Rush v Macro

To rush, or to macro, that is the question. Whether victory will come to the first who is bold enough to attack. Or if she would prefer the patient player who maintains his defensible position?

When you consider that the current map pool is horrible, the maps all tiny, I think its clear that she prefers, at the moment, the rush.

There are some maps, Metalopolis, that are so rush favored, you really can’t play them any other way. As a random player, this can be confusing, simply because there are so many different combinations.

I always like to start with the most volatile and rush friendly matchup. The ZvZ. One of the reasons why I think so many people have trouble assessing a ZvZ is that there are both one base and two base rushes. But after playing many games, I’m happy to report that on smaller, two player maps, the one base rush is superior.

You won’t find many one base rush builds on imbabuilds.com and I think that’s a sad thing. Because it implies that nobody plays the one base rush anymore, and that simply isn’t true. I mean, even Jaedong played one base rushes in a professional tournament. I mean, he lost, but still he played it, and played it well.

But that also ties into the other factor that makes me say you almost have to one base rush. That’s skills. I understand that on the pro level, one base plays kinda suck. But they sure don’t at the platinum level. Because you have to be pro level gosu to be able to defend a one base rush when you open hatch first. I’ve seen players like Taeja do it. But it never works out for me.

Everytime you try and copy something that’s beyond you, you wind up falling flat on your face.

Let’s switch to Terran. Now, on any map, in nearly any game, I have no trouble playing a reaper expand with confidence. But these new map pools, has me going gas first into a 1-1-1 tank push (for TvT). It just wins games, it just feels right.

Protoss is another race where the conventional logic has always been that the first to expand will lose. Which means you can expect to see a lot more 4 and 5 gate attacks. I love to threaten a 5 gate, and if everything isn’t perfectly in my favor, add on a Robo Bay and turn it into a Warp Prism drop, while having an observer for DTs.

Of course, when you plays as protoss, Stargate plays is almost mandatory. The potential of the oracle to flat out win games is so good, you have to roll the dice on that attack. Even if it doesn't work, you can hold onto the oracle for scouting and protection against cloaked units.

I played a PvP the other day where my opponent went for a proxy Starport. I found it, and killed three pylons powering it, only to have the single oracle it produced rock me. I had been going for a Starport in base. I made a phoenix, which eventually killed the oracle. But it didn’t matter. The damage was done. The game was close after that, but ultimately I lost. That didn’t feel fair to me, because I scouted his proxy, and I countered his tech, but still despite throwing a rock to his scissors, the lost drones and mining time gave him compensation.

I guess the point of the post is that, nowadays you almost always have to rush. Because if you don’t and your opponent does, you’re going to outright lose. If you do and your opponent does not, he’s going to lose. And if you both do, it should be a wash, it should be even, or at least come down to skill. With that kind of logic, who would ever open up 15 expand anymore?

Hyun's Roach Build

I took a look at my win rate percentages this season. I’m doing horrible. Sure my Toss play still looks halfway decent. But my Terran play is terrible. And my zerg play is even worse.

I blame it on the map pool. I kind of hate blizzard now, for letting us pick the map pool. It would have been fun for lols for a week, maybe, but they’re assholes for keeping it this way. I think it’s pretty clear that every starcraft player would prefer to play on the same maps that the pros are playing on. Instead we get metalopolis. It’s insane.

To deal with these new maps, I realized I needed a new build. I found it on imbabuilds.com. I use Hyun’s quick roach build. Basically you 15 hatch, 17 gas, 16 pool, 15 overlord. When the pool is done, you make two queens and continue to drone up to 28/28 supply. Then you make a Roach Warren, three overlords, and burrough research. When the roach warren finishes, you max out to 51/52 on roaches and attack. Often times, it can win you the game. It wrecks 4 gate attacks. It denies naturals. It’s a fantastic build, and I don’t know how I survived without it.

In the World of WarCraft, I met up with an old high school friend yesterday. We visited each other’s Garrisons. That was a first for me. It was a bit difficult, figuring out how to do that. You have to be in a group with someone else to do it. Then, when you right click your portrait, you have the option to visit the group leader’s Garrison. I see this being the new way for guilds to meet. Because for the first time ever, we now have easily accessible and private spaces. Having your own castle is great for cybersex, too.

My friend was mainly concerned with doing stable quests. I didn’t know that when you do enough and upgrade to level three, you get 120% speed on your mount. Which is awesome, don’t get me wrong. But it’s more of a grind. And it’s a grind I don’t really want to do. I finished all the available raids yesterday, too. The Wall raid was really quick and easy.

In summary, I’m very close to canceling my WoW subscription. Probably, I’ll come back after they release a few new raids. Until then, I have to work on my micro to win that StarCraft tournament.

WoW

So I think this is the post where I change the blog to Maul’s Blizzard Blog. Or maybe Maul’s SC2  WoW Blog. I like the second better. Partially because I don’t play any other Blizzard games.

But this morning, I really want to type about the World of WarCraft. First thing, I have to say is that this is the best version of WoW ever made. It really is good. They’ve taken out so many annoying aspects of the game, and streamlined the fun.

But long ago, I made a rule about how I play WoW. The rule is that I’m not going to play the same game, the same quests, the same dungeons, over and over again. I’ll play it through once, maybe again on an alt. But I won’t grind for them, like some sort of game bot monkey.

So I leveled my Druid to 100. I leveled all of his professions. I upgraded my garrison. I got Nat Pagle and all the fishing related stuff. What’s left?

Well last night I tried out the raid finder. I que as a Tank/Healer. I got healer, which I’ve never really played. The raid was MC, Molten Core.

Fun fact, Molten Core was the only raid I ever did in Vanilla. I remember after the grind to 60, I was eager to see what the endgame content looked like. MC was the raid. I couldn’t even keep my assigned tank up. We wiped repeatedly on the first mob of the raid. It was so bad, I decided then and there that WoW wasn’t fun anymore, and that I would stop playing.

So in a way, it was nice to return to MC, as a resto Druid, and avenge that earlier loss from a decade ago. We wound up bulldozing through the entire raid, and I upgraded my helm. It was a great raid, and I’m happy that Blizzard has streamlined the game to a point where a player can casually raid.

But there’s only one raid left. Once I do that, I fear, it may be time to cancel my WoW subscription. Maybe I’ll bring it back after they release more endgame content. It’s hard to say, because I really like playing. But then, there’s that StarCraft tournament at the University.

I started using my Razer Naga again. I want to win a StarCraft tournament.

Razer Blues

I won’t lie. Ever since Warlords of Draenor came out, it’s really been eating into my StarCraft 2 time. Before, I might have played set after set of StarCraft matches. Nowadays, I’m lucky to sneak even a single set.

What’s worse -- my wrist is hurting me from all that extra use. That’s what prompted me to make a post on Reddit about StarCraft and wrist injuries. I have no way of knowing how accurate the responses to my post were, but they were overwhelmingly in favor of wrist injuries being a non issue.

The pain in my wrist has forced me to think about the whole dilemma even more. Which made me remember, that when I play StarCraft or WarCraft, for years now, I’ve been using a Razer Naga. If you’re not familiar with the mouse, it’s got a small 10-key where your thumb is. It means you hit the buttons normally reserved for your left hand, with your right hand.

Take a look if you don’t know what I’m talking about.

Now I’m thinking, maybe it’s all the extra right hand button pushing work that’s making my wrist hurt. Ultimately, I stopped using my Naga. I have a perfectly orthodox mouse in it’s place. So far, my wrist is still sore. But after a week or two, I should know if it made any difference.

But in the meantime, I’ve adjusted to WoW rather quickly. Mainly due to the way Global Cooldowns work in that game. But in StarCraft, the transition is almost impossible. I’ve lost nearly every match I’ve played, since having to rebind keys. I can feel that the transition will suck. But I’m also wondering how I should bind the keys differently, since I’m making the switch anyway.

It makes it harder, because I play random. There are many guides on how to bind your keys for a particular race. But few to binding keys how a random player might organize them.

One trick I’m making use of, is with Terran. I used to bind Barracks to one key, Factory to another, and Starport to a third. Now, I’m binding all production to a single key and using Tab to switch between the three. This will make rallying troops easier, as with my old setup I used to have to rally each production individually.

Wish me luck, and if you know of any good solution, hit me up on twitter @maulakai

Wrist Injury Much?

There’s an issue with StarCraft 2 that needs to be addressed. The 500 pound gorilla in the room. And with upcoming Legacy of the Void, there’s never been a better time to talk about it.

Repetitive motion injuries. There, I said it. This has been an issue for me since day one. Look at the pros. Many of them have issues with their wrists. I typically play three game sets, then take a break to give my wrist a rest. But pros practice for *hours*. To me, it almost feels like they’ve got some kind of superpower. How do they play so much?

The true answer is that they’re hurting themselves, and on some level, we all know it. When you learn to play StarCraft, you’re probably driven by the graphics, by the coolness of the game, and by the intricacies of the strategy. But when you get to a certain point (I’m random Plat) the game really comes down to speed and micro.

I am a self described strategy game enthusiast. StarCraft is a game that, at the top level, demands perfect micro. And that’s okay, in theory. But with wrist injuries so prevalent, wouldn’t it be nice if they could modify the game in a way where it didn’t kill our wrists?

I’m not sure if such a game would be a tad slower, like WarCraft 3. If slowing the game down would make it more about strategy and less about Taeja level storm dodging and marine splits. But Blizzard is a very clever company, and if anyone can make a game that satisfies our strategy game needs while being nice to our wrists, it’s them.

TL;DR Blizzard should find a balance between fast paced strategy game and wrist injuries.

Dark Voice

I did it. I achieved the Dark Voice. I now have the coolest icon there is, pretty much.

One of the best things about having the Dark Voice is that it appears to remove your level number from the lower left hand corner of your portrait. Not all of the time, but some of the time. It feels almost like a small bug in the game that may well be intentional. Because, let’s face it, the Dark Voice portrait looks a lot cooler without your level taking away from it. And so few people have the Dark Voice, who would report that bug??

Now that I’ve gone back to playing as random, it’s taking some getting used to. Okay, Protoss was just like riding a bicycle. I could still pop off a five gate without even thinking. My PvP Stargate phoenix play is still there. It’s essentially the same micro as a mutalisk flock, and I know I’ll never forget how to control my mutas.

I did have to learn a new build for my zerg play. This time, I added Hyun’s three hatch to my arsenal. Let me tell you, I haven’t regretted it for a minute. Having been a 15 hatch kind of zerg player for a very long time, I feel like I’ve finally expanded my understanding of zerg, simply by learning this build.

For the first time, as I play Zerg, I feel like I’m now on time with the pros. At the time they get a big group of roaches, I can make a big group of roaches. When their mutas pop, my mutas are popping. It’s a great feeling. The big difference I had to understand was that you can use your first queen for a creep tumor, since you’re building so many hatcheries.

That exposes a universal truth of StarCraft, and strategy in general. You’re much more effective if you keep your options open, if you keep one of each type of unit, than if you only make one thing. Having one creep tumor is very valuable. Just like having one spine and one spore is so much stronger than having none.

I’ve also had a lot of fun playing a new Protoss build from imbabuilds. This one is the three stalker expand. I love it because the quick three stalkers and mothership core can be extremely powerful. I’ve won games with just those four units! It works against other Protoss, but also against Terran. It allows me to get a safe expand, and to go for my favorite Stargate play, all while ‘rushing’ my opponent with a formidable early force. Tends to be less effective against zerg, though.

v

Did anyone else watch that match of Taeja against Soo? I couldn’t believe it, which is why Taeja is my favorite player. He sure has a magic about him. He goes for a two rax on Overgrowth. Soo counters blindly, with a 14 pool. Yet somehow, Taeja comes up ahead. It’s incredible. I’ve been in that position so many times before, and I always die in situations like that. I always fall behind. But Taeja, almost supernaturally, manages to get the W. And against a player like Soo, who really looks like a bad guy in a movie. I can’t wait to see you play at Blizzcon, Taeja, make me proud!

Terran It Up

So I’m almost there. I’ve almost got my thousand wins as each race option for the Dark Voice icon. My goal is to have it by the time the new World of WarCraft is released. I’m less than seventy five wins away.

My terran game is still getting better. I’ve always been an early attacker kind of player. I see no reason to win a game off three base, where I could have won it off of two. When I first started playing terran seriously, I would go reaper expand every game. Then, recently, I started really working the two rax.

Then I remembered a Taeja game where he sent the SCV that built the first supply depot to scout on Nimbus. When the SCV scouted in the correct location, he had it built a bunker at the edge of the creep in the zerg main. The reaper went directly to the bunker, as well as a second reaper.

The interesting thing about this strategy is that it’s not setup to win the game. It’s setup to catch the zerg off guard, stop him from making too many drones, and ultimately to put him slightly behind.

At first I thought I could only use this strategy if I correctly guessed the opponent’s starting location. But then I realized, even if I got it wrong, the strategy is still viable. Part of why it works is how a zerg player thinks. Zergs hate to have to deal with early shenanigans. Zergs love to 15 hatch, and 15 pool and then maybe even another hatch. We are, after all, talking about playing on the most macro map ever.

The follow up is even more important. I always get the second command center, and transition into the 1-1-1. I’ve found the best combination to be a Barracks with reactor, and tech labs for the factory and starport.

Then you crank out marines, siege tanks, and cloaked banshees. When you have two siege tanks, you push out. The key is to siege the tanks in a location where they can shell the zerg’s main ramp. From there, you can produce a steady stream of tanks, banshees, and marines.

So far, I’ve been invincible with this push. You can expect a counter attack with roaches, queens, zerglings, and maybe banelings if no roaches. As long as you have decent micro, and keep your tanks alive, this unit composition is really tough to beat.

2rax ftw

It’s not like I’ve never seen a two rax before. I’ve seen plenty. But I’d never really played one before.

I think it’s a bit of a chicken and egg problem. When I first learned to play Terran, I learned the 888 and the reaper expand. I really like the reaper expand build, so I played that pretty much every game.

I’m a fan of building my second command center on the low ground. As a result, I’d lose a lot of games on two player maps. When I saw my low win percentages on those maps, I simply vetoed them.

But then I saw MMA play this weekend. I saw Flash have his perfect game, and I thought, I need that. I need the two rax in my arsenal. So I unvetoed the maps and began to play the two rax.

Woa! I’ve been winning games. Winning them quick. Sure, people whine about cheesing, but it’s a perfectly valid way to play. All the pros two rax, and maybe what was missing from my play style was the early rush.

You build a supply depot at ten, and when the SCV is doing building, you send him out towards your opponent’s spawn with a second SCV. Even on three player maps, this technique works well, so long as you place the buildings well.

Each SCV makes a barracks, and while they’re doing that I like to build another supply depot at the top of my ramp.

When the barracks are done, start pumping out marines. The SCVs then scout out the opponent’s base and build bunkers. I guess it goes without saying that this strategy works best in the TvZ matchup. I wouldn’t try it in TvT. I haven’t done it in a TvP yet, but I think it could work there, in the right circumstances.

I’d advise to make only eight marines, enough to fill two bunkers. If you can snipe a hatchery, that’s often enough to win the game later, you don’t have to win the game right then and there.

Monty Hall StarCraft Scouting Problem

The Monty Hall Problem, as it relates to StarCraft scouting. I was practicing my Terran play the other day, when I realized something profound. I sent an SCV to make a supply depot at 10 supply. When he finished making the depot, I sent him off to scout for the correct location to send my initial reaper. I sent him to the closest starting location, clockwise. This idea assumes we’re on a four player map. Which means we’re occupying one of the starting locations. Meaning our opponent is in one of three locations. Typically, StarCraft players choose to scout either clockwise, or anti clockwise directions, and avoid scouting the diagonal location. Imagine yourself as that first reaper. He wants the prize of an enemy base to harass. But he doesn’t know which direction to go. Isn’t he just like that a gameshow contestant on Let’s Make a Deal? http://en.wikipedia.org/wiki/Monty_Hall_problem He can go to one of three starting locations, and I point him to the other location, the closest starting location counter clockwise. But then your scouting SCV acts just like the game show host, and reveals the scouting location closest clockwise, to be empty. Now your reaper spawns and has an easier choice. He only has two locations left to scout. Where should you send him? And there it is, the Monty Hall problem. Obviously, this doesn’t apply if you see any early hints from your opponent, or if your scouting SCV guesses correctly, or if you’re on a map with more or less starting locations. But on those four player maps where your SCV scouts the wrong way, and then your reaper pops, there it is. The starting location you pointed your reaper to still has a ⅓ chance of being correct. But the other location suddenly (and counterintuitively) has a ⅔ chance of being correct. You should tell your reaper to change his mind and pick that other location. You'll be right more often than not. I think I saw Taeja use this logic during one of scouts. I’d love to hear if someone with more of a background on statistics has anything to add.

More Banshees!

So after having switched to Terran play, I started losing. A lot. I’m pretty sure I would have been demoted to gold by now, if the ladder worked the way you’d expect it to. But then, I made an adjustment in my game play. I did a few web searches on Terran builds, and I decided to make more banshees. That, it turns out, was a very good idea.

Against all races, I’m doing a 13 barracks, 13 refinery. Where I used to make two reapers, now I’m often only making one, unless I feel my opponent’s micro is sub par. Next I made a CC, while trying to spend that first 100 gas on a factory. From there, you make two tech labs and start two starports. Any excess minerals go to more marines, and I always like to have a widow mine.

But when you get that first Starport, you put it on the tech lab and make a banshee, and upgrade cloak. With the second starport, you make more banshees.

I’ve found that banshee harass is one of the best tools in the game. A single banshee can win the game, if left unchecked in your opponent’s mineral line. A single banshee can take on any other ground unit. With good micro, they’re even pretty damn good against marines.

But when I reveal the banshees, especially in TvT, I change one of the starports to a reactor, and start pumping out the vikings. I’ll do the same against zerg, if I see him reacting with a spire instead of a hydralisk den. The trick of the opening in TvZ is that if he goes hydra, you have to counter with MMM. If he goes spire, you have to go skyterran with constant viking supporting banshees, ravens, and battle cruisers.

But I even find success with this build in my TvP games. Stalkers are not the best counter to banshees. But many protoss players will make them anyway. Which is why it can be good to have a few vikings, to snipe the MSC, and even snipe observers to keep up your harass with the unit.

For the longest time, I played the game in a cold, calculating way. Carefully considering my opponent’s unit composition and choosing my units accordingly. But now, in my Terran games, I have a new mantra, and it earns me *victory*: More Banshees!

Quest for the Dark Voice

I was going through my achievements in StarCraft, when I remember the one about the dark voice. You have to win a thousand games with each race. Years ago, I remembered reading a post on the forums about how almost noone had it.

As a random player, I knew I wanted it. I knew I already had a thousand wins with Zerg and Protoss. So I looked up how many wins I have as Terran.

Six hundred and seventy four. Not as many as I would have liked. I never realized before how much I’d done the lion's share of my losing with Terran.

So I did the unthinkable. I changed from Random to Terran. And I vowed to get my thousand wins before going back to random, with the Dark Voice icon.

I also watched a few pro terran games. I saw a build that Taeja used against Protoss on a large map. He went CC first, into two rax, into a bunker. The problem is, I can never seem to pull off what Taeja can do effortlessly.

I tried another build that goes directly into three rax stim and combat shields. The point of that build is to run into a toss natural and snipe it. There is some promise there, althought I haven’t yet been able to work the kinks out. I think a slightly later timing with Medivacs might really push that one over the edge.

I’ve also made progress in my TvZ. My most comfortable opening, in all matchups for that matter, is going Barracks first, making two reapers to scout and harass. Then, against a zerg opponent, I reactor out four hellions, and snipe the zerg creep tumors, trying my best to deny his third base. I follow that one up with cloaked banshees. From there, you really have to tailor your attack to what your opponent is building. But I’ve found that in TvZ matchups, you often want to start adding vikings. If you can take a third or a fourth easily, I always go into three starports, two with tech labs for battlecruisers, one with a reactor for vikings. If you can keep your viking count high enough, you can have an invincible army.

Of course, his only real counter to that isn’t corruptors. It’s Hydras. So if he does start to make tons of hydras, you have to be able to counter some of that with good old fashioned Marine/Marader/Medivac.

TvT used to be my best matchup. I would also go for quick siege tanks and deny my opponents natural. But nowadays, I’m getting worked by cloaked banshees. Even missile turrets and an emergency viking don’t seem to help. That’s what really annoys me. That you can make a turret, and still lose tons of scvs.

Got any good builds? Let me know.

Thoughts on Zerg Builds ZvP ZvZ ZvT

I feel like my win percentages with Zerg have gone way up, now that I’ve changed my standard openings. I’d like to write a bit about each Zerg matchup.

ZvZ

This is the matchup where I’ve changed my opening the most. I now go standard 14 Extractor, 14 Pool. I drone until the pool is done, then make non stop zerglings. I’ll buy the speed upgrade with the first 100 gas. A baneling nest with the next 50 (usually I throw a drone in to replace the one building the baneling nest). From there, I que up my lings to somewhere outside of his base, but away from his scout. The first few lings I make into banelings. Then when the first wave of the queen’s inject larvae lings comes, I attack with my first wave.

I use all of the larvae off one base to make zerglings and banelings. But when I have enough excess minerals, I’ll expand and add queens, etc. Most ZvZ games I play are won or lost here. But if the condition stabilizes, you’ll usually graduate to roach/hydra, or muta/ling/bling.

ZvP

I still like to go hatch first. But enough toss players have punished me that tend to go pool first in all but the most favorable conditions. Hatch after pool. If you scout the toss player taking a second base, take a third. Stay gasless until six minutes in.  By that time, with lings and overlords you may be able to saturate the third. Unless you scout some type of a gateway attack. In which case, you should immediately build spines, roaches, and lings. I like to err on the side of caution. Even if the attack doesn’t come, a nydus worm follow up is often viable.

Ideally, I’d be able to drone until lair tech, and build five mutes. Depending on his reaction to the mutes, I’d grow a cloud of them, or tech switch to roach/hydra. Toss players today have shied away from corrupters, making hydras a great unit. When they add in colossi, you already have the spire to make corruptors.

ZvT

I always go 15 hatch 15 pool in this matchup. When I scout the terran throw down his second base, I throw down my third. I make queens constantly, but avoid making any extractors or mining any gas until six minutes in. The extra minerals allows me to get a wall off, roach warren, spore crawler, spine crawler, and evolution chamber, as well as all the queens. With the added creep spread, my queens are adequate to defend against the hellion/reaper pushes that are popular now with terran players in the early game.

Unless I spot an early push, I’ll spend the first 100 gas on lair tech, the next 100 on speed, the next 50 on a baneling nest. There’s nothing wrong with devoting excess gas to small roach force. If I build up too much gas, I’ll start  my upgrades. But really, I’m wanting to get my spire down as quickly as possible. When it’s done, I spent 100 gas on the +attack upgrade for flyers and build no fewer than five mutalisks.

Just like in the ZvP matchup, I pay close attention to how he handles my harass with those five mutalisks. If he handles it poorly, I’ll add to the flock throughout the entire game. But if he makes a bunch of towers and has good stim marine control and widow mines, I’ll keep the mutes in the airspace around his base to counter drop play.

From there, my favorite way to win is to simply deny third bases until he runs out of money. Good creep spread and a roach/hydra army are often enough to do this. Ultra/Hydra is my favorite unit composition, or at least the one I most often strive for.

StarCraft Weekend Ruined

I don't know about you, but I went to StarBucks last night and grabbed a coffee. I wanted to enjoy my Friday but playing games late into the night. But I got a bug, like many people are experiencing. Even this morning, Saturday, the 31st of May, I can’t play StarCraft. I’ve stooped so low as to play HearthStone and Heroes of the Storm. And while I had a little fun playing those games, I can’t help but feel a little off my game, for not having my usual fix of StarCraft. Come on Blizzard.

Zerg and Toss Strategy Changes

Once, while watching StarCraft games on Twitch, I remember hearing the commentator say something about a zerg player not wanting to make his extractors until about six minutes into the game. This little tidbit of wisdom stood out to me because I had been playing zerg very differently.

When I random as zerg, I always go for 15 hatch, 15 pool, 15 extractor. Following that up with baneling or roach attacks, as the matchup goes. Zerg, for the longest time, used to be my prime race.

But this season, I’ve been doing horribly with zerg. They are now my most losing race. So I decided I had to rediscover what openings to use this season.

Turns out, in ZvZ matchups, you now have to early pool. There is no one time to pool that is always best, it varies map to map. But I find that a 14 extractor, 14 pool, into speedlings and banelings tends to work well.

ZvZ nowadays is all about the baneling wars. And rightfully so. Two banelings can destroy your opponent’s economy. And a pool first build vs a hatchery first build can almost always have the zerglings streaming in before a proper wall off is made.

But in the other two matchups, I’ve followed the advice of that commentator whose name I can’t recall. And to good results. I’ve finally had my first ZvP win with a 15 hatch, 15 pool, no gas until six minutes. If a scout a four gate attack, I simply build a lot of spines. If nothing comes right away, I wall off with detection for the inevitable dark templar cheese. Also, when doing this no gas until six minutes build, its important to have constant queen production. They will help in the event of oracles or early gate attacks. You need that creep spread and transfuse.

I’ve enjoyed equal success using the same build against terran opponents. One of the best things about it, is it lets you choose nearly any tech path you want. Because some terrans build legitimate defense against mutes, and when that happens, you want to be able to transition into a composition that will allow you to keep a turtling terron on two base.

The only other development in my StarCraft play comes in PvZ and PvP. I’ve been using a nice little Core into Stargate build. Traditionally, in a PvZ I would nexus first. But that hasn’t been going so well for me. Hence the one base Stargate play, expanding shortly thereafter. The idea is to build phoenix and do a ton of harass.

As a natural mutalisk player, I really like going for early phoenix. I find I’m quite good at sniping queens, overlords, and drones. Even denying thirds.

What’s great is that the same build works in PvP. Part of the reason why is that you can easily snipe the mothership core. That alone usually keeps a protoss in his base. But what really makes the build work is how it counters other builds. Because it snipes the MSC, it can shut down blink play. And with the ability to pick up immortals, it can counter pushes with that unit, too. As a matter of fact, having a small cloud of around 5 or more phoenix shut down oracles, and even straight up gateway pushes, as they will remove as many units from your opponent’s attack.

Terran Game Theory

Today I’d like to write a bit about playing Terran. I feel like I’ve come to a new understanding about how to play Terran, and most of that is coming from watching Byun games.

The first thing I learned is that I should open with a 13 rax, followed by a refinery, into two reapers. Part of the reason why I do this is because I play random. That means I don’t like to reveal my race to my opponent through an SCV scout.

Before I was content to scout with a single reaper. But after seeing what a skilled professional can do with two, my opinion has changed. Two reapers are so much better than one. They can tank way more damage. The second reaper often allows the first, if he gets low on HP, to escape. With two reapers against zerg, I can always scout the drone count to see if an early baneling bust is coming. Two reapers can kill drones, queens, even stalkers, if you micro them right and get out while you can. They keep the game in your opponent’s base. About they only thing they can’t do is scout a proxy Stargate. But they can scout the main base to know if a proxy is likely.

Back to your own home base. So you make the two reapers, then what? You will soon have approximately 100 gas and 400 minerals. In my opinion, there really are only two options. You can start making a command center for your natural. I usually make it in my main base, but depending on the map and the matchup, you can build it directly at your natural. Or, with those minerals, you can build two-three more barracks.

With the hundred gas, you can build a factory. Or you can add a tech lab to your first barracks, and start researching Stim.

Depending on the map and matchup, at this point in the game, I like to choose either a Marine, marauder, medivac stim timing, or a more defensive build into siege tanks. I’m also a firm believer that a single widow mine is great to have in any situation.

Versus protoss and zerg, I feel the stimpack timing is better. I make three rax, two labs and one reactor. I research all of the upgrades. And I add on a factory and starport, and immediately start dropping. A good zerg player might have mutes at about the time you attack. If you go this route, it is vital that you do some damage. Against toss, due to photon overcharge, I find it’s best to snipe tech structures. Against zerg, I find queen snipes and tech snipes tend to work. Because a player with good control can usually save most of his workers. But if not, drone kills always work. The most important thing to realize is that you have to make it out with your units alive. Getting the tech snipe but losing the units is a loss. Killing drones but losing your units is a loss. You have to do as much damage as possible, while still keeping those first few medivacs alive.

In a TvT, I find it’s best to go the siege tank route. I make a medivac first, a viking second, and then resume medivacs. If your opponent surprises you with banshees, that viking will save the game. And even if he doesn’t, the viking can pay for it’s self by taking out the enemy’s medivacs, spotting for the tanks, and even landing and getting in on the fight if things are close. I should also note that while I still make two reapers in TvT, I wait until I have *both* reapers to move out. This is because often times your opponent is also sending out a reaper or two. If you can catch his reaper in your base, you can almost always win the fight with your second reaper and or SCV support. But if you move out as soon as your first reaper is built, then he has a good chance of catching you with your pants down. It’s very clutch if you can get into his base with both your reapers and snipe an SCV building a tech structure -- like his factory or expo.

I always move out when I have two siege tanks with a third on the way. Normally, your Terran opponent will also have siege tanks by this time. Here is where I stretch my advantage with positioning. I find that I can almost always abuse a ledge, and with superior positioning, take out a Terran. Sometimes this siege tank push can work against a protoss player.

I should caution that my worst matchup is TvP. The only times I ever win in a TvP is when I have insanely good luck with drops, and continue to pressure the entire game.

My next best matchup is TvZ, where I have a great mech build. Except it’s vulnerable to some timing attacks against a good Zerg player. For that reason, I’ve been trying to start with the stim timing attack, and transition into a bio mech build.

My TvT siege tank push wins for me more often than it loses. But it’s true that against an equally skilled opponent, it can be repelled.

Playing StarCraft Again

I’m not sure why, but I’ve been playing a lot of StarCraft lately.

the community isn’t what it used to be. Sure, when there’s a tournament going, there will be plenty of people watching on Twitch.tv. But nowadays, StarCraft often is so slow on the user count that you have to scroll to find it.

And here I am, still playing random. Only this time I seem to have learned how to play Terran better. It’s all about the stim timing.

It’s all about me watching a lot of Byun on stream. One of the biggest questions in playing StarCraft amounts to “What do I spent my vespine gas on?”

When you play Toss, you’re almost obliged to spend it on Wargate, a stalker and a MSC. With Zerg, it’s speed in nearly every matchup. But terran, the answer isn’t so set in stone.

But Byun would always spend that first 100 gas on two reapers. Traditionally, I had been buying only one reaper. But look at how he harasses! I saw him milk those reapers for all that they were worth. In every matchup, I realized, my reapers could be doing more.

So I started spending my first 100 gas on two reapers. Then I copied how the timing works out so that he spends his next 400 minerals on an early expo. Perfect. I love his openings.

Nearly every ZvT I play, I go for a baneling bust. One of the most difficult aspects of playing StarCraft is the reverse. It’s when you’re playing in a TvZ and you have to scout the baneling bust. Many players, at my level, don’t. They get angry and bm over you “Cheesing” them. But with two reapers, you can get an early initial scout. You can sometimes get a few drone kills. And  you can also heal up at a watch tower to come back in for the vital scout. That scout lets you know if you need to really bunker up on one base, or if you’re safe with two. It also gives info about whether you’ll want to build an early widow mine from your factory, or if it’s safe to skip it for more tech.

I notice a lot of 3rax play. I had been thinking it was more like four rax. But three seems to work better. Two rax with Tech labs, one with a reactor. Then you can factory and Starport into MMM. The trick is to give tech priority to Stim and Combat shield. You can get two drops going fairly early with this build.

The only drawback is that you really have to do damage. If you don’t, you’re likely to be behind. Zerg players can spot the drops coming with good OL placement. And against toss players, often a tech building snipe or a few drones are your best option, because you’ll always bait out the defence of the MSC and have to retreat.

The other way I like to play TvZ is to go full mech. Just double armory, double upgrades, mech. I prefer my army composition to start with 4 to 6 siege tanks. Then add on as many thors and hellbats as I can, while I expand to a third base. The idea is to sit on three or four base until you max out with mainly thors. The problem with this build is that it is susceptible to attacks, and pretty much cedes map control to the zerg player. Later on I’ll add on a stargate and raven and possibly battlecruisers.

I find the full mech army *with siege tank support* can demolish just about anything. It does poorly against swarm hosts, which is why I have been experimenting with adding on Banshees. But in the end, I know the build needs a little work. Least reason of which is because there really is no harass with this build. It’s all turtle. And all turtle can never win against a player with full map control and good game sense.

Still, even as I type it, I can see how we could add on more harass. Because the build still starts with the two reapers. If you can keep them alive, you’ve got a good followup attack with Reactored hellions. If you keep that count up high, you can deny the zerg expos. Which would make turtling a whole lot more playable. I’m definitely going to try it in later games.