for you.

Thanks to Professor Alan "Nimue" Feng for doing all the work of setting up and teaching this class. I hear that it's been approved again for next semester and that next time it will cover a mix of StarCraft and Street Fighter to show the interesting commonalities of strategy in different genres. Ok I didn't so much hear that as make it up, but the part about the class being approved next semester is true. I hope you enjoyed this extended coverage of the class.

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.
Comment in the forums
You can post about this article at