[OSL] The Usual Suspects

Game One started on Great Barrier Reef. 815 got orange zerg at 4 and sent his overlord towards 12, the wrong direction as Really was in purple at 8. 815 sent out an early scouting drone to the other base site to check for a forwards 8 rax and then opted for a 12 hatch when he scouted it was clear. Then in a very strange choice he sent another drone out to Really's base, skirting it around the scout scv's vision to meet its advance counterpart in the natural. Once there it was drilled across, unknown to Really who just opened with a standard 1 rax FE build.

815 made a quick third hatchery behind the natural of Really: not at the expansion where it would be the last place he'd look but rather close to the natural mineral line with no purpose but offense. Really clearly recognized that 815 had taken a hidden 3rd hatchery as the first-person view showed him sending scvs to look for it but it's fairly standard to drill to an island so Really wasn't worried or surprised by the missing 300 minerals. The creep quickly betrayed the hatchery and Really immediately drilled a marine across in an excellent micro trick but unfortunately there were 2 zerglings waiting and Really's other marines were not supporting his first from the far side of the minerals. The marine died and a creep colony quickly started. Another marine was drilled across but it was too late to stop the sunkens. At this point Really fell apart. 815 had 700 minerals invested in stopping a FE from mining but the hatchery couldn't send any units to the rest of the map and was poorly placed for mining. Really could have easily flown his command center to his mineral only third under the main cliff to get an advantage.

Instead, Really attempted to mine through so he could rush mnm through a tiny choke into sunken colonies. His scvs were cut down as he tried to take the last 4 minerals of the crystal in his path, easily covering the cost of the hatchery and sunkens in losses and disrupted mining. Although he did succeed in breaking the sunkens and got 3 free overlord kills which popped as the sunkens broke, his timings were completely thrown and 815's spire was out. Really then went on to humiliate himself further by suiciding his mnm into the sunkens at 815's natural and target firing a creep colony. Let's just look at that decision for a second. Sunken colony, 300hp. Creep colony, 400hp. Sunken colony, 40 explosive damage. Creep colony, no damage. Sunken colony, problem now. Creep colony, only a problem if you still have units there when it's finished. Marines will not auto-target a morphing sunken if there are sunkens there. It was a grave and embarrassing mistake.

The mutalisks delayed the engineering bay for a while and more surprisingly, in all the confusion Really had forgotten to kill the 3rd hatchery behind his natural. 815's mutalisk micro was somewhere between underwhelming and just plain bad but with good positioning against unsupported marines it did the job and got a decent number of kills. Unwilling to accept such an easy advantage 815 continued to pump drones and overlords out of his 3rd hatchery which promptly ran into Really's natural on their way to a rally point intended for mutalisks. The value of harass is somewhat negated if you insist on making your non offensive units run through his base.

815 finally lost his 3rd hatchery and made a new one at 12. 815's sloppy play put him ahead only because Really had no scvs. 815 was able to trade mutalisks for marines while pumping drones from expansions (while only suiciding half of them) while Really did not have the same flexibility. Really made an attempt to attack 12 and again went for the creep colony, making the same micro mistake he made before. The attack failed and then he decided to show us what it looks like when a Terran from original StarCraft plays a Brood War Zerg. Pure marines against lurkers. It went rather badly for Really. He lost all of his units and tapped out.

Really spawned in teal at 11 on Fighting Spirit against 815 in blue at 5. 15 opted for a rather unorthodox build: 12 hatchery, 11 pool, 10 gas. Really did another standard rax expand opener. 815 made 6 zerglings to chase Really's initial scouting scv. With the scv trapped in the main 815 made a quick hydralisk den at his natural to coincide with the completion of lair and Really didn't scout, even considering his fast expansion. Really didn't drill a second scv to the minerals in 815's base as the first died, despite the fact that with 60hp a scv is almost guaranteed to make it in if there are no sunken colonies in the way. In the mean time 815 has made a dozen speedlings and hidden them at 12 while trying to lure marines out onto the bridges with his initial 6. Unfortunately he and has to back off and seconds later his rushed lurker eggs are scanned.

The lurkers and the marines fight as if both are controlled by computers, the lurkers moving into their maximum range and firing and the marines standing completely still until the lurkers hit them and then AI retreating a bit. Still, despite setting up a great speedling flank 815 cannot engage because Really's firebats were too dangerous. Really slips a group of mnm out into the middle and 815 promptly takes half his lurkers and creates a hold lurker trap for them. This doesn't work out so well for him when the mnm turn round and together with mnm crossing the bridge from Really's natural, surround and slaughter his lurker contain. However Really's victory is short lived as he takes all his victorious forces and suicides them into lurkers for absolutely no gain. This game is absolutely tragic so far.

815 is going going 2 gas hive with fast defilers which wasn't a bad decision considering 815 is only one good swarm away from being inside Really's base. Really tries to slip some forces out through 12 to contain an expansion 815 hadn't yet taken but ended up running into cracklings with predictable results. Swarm arrived and despite Really's minefield (which was a good idea and would have been great if he hadn't spread them such that every single lurker took exactly one mine hit) the lurkers simply walked into the natural. The marines could only watch as the natural was overrun, and then again only watch as a swarm was cast over them and some zerglings tore them apart, another example of Really's refusal to micro. Really tapped out as his expansion was torn apart. 815 wins the series 2-0.

Kwanro started in blue at 5 on Great Barrier Reef while 815 got red at 12. 815 scouted the wrong way while Kwanro scouted correctly. 815 opted for a rather unorthodox build: overlord, gas, drone, pool. Kwanro simply went for a 12 pool gas which is about the safest economic opening in zvz. After scouting no 9 pool from his opponent he followed it up with a fast hatchery at the expansion, showing he had a clear plan to play 2 hatch, while 815 went 1 hatch with a weaker economy but a significantly faster lair. This being ZvZ, both players proceeded to go pure zergling, 815 leading and forcing Kwanro to match his count or be overrun. However despite his faster zergling speed 815 could not do any damage because Kwanro made exactly the number he needed and held the timing attack which hit the moment the hatchery completed (the weakest point). From then onwards 815 was on the defensive as Kwanro had double his production capacity. This allowed Kwanro to pressure 815's larva, threatening with zerglings to force 815 to make zerglings himself and therefore not save larva for mutalisks. As Kwanro had twice the larva income this trade could only work out well for him.

Although they had a similar economy 815's money was slowly piling up because he had no second hatchery to spend it. Despite 815 having an excellent concave at the top of the ramp Kwanro simply suicided zerglings up it, trading them inefficiently secure in the knowledge that he could reinforce twice as much as his opponent and any reinforcement by his opponent would serve to delay mutalisks despite 815's faster spire. Eventually 815 got 2 mutalisks out but could not do anything with them because of stray speedlings wandering around his base. By the time he had cleaned up the speedlings Kwanro's mutalisk count already rivaled his own and the difference was growing. They fought a mutalisk battle which Kwanro won and that was that. GG.

Kwanro's opening was designed to maximize mineral income in the first 90 seconds or so purely for the second hatchery. He never had any intention to mine 2 bases because anything beyond 8 drones mining is pretty much a waste in ZvZ. He then fought with production, attempting to trade at an even rate whenever possible to create losses 815 could not replace. It wasn't really a fast expansion, although placing it at the expansion obviously created options for him had the game turned out differently, it was an early investment in unit production. While 815 opted for a speedling rush, Kwanro had the advantage in production which won him the game.

Kwanro got purple at 11 on Fighting Spirit while 815 got red at 1. Both of them scouted incorrectly, Kwanro sending his overlord cross positions (so slightly more correctly than 815 who sent his to 5). 815 did that overlord extractor drone pool thing again which gives you fast lair and zergling speed but also loses the game. Kwanro just opted for a 12 hatch because it's a big map and 815 probably isn't 9 pooling. Besides, 12 hatch isn't necessarily a death sentence against 9 pool anymore. 815 made some lings and tried to attack but as a result of his weird build, they weren't fast enough to get anything accomplished. 815 was able to threaten the natural for a while but with 4 zerglings hatching for every 2 of his, there was only so much zergling speed could do.

Kwanro got a relatively late spire compared to 815's rushed spire but, as before, was able to place tremendous pressure on 815's larvae, despite a concave on the ramp. Kwanro just suicided them up and hit the timing window where 3 larva were saved up for mutalisks but not hatched yet. He broke up and 815 realised he couldn't wait and used the larva for lings but still couldn't get them in time to stop drone losses. By the time he finally contained the situation with a sunken 815 was down to 6 drones and Kwanro had spore colonies and 15 drones. Then they both made some units. Kwanro made more. GG, Kwanro wins 2-0.

According to Midian, 815's overgas into pool build is like a overpool gas standard spire rush build but with faster zergling speed and slower zerglings. He didn't think much of it either because the lings are too late to have any rush value and the economy is too weak to give a second hat. The only thing it has going for it is fast zergling speed but that micro advantage isn't as good as the numbers advantage you give your opponent by late zerglings allowing a 2 hatch.

GuemChi got teal at 4 while Killer got green at 12. Guemchi scouts correctly while Killer goes for overpool and builds up larva for lings/a fast expansion. Guemchi is a baller though and blocks the expansion, killing the drone with 3hp left on his probe and by this point the lings are too late to stop a 13 nexus after forge. Guemchi cuts probes to get a fast gateway to seal the wall up and secure it with probes so his expansion is actually faster than Killer's. His 1 kill probe lives on in Killer's base and sees the gas go down as well as the 3rd hat at the mineral only below Killer's main cliff. Killer responds by walling his ramp with zerglings and making a fast hydralisk den but Guemchi isn't bothered in the least.

Guemchi sends out a zealot to force some additional zergling production but then retreats it immediately. He gets +1 air attack and a fast stargate as core completes and adds a 3rd cannon for a bit of safety. However the 3 hat hydra is unscouted and can be very dangerous against only one gateway and no fast tech. The hydralisks attack into 2 cannons with a 3rd on about 80%, 3 slow zealots and 1 dragoon. However despite being an instant zerg victory on paper Guemchi's response is better than perfect. Guemchi immediately drills almost all his probes up into the hydralisks while a remaining probe lays down 3 more cannons. The probes drill into the hydralisks time and time again, prolonging the life of the cannons significantly and doing a lot of damage in their own right. As the dust settles very few probes have been killed and the probes are actually microing down hydralisks. Killer tries to break in one last time, despite the defences now being considerably better than they were the first time he failed. The probes instantly drill again (Perhaps Guemchi has permanent vision of something and is just doing the f keys probe evacuation trick) and the hydralisks are massacred.

Any credit Killer had is taken away at this point due to his failure at pattern recognition. Guemchi just gets bored of this and to ensure his safety while he makes new gateways and waits for ht energy he simply makes 9 cannons at his front whereupon Killer concedes and starts pumping drones and heading for lair. Having stabilized with very few probe losses Guemchi starts storing up high templar energy and goes to 6 gateways while Killer adds two more hatcheries, one at the natural at 9.

A storm claims 3 hydralisks on hold position containing outside Guemchi's natural but Killer continues to contain, strengthening it with a few lurkers to go with the hydralisks and overlord. Killer gets a bit optimistic, adding a 6th hat at his mineral only and a 7th outside Guemchi's natural to go for a permanent contain. But Guemchi's first observer comes out and he presses t all over Killer's army. Killer manages to trade all his hydralisks for an observer kill which allows 2 lurkers to hold the line for about 10 more seconds before the next one arrives. Guemchi's army quickly finishes off the contain and pushes into Killer's natural, killing the sunkens and many drones before retreating.

Killer is left trying to limit the damage on several fronts as pressure mounts against 9 and his mineral only. At the same time Guemchi takes his back mineral only at 6 and a dt scores 10 drone kills at 9. With 9 shut down Guemchi has only one front to pressure and starts throwing storms at the mineral only, losing the battle but applying more pressure than Killer can endure. Another dt there scores several more drone kills and the constant aggression gives Killer no time to recover from the harassment. Guemchi finishes off 9 while taking his mineral only under his cliff and Killer attempts to trade bases. Storms punish his army and Guemchi's main force returns in time to clean up and end the game. GG.

One of the best defences of a 3 hat hydra in recent memory. After that he simply played safe and never allowed Killer to stabilize the situation. The dt's killing drones were a bonus but that was a consequence of his aggressive multitasking, not a lucky break. Extremely well played by Guemchi. Not so much flair but he did the thing he needed to do to win and he did it well.

Guemchi started at 11 in brown on Fighting Spirit vs Killer in orange at 1. Killer opted for an overpool into expansion again and Guemchi first scouted and blocked it again, this time with a pylon. Unlike game 1 Killer did not wait until after the hatchery to build lings, realising on about 400 minerals that he was wasting time. This game he went lings then hatchery which isn't ideal for the zerg but is better than nothing. With close positions and the minerals lost due to the pylon (cancelled) Guemchi decided not to risk the timings with a 13 nex. That's fair enough because although 1 cannon is definitely done in time for overpool at 11 on Fighting Spirit you need to pull 5 or so probes to block safely and it's just more expensive than it's worth to cut it fine when you don't need to.

Killer's first 6 lings killed the scout probe while he went pure drone. Guemchi immediately sent out another scout probe but got it trapped and killed immediately. However the third safely made it up to 12 and took the long route in undetected, scouting the 3rd hatchery at Killer's natural and almost making it into the main. A look at the drone counts calmed any fears he had and in this game Guemchi stayed on 2 cannons, despite zergling speed. Killer went for a 4th hatchery and a dozen speedlings after lair before adding a 5th. Meanwhile Guemchi walled in with zealots and a pylon and made a fast robo. However Guemchi sloppily lost his first corsair (which can be huge depending on builds) to scourge trying to kill a second overlord.

Guemchi massed corsairs and teched to +1 so losing that corsair would definitely mess with his timing. You cannot risk a shuttle without a critical mass of corsairs simply because the cost of losing it is too great. Killer scouted the reaver and created a net of zerglings around Guemchi's base to scout the leaving shuttle, burrowing them to remain hidden. Despite knowing the reaver was coming he still managed to lose 3 drones at his natural, burrow proving no defence against splash damage. In a display of really shortsighted decision making Killer burrowed some hydralisks at his natural mineral line so when the reaver came back they could ambush it.

The reaver dropped and the hydralisks popped up.... in the middle of the drone line. The reaver promptly fired and the 2 dead hydralisks took 3 drones with them before the reaver was evaced by a shuttle. Guemchi continued to produce significant numbers of zealots and attempted to take a 3rd at 12, similar to Stork's build last week. However unlike Stork's build last week Guemchi's shuttle micro was lacking and he managed to lose his shuttle a second after picking up a reaver to protect it which promptly left the other reaver vulnerable and shortly dead. A third reaver made the long walk from the natural to 12 and secured it but without the shuttle to quickly redeploy reavers as Stork had done his defence was weak. The hydralisks simply pulled back from 12 and went straight to the natural, catching Guemchi off guard. The high templar Guemchi had just produced had no energy and the reaver which followed had no scarabs. With his front broken Guemchi lost all his probes and his robo and GGed out.

What Guemchi was attempting is what Stork made look really easy last week. To hold 3 bases against a superior force by redeploying reavers at lightening fast speed while doing reaver shuttle micro and securing air control. Stork could do it. Guemchi couldn't.

GuemChi got purple at 11 on Eye of the Storm while Killer got orange at 5. Neither player opened with anything remotely interesting, Guemchi double scouting as standard while Killer went for 12/11 with a quick third at 3 o'clock. Guemchi expanded and teched fast with a single late cannon appropriate for the level of threat Killer offered. Killer headed for a fast lair while Guemchi skillfully microed a probe into his main and disrupted mining at 3 with a zealot, paying for itself in delays and attention. Killer did a very standard 3 hatch spire 5 hatch hydra den with normal timings, losing 2 overlords but powerdroning heavily and getting a corsair kill. Meanwhile Guemchi did not continue corsair production beyond the safe anti-muta 2, instead pushing out to hit 3 at a weird timing just before sunken colony, catching Killer's superior army out of position. Despite the fact Killer had the units and cleaned it up the attack did more damage than it cost.

Guemchi had evidently resumed corsair production and attempted to do some dt corsair harass at Killer's third with 3 corsairs but overlord speed blocked him doing any real damage. A group of corsairs proceeded to harass overlords while Killer's army harassed Guemchi's natural with quick lurkers. Zealots and archons died for very few kills but there were so many overlord kills that Guemchi was probably even in the exchanged and seconds later he was able to break out of the lurker contain by killing the overlords and using dts to attack the lurkers. Unfortunately for him reinforcements arrived and he was quickly contained again as more lurkers morphed. Despite having the units to break out of the contain at any point before the lurkers finished Guemchi waited until the last moment (probably waiting for ht energy but his zealots without storm would do better than against lurkers with storm). He attempted to push out but going zealot heavy against lurkers never works all that well and he lost his army.

Attempts at drop play and a hidden expansion by Guemchi were shut down leaving him resorting to more zealot against lurker play with the same degree of success. Killer refused to overstretch, staying on lair and just trying to hold the contain until Guemchi mined out. Guemchi for his part abandoned the futile zealot breaks and opted for a reaver siege with dragoon support which was shut down by considerable hydraliks casualties. Another attempt to push out with dragoons was blocked by killing the observer as the robotics facility was now divided between observer and reaver production.

Killer continued to trade hydralisks for a contain but despite heavy losses to reavers and storm his relative psi grew through the battles. After one last attempt to push out was squashed Guemchi GGed.

Guemchi overstretched his gas. He attempted to go corsair heavy, dt heavy while getting a fast robo and archons. He was ill suited to deal with an established lurker contain. He could have easily avoided the contain if he'd pushed out before it was fully set up. But he missed the window and Killer simply hard countered what he was doing. Oddly enough the halt in corsair production and the first lost corsair may have been critical. His first corsair dt play had 3 corsairs whereas the critical mass to just attack move scourge is 5. A good Protoss knows the timing for scourge relative to corsairs, if they plan on using the corsair later they don't risk it for a pointless kill.

Newer progaming fans won't remember this but there was a time at Savior's height when Hwasin beat him TvZ and everyone was like "wow, this guy is sick" and then he got into really tough draws over and over and never got anywhere despite being good. Hwasin never really got the credit he deserved for his skill. Anyway, Hwasin got blue at 12 on Great Barrier Reef while Killer got teal at 8. Killer went for 12/12 while Hwasin went rax FE with cc and a second rax before any marines.

Killer drilled a drone across his nat for a quick mineral only third which is very unusual for ZvT on here where 2 hat play is more common and taking a 3rd island is normally a hidden gas. Hwasin stopped rax production all the time to get tech up which was ballsy if done right, opting for production facilities and tech over actual units for the flexibility. He needs to spend the minerals on that stuff at some point anyway so he's not behind by switching the order, it's just a different route to the same place.

However unfortunately for him Killer got zergling speed, made a load of speedlings and killed him. He did it blind which was kind of amusing but it doesn't matter, Hwasin was pushing his luck and got punished for it and that's probably fair. A blind counter every now and then keeps a Terran honest. GG.
Hwasin was relying on the fact that everyone knows you can't simply attack move into a camping FE Terran early game. It was ambitious and would have been great if it had worked.

Game 2 on Fighting Spirit. Killer got red at 11 while Hwasin got blue at 7. Killer went for 12/12 again(like 12/11 but with an extra drone) and scouted Hwasin first with a drone. Hwasin for his part opted for rax FE but this time with marines rather than just assumptions about Terran camping ability. Probably wise given the previous game. Killer made a 3rd hatchery in base on 14 control and followed it up with gas in a heavy macro build.

Unwilling to be unambitious Hwasin sent out several of his first marines overlord hunting blind but fortunately was unscouted and got them back safely. Meanwhile he made 3 rax quickly with an academy and kept them pumping, the hard counter to what Killer did in game 1. Killer got lair and dual teched to spire and hydralisk den, unaware of the sunken break on its way. A screen of a sensible marine count went out ahead of the main force, concealing Hwasin's numbers and although Killer did have a speedling flank ready for the sunken break they were cut down by firebats. 2 sunkens stood no chance against the mnm force and the drones were slaughtered.
A mutalisk allin counterattack had reasonable success but given his base was destroyed it was only to amuse the audience. Killer GGed.

A classic sunken break, reasonably well executed and well concealed.

End of the night on Eye of the Storm. Killer got blue at 11 while Hwasin got red at 5. With one game to redeem a depressing night, it was in Hwasin's hands now. Killer opened 9 pool no speed, a build only useful against a 14 cc which is very viable on Eye of the Storm so you can see where he's coming from there. Even with rax cc players often cut marines for a faster cc so maybe. Anyway, Hwasin scouted it and despite only having 2 marines was able to wall with scvs and hold the front while building a cc in his main to fly out. At the same time Killer took his natural and made a 3rd hat in his main while pumping drones.

Killer killed the scout scv and made a dozen zerglings but Hwasin was taking no chances this time, opting for an old fashioned defensive build with fast ebay, fast academy, constant marine production with a bunker. There have been metagame shifts inspired largely by Flash towards aggressive 2-3 rax play against lair in order to keep the mutalisks away through aggression rather than defeating them through a fast +1. However with Killer being an unknown factor with 3 hatcheries I can't fault Hwasin's defensive play.

Killer opted for some strange twist on a lurker rush, taking just 1 gas but building no spire. It was quickly scanned but the 3 hatch and 1 gas aspect meant it was very flexible for other transitions. At the same time he expanded to 7 but the 2 hydralisks he sent towards 7 to lurker were caught morphing in the open and killed. The presence of the lurkers revealed an expansion and the lack of them left it vulnerable. 7 was isolated from reinforcement and killed while attempts to send lurkers just threw units away.

Having kept his opponent on 2 bases Hwasin made no attempt to expand himself, instead upping his production in line with his economy and planning to just contain expansions and wait for science vessels to defeat a gas starved zerg. Killer quickly teched to hive and made another expansion attempt, this time at the main at 1. Lurkers made it in time to defend this time, cutting down a few lost marines sent to harass it. However when Hwasin's main force arrived with tanks and firebats the lurkers and speedlings were killed to a zerg with almost no losses. That left the way open and despite a nydus (which finished on 26hp), the base was destroyed. Hwasin took no chances against hive, teching to spider mines and going 2 starport vessels. With only 2 gas this severely limited the zergs options giving him no gas for scourge or ultralisks or to replace fallen defilers.

Another zerg expansion attempt at 9 was scouted and destroyed with Hwasin having uncontensted map control. Only when he walked his army directly into lurkers under dark swarm and then tried to retreat his army into the lurkers did he actually suffer losses and even they were insignificant, leaving him 120 to 50 supply ahead. Bored of this game Hwasin just walked into Killer's natural and killed everything forcing Killer to GG.

Catching those first 2 lurkers was fortunate for Hwasin and won him the game. Seeing them told him Killer was trying to take an expansion in the area and killing them left the expansion undefended. Killer could never get a 3rd gas going and with only 2 gas it just wasn't going to happen for him.