Transcripts (of Day[9]'s Podcasts) - Page 3

The f-keys not only make everything much quicker, but you’re much more accurate and less likely to make those mistakes. There’s tons of uses for F-keys, and the most important thing for you is that you just start using them. The easiest starting point if you’re uncomfortable with the F-keys, is to use them on your main, natural, and then second expansion. This way, you can be moving workers around much more cleanly using these F-keys. From there, as your hands develop more comfort with rapidly hitting F2, F3, and F4 you can branch out and assigning more uses to those F-keys now that your hands are comfortable. Now that we’ve discussed F-keys, I want to come back to discussing hotkey setups – what should your control 1,2,3, up to 0 keys be to allow you to play efficiently. First of all, your hotkey setup should be something you use to play more efficiently. So, if the setup doesn’t make up logical efficient sense, it’s probably bad. For example, let’s say you’re a Protoss player, and you hotkey your nexuses as 1, 2, and 3, and your units as 8, 9, 0. If you want to be making probes you have to do 1p2p3p and then to attack you do 0a9a8a. In both situations, you’re reaching across the keyboard in pretty much a really painful manner and that’s like totally bad for play because not only will that slow you down, but your hand will hurt, so that would be an example of a really bad hotkey setup.
So if you want to make a good hotkey setup, you have to consider what hotkeys are going to be pressed a lot for each of the races. If you’re a Zerg player, for example, the s key is used all the time to make things, you know, like sz for zerglings, and sh for hydras, and so on. Terran players need to consider the m key a lot, because they’re constantly making marines against Zerg, and the t and v keys for tanks and vultures. Protoss players need to worry about z and d for zealots and dragoons. On top of commonly pressed keys, all of the races need to consider the really really important keys that they’ll need in an instant. The most obvious example is Terrans with scanner. Even though you’re not necessarily constantly scanning all game long, you do need to be able to do it quickly, like if you see lurkers or dark templar. Whatever hotkey setup you want to form, you should just take into account all of these factors when constructing things.
But by far the best word of advice I can give anyone who’s trying to make a hotkey setup is not to be afraid to just start using random hotkeys for anything you might think is important. For instance, when I’m playing Zerg, early on, I don’t have too many units or hatcheries, so I use the 4 hotkey as my quote-on-quote important building. So, for instance, I will hotkey my spawning pool as 4 early on. That way, when I’m microing zerglings around my opponent’s base, I can just hit 4m to upgrade zergling speed and I never have to leave vision of my zerglings. Likewise, I will use the spire as 4 in Zerg vs Zerg. So that way, if I – (again, I’m microing in my Zerg opponent’s base) I can just constantly check the progress of my spire so that way I can know when to stop making units and start saving larvae for mutalisks. In fact, any little nuisance that you can think of in your play, try to create a way that hotkeys can solve that problem, however big or however small that problem might be. For example, early game, when I’m just making my first four or five drones, it always really annoyed me that I had trouble lassoing the drone right as it spawned from the egg, and I always wanted to get that drone moving towards minerals as fast as possible, and on occasion I would screw up. So, every time I build a drone now, I hotkey the egg as 3, so I have my scouting overlord as 1, my next overlord as 2, and any drone that’s about to pop out, I have it as 3. So when it hatches, I just click 3 and then start spamming on the mineral patch. There, problem solved! Naturally, your hotkey setup progresses throughout the game, so for example I just said that I used 1 and 2 as an overlord and 3 as a drone. Later on in the game, I have 1, 2, and 3 as units. And it’s totally okay to consider how these progressions will work in your play – when to override old hotkeys and put in new ones.
With that in mind, let me present three hotkey setups, one for each race. That way, you guys can use this as a reference point for your own hotkey setups. Let’s begin with the race that I play – Zerg. My general hotkey setup is that I have 1234 as units and then 567890 as hatcheries. Notice how easy macroing is when you have 5 and up as your hatcheries, because you can use your pointer finger on the number keys, the ring finger on s, and then the pinky finger on z. Contrast this with someone who uses 1 2 or 3 as their first hatchery placement. It’s really awkward to try to use three fingers. Most of the time, you end up being forced to use too two, which will slow you down tremendously, considering you’re building hundreds and hundreds of units every single game with Zerg. 5sz6sz7sz, going all the way up the keyboard, I don’t ever have to rearrange the way that my fingers are placed. I just simply stretch with my index finger a little bit more. The same goes for hitting 5sh and 5sm for hydralisks and mutalisks – I get to use three fingers in each of these circumstances. The 1234 for units is pretty self-explanatory. The a-key is right there and that allows me to maneuver my units totally effectively. I always have my main, natural, and second expansion hotkeyed to F2, F3, and F4, repsecitlvely. That way, when I’m making drones early, I can very easily split them around where I need to by lassoing the drone and hitting F4 to move it to my second base or lassoing the drone and hitting F2 to move it home. Likewise, if I’m getting storm dropped, I can just go F2, make a box, F3, send them to my expansion. There’s a few other little things I do with Zerg – for example, I usually have 0 as my important units such as defilers or queens, that way it’s very easy to access, and that’s not a unit that I’m going to be spamming a whole lot and remicroing as I would be with zerglings, lurkers, mutalisks, and so on and so forth. That way, if I have maybe four or five hatcheries, and haven’t quite hit six and getting ready to go into macro mode, I can use that 0 key to control my defiler or queen a little bit more effectively. I choose not to use F2, F3, and F4 to macro because I like to expand a whole lot as Zerg, and if I have four hatcheries at four different bases, it takes a long time to hit F2 and then make something and F3 and make something and F4 and then make something and then click on the other expansion and then make something. Considering larvae spawn every thirteen seconds, it’s way easier to just have all those hatches hotkeyed throughout the entire game.
Now let’s discuss a basic Terran hotkey setup. For Terran, my basic hotkey setup is fairly similar – I use 1 to 4 as units, 098 as scanner, and then 567 and 8 as unit-producing structures like factories or barracks. For the most part, with Terran, I tend to macro by clicking on my structures. I’ll hit 55 to ping back to a cluster of barracks, and then just spam m all the way along. Early on, though, I do have 567 and 8 hotkeyed as barracks, so for example, if I’m being very aggressive, I can just hit 5m6m7m8m, and you’ll note that that’s fairly easy to do if you place your thumb on the m-key. If I’m doing something more aggressive, like doing a fast dropship, I generally keep 5 and 6 as those factories still and then 7 as the starport. In general, though, 5 to 8 is just the unit-producing structures that I will occasionally need to build from when I’m not looking at them. Most of the time, though, I just keep all my unit-producing structures clustered, and then I hit 55 to ping back to it and to spam-build everything. If I end up with more clusters, I use 6 as the next cluster – so for example if I have a whole bunch of factories in my main and then a whole bunch of factories at an expansion, I’ll keep them both clustered, and then I’ll hit 55, build from the first cluster, 66, build from the second cluster, and so on. Again, 1234 is totally standard for units – 1t2t3t4t is fairly easy to hit if you move your thumb up, or you can even use your index finger. And then, naturally 0, 9, and 8 are all hotkeyed to scanner because those are the sorts of things you need to ping back to immediately. I don’t actually ever have my command centers hotkeyed, I find it very easy to do F2 s F3 s F4 s, use those screen hotkeys over my command centers. Not only is this useful because it helps you dodge storm and reaver drops, but that motion – hitting F2 s and F3 s – actually just feels very pleasurable and is very straightforward to do.
Now let’s move on to a Protoss hotkey setup. My Protoss hotkey setup is actually very similar to my Terran setup. I have 1234 as units, and then 5 to 8 generally as my unit-producing structures, and I also macro in the same way. I hit 55 for one cluster of gateways and spam along that and 66 for the other cluster of gateways, and I spam along that. I do tend to have my nexuses hotkeyed as 098. Again, because I’m making a probe every thirteen seconds, it’s very easy to go 0p9p8p to spam probes very easily. Again, I use the F2 F3 and F4 keys to center over my three nexuses at my first three expansions, to move probes around for storm drop. With both Terran and Protoss, I do the re-rallying trick I told you about earlier, where I hit F2 over the cluster, and then F4 where I want to rally. That way, I can re-rally these huge clusters very very quickly.
I want to point out that none of these hotkey setups should be considered as the best hotkey setups for that race. I’ve heard a lot of Terran players use 1 and 2 as scanner because the 1 and 2 are right next to the s, and then they use 3, 4, and 5 for medic marine because it’s very easy to go 3t4t5t, for stim that is. Or a lot of Protoss players like to begin their gateways at 4 because they never really have that many units early game so they don’t necessarily need the 123 and 4 hotkeys all game long.

The most important thing to take from the examples that I provided is that I'm making full use of the keyboard. Every number from 1 to 0 had some use in every scenario. I was even using the F-keys in all three cases.

The best thing that you can do is to think of a complete hotkey set-up and stick to it religiously. For example, again, let's say that you're one of those Protoss players who likes to click on the probe to build it, instead of hitting the p key. You might say to yourself, "Well, let me use 0 and 9 as my nexuses, so I start doing 0p9p and I don't waste that mouse action." What's most important is that you just stick to it, that when you're in-game, you don't get lazy and start clicking on something. What'll eventually happen is, though you begin slow with your hotkey set-up, you'll begin doing it faster and faster and faster, and as your APM increases, the fact that you have an efficient hotkey set-up will allow you to play all that much better. An efficient hotkey set-up is the key to having good micro and macro simultaneously. There are many, many players who are good at just microing or good at just macroing, but by having a really good hotkey set-up, it allows you to be in many places at once.

Since I've spent so much time discussing the keyboard, I wanted to move on to the mouse. Remember how I talked about the pivot point for your keyboard hand being that small bone in the bottom-right corner of your wrist? Well, we want to use that same pivot point for our mouse hand: we want to be putting all the pressure for our mouse hand on that same bone that now is in the bottom-left corner of our right hand. That's why people always joke about the "progamer callus" being that bone, right there--that's because players tend to keep all the pressure of their hands there, first of all, because it's very difficult to cut off circulation when you're putting pressure just on the bone, and also because it's a very solid pivot point to allow you to control your hands in a very straightforward manner.

In fact, if you're at your computer right now, I want you to pick up your mouse hand and place it right at the edge of your table so that you feel pressure right on that bone. And then try moving it around: notice how easy it is to stop accurately and to move it around in quick bursts. Contrast this by moving your mouse hand way far up the table, to the point where, like, the middle of your forearm is now at the edge of the table. It's actually a lot more difficult to move around and control, because you don't have that solid base pivot point as that bone. I still remember the first day that I moved my mouse hand from way up on my desk all the way to the front of the desk. I had a tremendous improvement in my micromanagement, primarily because I was able to easily stop my mouse pointer once I hit a specific location. So, for example, if I wanted to click directly at point x with my mutalisks: before, when I moved my mouse hand to point x I would tend to shake a little bit, and I would be able to get close to x but not quite there: my mouse hand would always waver around at specific locations, but by having this pivot point, I was able to stop much more accurately.

Naturally, any discussion of a mouse comes with a discussion about sensitivity. First of all: do not use mouse acceleration at all. Do not use it; it is horrible and hugely detrimental to your play. You'll find you have big difficulties remaining consistent with your mouse movements if acceleration is on. Many operating systems call mouse acceleration "enhanced pointer precision". You want to make sure "enhanced pointer precision" is unchecked--I know Windows XP does this. In fact, if you're using Windows XP, it's actually impossible to remove mouse acceleration just using the basic menu options. So if you go Google "remove mouse acceleration XP", it will give you a registry key that you can enter into the registry editor that will permanently fix mouse acceleration. I actually do this at every tournament I go to: I manually go into the registry editor and put in the correct smooth-x and smooth-y curves so my mouse responds exactly as I want it to.

In regards to sensitivity, I notice that there's a lot of players who jammed their mouse sensitivity to the max once they saw players playing at 300 APM. Many 300-APM players don't actually have very high mouse sensitivity at all, and, in fact, if your mouse sensitivity is too high, it's really difficult to do those fine-tuned, precise clicks that you need to. I would recommend the mid-range sensitivities: don't go for anything too high, and don't go for anything too low, like what Counter Strike players use. You want to hit something in a nice middle range; I tend to be at the high end of that middle range, but, again, I've practiced it for a very long time and I'm very comfortable with it. Just choose one and stick with it until your mouse hand feels really, really good.

Once you're comfortable with your mouse hand, there's two general things you need to be very good with. One is making huge, fast sweeping movements, like if you want to make a big box around a bunch of units, or you need to send mutalisks from the right edge of the screen all the way to the left edge, immediately. The second is fine-tuned little movements, like if you want to right-click on just the exact SCV that's at low health, instead of the one that's just a few pixels away from it. Both of these require a whole lot of practice, so, naturally, you can improve by just playing a lot, but I strongly recommend playing some reflex games, like ones you'll find at missionred.com, and, in fact, I strongly recommend Minesweeper as an excellent game to help improve your mouse accuracy. When you play Minesweeper, you have to click at such a high rate, at such small boxes, that your hand becomes really, really sensitive to fine-tuned movements with your mouse hand, and, in fact, there was a period of time when I played a whole bunch of Minesweeper, and when I went back to playing a lot of Starcraft I found that I had a much easier time controlling mutalisks and zerglings and clicking on exactly what I wanted to, even when I was nervous.

Even when you're just by yourself at your computer, chatting on AIM: if you're going to close the window, try to move your mouse pointer onto the X to close it as quickly and accurately as possible. For example, right now, when I'm going to finish this Audacity recording, I need to move my mouse hand to the stop button. I'm going to try to do that in as accurate a whip motion as I possibly can. Any time you're at your computer doing something with your mouse, just try to be precise with it. And the more conscious you are of this, the more rapidly you'll improve your control in Starcraft, without ever even having to play a game.

Last, if you want to improve your micromanagement with certain units, I strongly recommend downloading micro, single-player maps, or just playing against a computer and controlling the specific unit you want to. So, for example, if I want to practice mutalisk control, I will just open a 1 vs. 1 against the computer and micro it around. I also have a mutalisk training map of mutalisks against scourge. If I want to practice my vulture control and I don't have a Use Map Settings map, I'll just make one--they're actually very straightforward to make on your own--and I'll just plant a zergling at spread-out locations on the map and have one vulture. And after you play this for, maybe, 4-5 hours, just doing the control of one unit, you will have almost mastered that unit. I actually did this with my mutalisk control: I just spent 4-5 hours trying to make sure my clicks were just right to kill scourge, and after just one day of practicing, I had a tremendous boost in my confidence in Zerg vs. Zerg.

That about wraps up the basic discussion of mechanics. What we've covered in this is

• How to hold your mouse and keyboard hand
  
• and how to make sure that you have a good hotkey set-up
  
• and how to practice fine-controlled mouse movements.

I hope this is enough as a good base-point that will allow you to improve your micro and macro and help you win more and more games, without ever even having to really change your strategy.

That's all for now. Thanks everyone, good luck, and cheers.

[Beginning Music: When Things Fall Apart by Proem]

Hey everyone this is Sean Plott, AKA. Day[9], presenting today the hallmarks of expert play part 3: Having a Good Mind Set.

In virtually any pro-gamer interview you’ll read or see online, whenever they ask “How were you able to win today?” the pro-gamer always seems to respond, “Oh it’s because I practiced so much I haven’t slept or eaten in 3 weeks because I’ve only practiced tirelessly to win this match.” Of course, I don’t think anyone would argue against the fact that Starcraft is an incredibly difficult game that requires a huge amount of practice to maintain your skill level let alone improve a bunch. However it’s really easy to put too much emphasis on practice. I’m sure everyone has heard of the story of the kid who sleeps all the way through high school because when he gets home at 3 pm he does nothing but play Starcraft ‘til 5 in the morning because he wants to be the best. However after a year and half of doing this, that kid is still C- and can not seem to get any aspect of his play better no mater what. So, though practice is very important, many people such as the kid I just mentioned will not improve based upon the huge amount of hours they play simply because they have the wrong mindset. These are the players that can not reflect upon their game and analyze it properly and as a result stagnate for very long periods of time.

Naturally, with an introduction like that, I’d like to present on how to have a good mindset when approaching the game. I want to provide a structure to your thought process so that way you don’t end up in any sort mental prison that will prevent you from improving. Since, having a right mindset is sort of a vague broad term that encompasses a huge aspect of play, what I’m going to do is present 5 ideas, each of which can only help to improve your play, and hopefully other areas of your life too.

So with that in mind, let’s begin

Number 1.

Number 1:

Always have a plan.

Note that I did not say always have a build order. Those are two very different things. A build order is just a set of instructions that you follow to optimize something. The plan is generally what that something is. You always need to know what your plan is first. So let's distinguish between these two even more with an example.

Suppose an intermediate Terran player has a build and versus Protoss on Destination that allows him to set up a contain outside the Protoss natural. Virtually anyone who would watch the game at this point, when the contain is all set up, would say oh yeah, the Terran is in a really good situation right now. The Protoss couldn't possibly break the contain across the double bridges the Protoss will have to do some sort of shuttle harrass or expand to that back mineral natural. Either way the Terran player is in a really great position. It's normally at a point like this when you see a Terran player totally collapse and lose the game in a silly fashion.

For instance this Terran player will overcommit to his containment and when the Protoss starts doing counter harrass drops they'll do way more damage than they're supposed to and the Terran will end up losing. Or sometimes you'll get a Terran player who'll get excited at the fact that he has this nice contain and they'll try to push into the natural and get continuously pushed back by dragoons, reavers and zealots. Note that I am NOT saying that the Terran player shouldn't push into the natural, and I'm NOT saying that a Protoss counter attacking will always work. Rather, what I'm saying is that the Terran should have already had a plan of what he wanted to do after that contain was set up. Whether he wants to play passively or aggressively is fine. It's most important that the Terran player had that all laid out in his mind in advance.

Moreover, when I say plan, I do NOT mean a series of responses or counters. Thinking of Starcraft in terms of responses and counters is actually this toxic mindset that completely polluted most forums on the internet.

Thinking too much in this "if he does A, I'll do B" mindset can result in what I'll refer to as very bumpy or very choppy play. Let's take Protoss vs. Zerg for instance. You'll see a lot of players will think too responsively and do something like "oh my opponent is making a lot of mutalisks, I'll make some corsairs now so I can hold that off"; and then their opponent starts getting some hydras and they'll say "I'll get storm to make sure I can hold that off at my front and maybe I'll get a few extra cannons" and then they'll get some dragoons when they see lurkers and they'll have to get some observers too and then they wanna make sure they get zealot leg speed.

What ends up happening is that they end up with this nonsensical silly timed army that can't really accomplish any task when they get to the end of the game. Those observers will come too early and they can't really do enough damage to the lurkers. You'll see this sort of player get storm for high templar and then have to morph them into archons to hold off some mutalisks. They'll have a whole bunch of corsairs that were also used to try to counter those mutalisks but now all the mutalisks are dead and they have a whole bunch of corsairs that can't really do anything for the remainder of the game because the Zerg is very good at defending his overlords.

This is exactly the sort of problems you'll encounter when you think too responsively. So, let's run through an example of having a really good plan. Let's take Terran vs. Terran on Destination. One of the key things that I immediately noted when I saw Destination is that the left and right expansions are crucial in Terran vs Terran.

Because they're high ground and they're both in the middle of the map, if you can control the ramp at both the left and the right expansions then you will end up with a resource advantage at the end of the game. In fact, you'll see a lot of players in Terran vs. Terran setting up siege tanks and turrets there long before they expand just because it's such a key tactical point that, by securing, guarantees late game economic advantage.

So let's say I wanted to play Terran vs. Terran on this map. Note that I suck at Terran vs. Terran. I'm awful at it. So, what I would do is first say: "What's my plan on this map?" "Okay, it is to try to secure the left and right expansion ridges so that way I can win in the late game with Battlecrusiers or whatever. I just want to win by starving my opponent to death." At this point in time that's when thinking of a build order comes into play. Then thinking of these responses or counters comes into play because now that I have my plan in mind I can make sure that any response or counter I can do doesn't violate my underlying game plan. I can make sure my build leads all the way up to securing that left and right expansion ridge area. The build orders that I'll explore in the game would be something like getting many factories with vultures and mines early on to see if I can secure the middle region with vultures then get a whole bunch of tanks in the midgame before my third expansion or something like that. Whether it works out or not I'll find out through practicing, but all of this will be dictated by my underlying gameplan.

And let's further say that I'm practicing and I have these ridges secure and I'm in the middle of the game and suddenly it seems like one of his expansions might be vulnerable if I walked my tanks forward. Is that a good idea to do? Well, probably not because it completely contradicts my gameplan. Why would I give up control of a ridge to kill off an expansion when my entire build up to that point has been geared towards securing that ridge.

Obviously if I can kill the expansion and make sure my ridge is still secure at the left and right expansion, then yeah, by all means I'm gonna do it. But I'm always gonna question all my decisions first to see if it screws something up in my overall game plan.

Again, I'm AWFUL at Terran vs. Terran; however, I will have far more success than a player who is talented mechanically purely because my gameplan and learning is guided by this pregame decision and my underlying game plan.

Number two: (8:30 timestamp)

Hello everyone, it's your friendly neighborhood Sean Plott aka Day[9] and today I want to spend a little bit of time talking about Terran's midgame in Terran vs Zerg on Heartbreak Ridge. Whenever you're making a plan in Starcraft for what you wanna do it's critical that you think about the mid and lategame first, develop a really good plan for that area, and then do the opening last, think about what your exact opening build is going to be. This podcast is not going to explain that concept in detail, but instead it's going to be an example of using that sort of mindset to formulate really strong play in Terran vs Zerg on Heartbreak Ridge.

So, let's do the most natural thing and look at the map. Generally when people look at maps, they suck, right? They are so bad at looking at new maps. The first thing that people do is think "Hmm, how can I do my python build on this map?" And when that python build doesn't work, they complain about imbalance. And the second big mistake people do when they look at maps is to focus waay too much on tactics. They obsess about the fact that you can put a tank hereor there, or that mutalisk harass is really good here or "Ooh, how do you stop a reaver drop from coming through that?" and some of those considerations can be somewhat useful, but I'm choosing Heartbreak Ridge because the aspects of the map I'm going to focus on have a heavy impact on the strategy we're going to do, as opposed to the tactics or little abuses. So if you don't have Heartbreak Ridge open right now, don't worry! 'Cause we are only going to be talking about the broad characteristics of Heartbreak Ridge. First of all, Heartbreak Ridge has a circular dynamic to it. There's the top lane and the bottom lane and the middle. There's no defined 'center' like there would be on Python or Luna. Second, the mains are really far away from each other on Heartbreak Ridge. Compare this to a map like python, if you're 6 vs 3, the naturals are actually really really close in terms of walking distance. Moreover, on Heartbreak Ridge the distances seem even longer between the bases, because there's that um, I call them spokes, it's the little piece or ridge that goes up and then down and then up and then down, those little spokes coming off the middle make it even harder to get to your opponent's base because if there're lurkers up there it delays it even more, even mutalisks harassing from on the top of those spokes can be really hard for marines to move across, so the distances between the bases is very long. The last and perhaps most important point about heartbreak ridge is that expansions ore really difficult to take. in particular, Zerg has a really hard time taking a fourth gas in any matchup, especially in Zerg vs Terran. And even if you're playing Terran, it can be a little tricky and awkward to take that third gas that's right beside your natural. The first conclusion you should reach form this is Zerg is going to be very scary and very comfortable going three gas lurker/ling/defiler. Because there is the circular dynamic in the middle, Zerg can throw down swarms to hold off whatever push you're making in one lane, and then counter attack with lurkers and lings in the other lane. Moreover, because the lanes themselves are fairly narrow, Zerg can actually push Terran back. On a map like python when Zerg throws down a few swarms the Terran can go around. Uh, this isn't possible on heartbreak ridge. So that lurker/ling/defiler combo is very intimidating in the midgame. The second major conclusion you should reach is that Zerg is going to have a very hard time getting a fourth gas to either get a bigger army or transition to ultralisk or do some other late-game plan.

Now that we've analyzed Heartbreak Ridge a good amount, it's time to put on our Terran thinking hats and decide what our midgame plan is going to be vs Zerg. The first and most reasonable thing to do is just to say, 'is playing standard any good on this map?'. Well, let's think about it. The standard Terran vs Zerg midgame involves pushing out with a pretty decent group of medic marine, three or four tanks, and one vessel. The timing of this push is decided by when that first vessel pops out. Now, in your experience playing Terran, or even watching Terran vs Zerg, you should know that Zerg is always trying to get defilers out right when that first push hits his front. So from the Zerg's point of view, he's harassing, he's positioning lurkers, he's trying to delay this so that way Zerg can get defilers, throw down swarms, and push the Terran back. On Heartbreak Ridge, when Zerg gets these defilers out, we already know that Zerg is going to be in a really good really scary position. Are we suited to deal with that if we do a standard opening? Probably not, we have some tanks that aren't going to be as effective anymore because there's two lanes to defend, and more importantly we don't really have a very high vessel count to deal with an aggressive lurker/ling/defiler push. The only real way to make playing standard super-effective is if you do some little twist early on, if you move out very early with your marine and medic group to try to mess with Zerg before he gets going or something like that. So now we need to consider some alternatives. A first very reasonable one is to think, 'Hey, maybe I should just cripple Zerg before he ever gets his scary midgame army going.' As I mentioned earlier, it's a little awkward to defend that third gas, even for a Zerg player. So, I've seen Terran players get five or six barracks even before they get their factory. They end up with this unbelievably huge marine and medic army and they just pin Zerg down with so much, then the idea would be to make quick tech up to vessels after that. Obviously there's going to be some issues, for instance if you don't cripple Zerg, if Zerg's very good with defending and positioning himself, you'll end up with even fewer vessels against those defilers than if you had played standard.

However, the alternative I really want to spend a long time talking about is Fantasy's play against Jaedong in the Bacchus OSL semifinals game one. I absolutely loved Fantasy's midgame structure, for three reasons. First of all, fantasy's midgame was designed to be very very effective against that three base lurker/ling/defiler that's oh-so-effective on Heartbreak Ridge. Number 2, the way that Fantasy combated it took advantage of the fact that Zerg has a hard time taking a fourth gas. And three, the transitions leading up to the midgame were absolutely beautifully planned. Fantasy's midgame plan revolved around getting elements of lategame Terran play very very quickly-- in particular, based on getting a lot of vessels and vultures with mines very very quickly. by having these two additions to a very big army, Fantasy was able to combat excessive defilers with all those vessels, and Fantasy was able to put mines along the top and bottom lane to prevent any sort of counter attack. The obvious danger to devoting time and money to getting those late game things is that you're slightly weaker in the early and midgame, you can't quite put a lot of pressure on Zerg. But, that's okay on Heartbreak Ridge, because defending a fourth is so hard for Zerg anyways. Zerg is pretty much committed to doing a three base strategy for a fairly long time. and the last point about the transitions you'll hear all about as I go through the game from start to finish. The most important thing to note in this game is that fantasy went for a starport very very quickly. For a long time in the game he actually had one barracks, one fact, and one starport, each makin' stuff. The clever part is getting to that situation without dying. So first of all, Fantasy opens up with a very very standard Terran opening, which is to go straight for a factory, just do a one barracks fast gas, and get a vulture out and use the vulture to do a little bit of harassment, but primarily to expand and defend the expansion from zerglings. After expanding and poking around with that vulture a little bit, Fantasy gets an addon on the existing factory, and begins building a starport. At the addon, fantasy upgrades mines and continues to make a handful more vultures. This continues along the idea of making sure the um, the expo is well defended. Fantasy now can plant mines at expansions, plant mines at the top and bottom lane, and get a little bit more control over Zerg. Now, here comes some serious brilliance. Fantasy was doing very ordinary stuff given what I've described his midgame to be. He was going for science vessels, halting vulture production at the factory and starting to build tanks. He was starting to build up marines and medics, but Fantasy also threw in a dropship to do vulture harass with those vultures he had been making early on. This is so brilliant because notice how it does not disrupt anything else that Fantasy is trying to do. It flows in seamlessly. While the science facility is building, so Fantasy can get vessels, Fantasy is able to build the dropship. It doesn't delay the time of anything. moreover, Fantasy already had those vultures out because he used them to defend his expo early. He didn't suddenly have to build some vultures and suddenly have to stop building science vessels to get the dropship. Everything was already there and already in place. That is perhaps the key difference between extremely good players planning transitions and extremely bad players. Bad players tend to just jam something in because it looked cool and looked effective in the game, and they don't put any thought into the overall interaction with everything else that's going on. At this point, Fantasy was able to do a decent amount of damage with those vultures. Meanwhile, back home he was adding on a whole bunch more barracks and was already producing tanks and science vessels out of that fact and starport that were build so early on. When Fantasy does finally push out, he has three vessels instead of one, and six tanks instead of three to four. Note that Jaedong was able to hold off this push, but Fantasy's play was already in full lategame swing. Vultures started planting more mines all over the place, that vessel count that was already big continued to get bigger, and Fantasy eventually wore Jaedong down, despite the fact that Jaedong was in that very comfortable, very scary three base lurker/ling/defiler mode. Fantasy was able to wear Jaedong down because of that fourth gas problem. Jaedong was having a really hard time taking a fourth gas given the structure of Heartbreak Ridge, and given Fantasy's midgame plan.

Now that was all fun talking about Fantasy and Jaedong's game, but what I really wanna do is stop for a moment and think about how you would have been able to formulate Fantasy's gameplan exactly on your own, because it's one thing to be able to look at other players' games and say 'Okay, I get what's going on there', but it's so much more useful to be able to stop and go 'Oh, I see what his thought process was, I understand his mental tools now, and I'm gonna use those to develop my own style and my own sort of play, so that way I can conquer maps that are not Heartbreak Ridge.' To do this, let's strip all the complicated stuff away from the midgame, and boil it down to our two basics wants. We want vultures with mines, and we want lots of science vessels for our midgame. Let's think about those vultures first. typically in Terran vs Zerg you start getting vultures with mines after you stop making tanks out of your factory. We've already discussed the Terran standard midgame on Heartbreak Ridge, and we decided that that wouldn't be quite as effective, that we're getting those mines a little bit too late. So, right now we have already figured out that we need too be getting vultures with mines early on. But we don't want to just jam vultures into our early game plan without thinking about how vultures interact with everything else. For instance, here's a really really bad way to go vultures-- you have opened up with an early expand. You have three barracks making marines and medics and you've been holding off muta harass for just a little bit of time. Suddenly, you add on a factory and start making vultures with mines. First of all, vultures without speed at that stage of the game are going to have a really difficult time leaving your base with mutalisks planted all around. Moreover, the Zerg has a pretty good bit of map control so it's gonna be difficult to be able to plant mines anywhere you want to without him being able to just easily walk an overlord over there and take it out. So the only potential need those vultures could satisfy is keeping your front more defended from attacks with mines. However, you already have three barracks with marines and medics, likely a bunker at your front, you're already defended. The vultures, in a sense, are satisfying nothing that you need it to. A good way to think about how to put vultures into play or to think about some need that you have early on that some other units satisfy, and then to try to see if vultures can swap out for those units. For instance, we make a handful of marines and occasionally a bunker to defend an early expansion as Terran. Why not make a vulture so instead you can hold off any zerglings early on. That's exactly what Fantasy did in this game. He used the vulture to defend himself early on. But we've also said that we need to get mines early too. Well, we know that Zerg can be seriously threatening in the early game, so if we just get a few more vultures from that factory we've already built, then we can also get mines to hold off any sort of bigger attack, for instance lay down mines in case we're worried he's doing a ling allin, plus getting mines this early helps us control expansions we know exactly when the Zerg is intending to take an expansion. What we've done is listed some needs, some goals we have, and the vultures and mines are satisfying that early on already. With our front fairly comfortably defended, we can start getting science vessels right now, and then we can worry about adding on more barracks later. Already it seems like we've satisfied our midgame goals, we have those mines and vultures already in place for when we need to start producing them in the lategame, we also haven't made vultures in a way that screws anything up, and we're happily on our way to lots of vessels. So lets say you've done a good bit of thinking by yourself and you've decided that this is the build you want to do. You should go and start testing it and what'll happen is you'll notice a few things in that early and midgame stage. The first thing that you would probably notice is those vultures suddenly start to lose use. Not the mines, but the vultures themselves, they can't get in past the sunken colonies, Zerg probably has some muta around, that would kill them off very easily, it feels like the vultures are just hangin' out for a little bit of time. Moreover, you should also start to notice that it feels like you're not putting much pressure on Zerg. Normally you would have three or four barracks making medic marine, and you can always have the threat of attacking Zerg early on. But suddenly you have one barracks, one factory, and one starport, and zerg is pretty much free to power as much as he wants. The immediate next step of logic is to say 'Hey, can these useless vultures right now put any extra pressure on Zerg?' The answer is, generally, not because we've already said that there are sunken colonies and mutalisks everywhere. Because generally we don't have a starport when we can get a dropship. However, we're already on our way to a starport to get science vessels, and if we get that dropship, it doesn't slow anything down. All of a sudden, everything is starting to link together so these vultures that were useful early on are now useful in the midgame for harassment, and are useful right after the midgame because they have mines to hold off those defilers. That starport that was only useful right when the science facility was made suddenly is useful on the way there because we can make a dropship to it. Everything has perfect efficiency from the moment it's construed. It's not some temporary holder until you can get your strategy going. Nothing I've said takes a genius to figure out. All I did was said, 'Here's what I wanna do in the midgame', and then I said 'Hmm, here's how I wanna put vultures in my game so that they're actually useful, and here's how I want to get the science vessels out so that they're actually useful. And hmm, I have a few problems now. How do I solve them given my current setup?' And by solving them, I pretty much have described fantasy's play exactly. Totally frickin' awesome. Now, what I want you to do is to look at a new map that you've maybe been having trouble on, or a new map you're excited to play. Think about what would be really good in the midgame and then all your decisions afterwords are guided. All your decisions know that they're trying to flow into your midgame plan, and all the problems that start arising have very natural, logical solutions. And in the end, you will be a better Terran player.

Before I end this, I want to briefly touch on a question I know tons of people are going to ask. Their question is going to be, 'Well gee, why wouldn't I use Fantasy's build on a map like python?' My immediate response is 'Go ahead, try it, absolutely, there's no reason that you should dismiss a strategy on a certain map because you can't figure out a perfect way to play it.' What will probably end up happening though is because you're putting so little pressure on Zerg early on, Zerg will have a much easier time taking his fourth gas, because on python it is pretty easy to take a fourth gas on that map. So though your army will come out really strong with six tanks and three vessels and a bunch of marines and medics, Zerg will still be getting defilers as quickly as he normally would have, but he will also have that fourth gas under his belt. So, that feels like the danger of doing that Fantasy play on a map where it's easier to get a fourth gas. In a sense, that's exactly why um, standard Terran play is so effective on python. Terran can ramp the aggression up really early on with that first Terran push, and now Zerg is struggling to take his fourth gas once again. And Zerg doesn't have a lot of good counter-attack paths because it's a nice wide open center that Terran can easily control after he takes his midgame army. So, yeah! Hopefully I've given you some nice new tools to analyze games. If you're still struggling to look for a starting point, I'd recommend going back and looking at the Jaedong vs Fantasy game and thinking 'How would fantasy have responded if Jaedong had done 2 hatch muta? 3 hatch muta?' and start going through more, um, unorthodox strategies even, and think about how Fantasy's strategy would be flexible enough to deal with it. And with that, I end my first Terran exclusive podcast. Thanks for listening everybody, cheers!