StarCraft II

[Megrim]

Links to replays plx.

Replays from pro-level games aren't typically available, but we should be able to get vods soon.

[bhodi]

So, I played in a tourney this weekend. Made it past the qualifiers, lost in the first round.

This is what a 1700+ diamond player looks like. I fought him back and forth but his macro is clearly superior. Watching the replay, it's painful to see the amount of drones I missed and getting briefly supply blocked mid-game.

I can, of course, play better than that; here's one of my qualifiers and a near-flawless 2gate robo (excepting the sub-second nexus drone delays). Note that I still only play at ~60apm and can fairly easily smash a 1100 diamond player, at least moving into the midgame.

The two builds are functionally the same, only in the top one I ended up having 2-3 second delays on probe production multiple times, putting me ~10 probes down by 15 minutes in and ultimately losing the game because of it. It's pretty amazing once you get into higher levels how much little mistakes cost.

[Mosesandstick]

The I love you would have really put me off.

[Ironwood]

I do like all the new coloured icons.

[Malakili]

So, I played in a tourney this weekend. Made it past the qualifiers, lost in the first round.

This is what a 1700+ diamond player looks like. I fought him back and forth but his macro is clearly superior. Watching the replay, it's painful to see the amount of drones I missed and getting briefly supply blocked mid-game.

I can, of course, play better than that; here's one of my qualifiers and a near-flawless 1gate robo (excepting the sub-second nexus drone delays). Note that I still only play at ~60apm and can fairly easily smash a 1100 diamond player, at least moving into the midgame.

The two builds are functionally the same, only in the top one I ended up having 2-3 second delays on probe production multiple times, putting me ~10 probes down by 15 minutes in and ultimately losing the game because of it. It's pretty amazing once you get into higher levels how much little mistakes cost.

Will watch the replays later, but I have to say its promising that you are able to play at that level with ~60apm.  That is roughly where I  play at and I am currently high platinum (wondering if I'll get moved up soon actually, because I've been matched up almost exclusively against diamond players lately, with a reasonable even record of wins and losses, but I digress).  That gives me some nice motivation to keep refining my play and improving on my builds without stressing so much over hand speed, at least at my current level.  Obviously it eventually will become necessary if I really want to break into higher levels of play (which may never happen, heh), but I am happy there is still lots of room for improvement with my current physical skill.

[bhodi]

The I love you would have really put me off.

There needs to be more love in the world.


My average is 60apm, I do spike to 150-200 when "stuff" is going on. I just don't spam buttons for the hell of it.

[Malakili]

My average is 60apm, I do spike to 150-200 when "stuff" is going on. I just don't spam buttons for the hell of it.

I am mostly similar in that regard, though there is something to be said for maintaining 100+ at least.  Players who are constantly able to keep their units positioned well, continue macroing, etc are not just spamming to keep their APM high, it actually pays off even when its not micro in battle.  Still, there are a lot of spammers out there who think APM is good in and of itself.

[bhodi]

I am not competitive level, nor do I intend to be. I just know some build orders and have enough practice to be decent at most matchups.

[bhodi]

Links to replays plx.

They're all on this guy's stream. Here's the first one. Watch them before they're yanked!

[Megrim]

Thank you muchly.

[Fordel]

Watching the games again, it seems like the Terran was just completely inflexible; while the Zerg was the exact opposite, bending his play perfectly every time.


I know it's easy to say this in hindsight, but how many failed tank/marine drops do you need to clue you in "maybe I should try something else?". He couldn't get those early harass drops off, which just then seemed to cascade all the way down his entire game plan.

[Margalis]

Watching the games again, it seems like the Terran was just completely inflexible; while the Zerg was the exact opposite, bending his play perfectly every time.

I don't even know if you can call it bending as much as straight up exactly knowing what his opponent was going to do well in advance. It wasn't reactive, it was predictive. For example on Kulas he had the OL positioned, the spinecrawlers in place and the Roach Warren up well before there was any indication that a drop might happen. He just knew what was going to transpire.

Now on Kulas and Lost Temple that's pretty predictable, but he had the timing and defense perfect. When I play on those maps I lose to those same tactics, even when I know they're coming.

[Malakili]

Watching the games again, it seems like the Terran was just completely inflexible; while the Zerg was the exact opposite, bending his play perfectly every time.


I know it's easy to say this in hindsight, but how many failed tank/marine drops do you need to clue you in "maybe I should try something else?". He couldn't get those early harass drops off, which just then seemed to cascade all the way down his entire game plan.

In a tournament you are 100% playing your strategy and losing than you are trying to totally wing something.  I'm not saying don't make choices by what your opponent is doing, but if you have a strategy, stick with it.  Heck, I just watched the 2 hour long day9 daily number 100 where he talks about his life in Starcraft, and this is something that came up several times.

[sinij]

I can, of course, play better than that; here's one of my qualifiers and a near-flawless 1gate robo (excepting the sub-second nexus drone delays). Note that I still only play at ~60apm and can fairly easily smash a 1100 diamond player, at least moving into the midgame.

Can you explain to me how this can possibly work against 1 base zealot/stalker play? In the replay opponent got greedy and went for expand, if not for that you would have been drowned in zealots making your immortals nearly useless.

[Margalis]

In a tournament you are 100% playing your strategy and losing than you are trying to totally wing something.  I'm not saying don't make choices by what your opponent is doing, but if you have a strategy, stick with it.  Heck, I just watched the 2 hour long day9 daily number 100 where he talks about his life in Starcraft, and this is something that came up several times.

Day9 is a Starcraft scrub compared to Korean pros. "Stick with your obviously losing strategy" is horrible advice for anything, especially given the one game ITR won was the one where he went for a push/contain while his opponent was still on a small number of bases rather than a cute harass.

[Muffled]

Just to pick a nit, I think Fruitseller deciding to try to directly kill tanks with banelings had a lot more to do with HopeTorture (ugh that name) winning that one game than anything else.

[bhodi]

Can you explain to me how this can possibly work against 1 base zealot/stalker play? In the replay opponent got greedy and went for expand, if not for that you would have been drowned in zealots making your immortals nearly useless.

Sure. I assume when you say '1 base zealot/stalker play' you actually mean a 4gate, so I'll go with that. I also totally lied, I just watched that replay and I actually did a 2gate robo. I was confusing it with the other replay against terran where I did 1gate robo because I saw he was going heavy marauders. The reasoning behind what I did in this game is a number of things all put together so I'll try to explain as best I can without making it too convoluted. Let me talk about this specific instance and then I'll talk more about 4gates in general afterwards.

First, map specifics. The two important things to know about this particular map and our matchup (PvP) early game is that the map is gigantic, there is decent choke going into the natural and also the main, and you're cross spawned. For early game, this means a long rush distance (even with the ramp cutins) and lots of places to proxy.

When I scouted him at 3:40, I knew he went 13 gate to cyber just like I did. No funny business. If he had done a proxy 2gate, I'd probably have had a hard time with it, but the zealots would have already been in my base (or nearly so) by then. To help with this, I always go for the early zealot pre-stalker instead of an early 2nd gateway, so I wouldn't have instantly died, though I'd have probably had a hard time of it and it'd come down to who's micro is better.

The biggest thing is what I scouted at 4:35. See that pylon up in the upper left that went down right as my probe died? There's only one reason to put a pylon in such an odd position - he's trying to hide something. That something is almost always a stargate. When it's not a stargate, it's a dark shrine, or in later game a robo bay/templar archives. In this case, it was way too early for anything but a stargate. At any rate, now I'm forced to check, because an odd pylon is only a suspicion, so I build a robo instead of a 3rd gate. I immediately build an observer and then an immortal (so the building isn't wasted) and I'm on my way over (first heading north to check for proxy pylons) when i see his first push. He's got the correct number of units for a 2gate early aggression, so I again know there's no oddities going on.

At this point I elected to pop his force before he got that pylon up. In this circumstance, I put my immortal forward to absorb the brunt of the force, waited until he was stationary (and his attention was elsewhere for a second) then pounced. As a result, I was able to close before he he could snap back and thus I drove him off, killing the pylon. I follow him back to his base, see he's expanded and that he's extremely stalker heavy (so he's not going void rays, I was a bit confused by this, I had assumed the phoenix was an illusion), but at any rate I come in from the bottom, force field, GG.


As to fighting 4gate directly, if he had done an ACTUAL 4gate, I'd have never built the robo - I'd have been forced to cannon up and/or build another 2 gates of my own to match. If I had missed it, or if we're talking in abstract, 4gate is always going to beat a 1gate robo, and might beat a 2gate robo. Generally, the answer of 'how to beat it' is to force field your entrance until you can get more immortals. Even with heavy zealots, immortals TEAR through stalkers, especially in numbers. In general, it's down to your micro. 4gate is pretty tough to beat right now, but it's also insanely popular and easy to see coming. You've got a shot if you haven't expanded it and can FF your ramp/entrance.

Combat when you have immortals against another protoss with Tier1 push should be 1. Attack move into his army, pray he doesn't have chargelots. 2. Select your sentry, guardian shield. 3. select your immortals, shift right click each stalker in his army. 4. force field his zealots off your guys if he's got way too many, hope your immortals can whittle down his stalkers. 5. Retreat/kite the remaining zealots.

[bhodi]

In a tournament you are 100% playing your strategy and losing than you are trying to totally wing something.  I'm not saying don't make choices by what your opponent is doing, but if you have a strategy, stick with it.  Heck, I just watched the 2 hour long day9 daily number 100 where he talks about his life in Starcraft, and this is something that came up several times.

Day9 is a Starcraft scrub compared to Korean pros. "Stick with your obviously losing strategy" is horrible advice for anything, especially given the one game ITR won was the one where he went for a push/contain while his opponent was still on a small number of bases rather than a cute harass.

Yes, it's very easy to get tunnel vision and you should be flexible based on an evolving situation, but you have also practiced a build for a reason. Trying to deviate from what you've spent weeks and weeks practicing and just trying to 'wing it' can lead to disaster - especially in the higher ranks of play. When you have memorized standard timings for expansion, pushes, counters, and transitions, deciding to throw it all out the window while you're already in a position of weakness due to some unforeseen surprise is probably going to lead to a loss.

Good players have extremely flexible builds that easily transition into common counters, and often their 'response' to what you might see as a surprise attack is simply a branch of play that they have planned for and can easily transition to counter (see: mutas, banshees, fast expand contain).

Also, Day9 is not giving advice to Korean Pros, he's giving advice to average Joes. A well executed inferior strategy will roll over a poorly executed superior one every single time. See the episode "there is no such thing as a hard counter" for examples of this. It's INCREDIBLY easy to panic and forget basic things like worker and unit production when you are trying to mentally figure out what to build and how to counter the nasty curveball your opponent just threw at you. Hell, look at game 1 of the finals for examples of this. HopeTorture fell apart and was floating 1k+ minerals even while FruitSeller was still pushing in on his base. He fell apart and he's a PRO. How do you think you'll do?

If you want to get a better idea about the decision making, here's a decent stream - this is the finals of the tourney I was in. Listen as he explains the mentality/methodology and what the players are thinking. This is a lot more in depth than a normal cast, and you may not have heard it's like before.


Edit: Oh look, day9 just said this exact same thing 3 minutes into the daily #192. Watch that for a more detailed explanation of what a 'build' is. It's not 10 depot 13 rax.

[Margalis]

Edit: Meh, who cares.

[sinij]

Appreciate response. Will ask more questions after I digest it.

Have another question - I am in AT Z+P and we finally moved to solid Diamond, so it no longer matches us with plat or gold. Personally we are High-Gold, Low-Plat players, but with vent and shared control we can out-micro most players. Problem I have now is that games went very predictable and linear - it is 100% about tweaking econ and nobody attempts anything but pump-pump-pump fights. My personal 1v1 play style is pure aggression, if I ever get into late game I don't feel I was playing well. Do you have any tips on how to spice it up? I am literally falling asleep (and my game slips) after 5-6 30-minute pumping matches.

[sinij]

shift right click each stalker in his army

Now I feel dumb, this never occurred to me... I usually babysit my units and focus-fire them via control groups... one target a time.

[bhodi]

Do you have any tips on how to spice it up? I am literally falling asleep (and my game slips) after 5-6 30-minute pumping matches.

For the tired, there's nothing you can do except take breaks; playing games are draining. There are always things you can do to break the 2v2 tier1 pushathon, though. The snarky answer is just to do a Z+T zergling/reaper which is 100% unstoppable right now.

Drops and distractions are key; even the best people fall apart under pressure, so while your teammate is coordinating defenses, do drops and harassment timed for when they attack. You'll end up destroying a lot of workers. Share resources and control. Have the zerg give all gas to you, and rush for motherfucking carriers! Switch off control of your army and have only one of you control it and coordinate defenses. Go mass zealot+cannons, have the zerg fast expand and give all gas to the zerg for early mass muta off of 2 base queens! There are plenty of variations that do plenty well against a coordinated 2v2 t1 push. Expect good (?) things to come in the next patch, since blizzard has said they'll be focusing on 2s balance.

[sinij]

The snarky answer is just to do a Z+T zergling/reaper which is 100% unstoppable right now.

Our answer to that - static D for P ( 13 forge) and early pool 2x queens + sunken for Z. Don't really have a problem with it at all, esp when reapers are proxied and they are naked to fast counter-push. Works best on shared choke maps, but doable with complete wall-off for P on other maps. Don't try early stalkers, you won't get enough in time and your partner can't save you in time before you take too much econ damage.

What is really unstoppable is 2x 8-pool on a wide choke map. If you are good at kiting workers with lings they are Fucked and will never mine. We used to play Z+Random and any game we got ZZ it was auto-win.

[sinij]

As to my personal outlook on SC2 play - don't try odd strats you see in world-class players vids. They had to practice to pull it off, there is no chance you, Joe-average can wing it. Stick to basics - focus on econ, cheese-proof your base (use static D) and pump out army you can control. If you are 20APM A-move Silver player trying to go with micro-intensive build and/or harass is counter-productive.  While you are trying to make it work your economy will fall behind a lot more that damage you can inflict.

Know your limitations, practice & stick to the plan and you will do a lot better than otherwise possible.

[Malakili]

As to my personal outlook on SC2 play - don't try odd strats you see in world-class players vids. They had to practice to pull it off, there is no chance you, Joe-average can wing it. Stick to basics - focus on econ, cheese-proof your base (use static D) and pump out army you can control. If you are 20APM A-move Silver player trying to go with micro-intensive build and/or harass is counter-productive.  While you are trying to make it work your economy will fall behind a lot more that damage you can inflict.

Know your limitations, practice & stick to the plan and you will do a lot better than otherwise possible.

This is very true.  After I started trying to play with tons of drops and harass and others stuff that protoss isn't especially great at and required lots of micro, I had a lot of trouble.  I went back to playing very simple macro and expand strategies and getting them down pat, and now I've gotten a lot better.  Better enough in fact that I'll be incorporating some of that stuff back in again soon I think. 

[sinij]

Ugly truth is that difference between Plat and Diamond is 2-3 extra workers per fight. It is *all* about econ. You can compensate in other areas, but you don't have to - if you have stronger econ you can crush your opponent by A-moving your army in 9/10 situations.

[Samwise]

I'm only further convinced that I never want to play SC2 competitively, which probably means avoiding multiplayer in general.  When it's described as practicing and executing basic strategies and keeping APM up and all that it just sounds incredibly boring.

[bhodi]

Nah. You can have fun with it. 4v4 is where it's at. Get a few t1 for defense and then



It's so much fun. Also, I love custom maps.

[Samwise]

Maybe it's just 1v1 I need to avoid.  I guess most of my fun in SC1 was 4v4s and 8-way FFAs and stupid shit like that anyway.

[Malakili]

I'm only further convinced that I never want to play SC2 competitively, which probably means avoiding multiplayer in general.  When it's described as practicing and executing basic strategies and keeping APM up and all that it just sounds incredibly boring.

Thats what you do normally playing the game, this is just...acknowledging it and trying to get better at it.  I don't know why you'd be upset about "executing basic strategies" in a strategy game?

I mean, if you don't like multiplayer RTS, then you just don't, i'm not going to try and convince you, but I think you are looking at it the wrong way.

[K9]

I'm only further convinced that I never want to play SC2 competitively, which probably means avoiding multiplayer in general.  When it's described as practicing and executing basic strategies and keeping APM up and all that it just sounds incredibly boring.

I haven't played in Platinum or Diamond, but I'm feeling that Gold is a really nice sweet spot. Bronze has a lot of goofy cheese and flat-out terribad players (beating someone who only has 7 SCVs for the whole game is not really satisfying). Siver is a bit better I guess. Gold seems to be the level where you get adequately challenged without having to start getting too crazy. Perhaps it is a gradient and if I was in Diamond I'd be happy, but I think you can find your sweet spot on the ladder and providing you're not someone who feels put out by not being "the best" you can have a lot of fun.

Also, Income Wars is lots of (silly) fun

[Samwise]

I'm only further convinced that I never want to play SC2 competitively, which probably means avoiding multiplayer in general.  When it's described as practicing and executing basic strategies and keeping APM up and all that it just sounds incredibly boring.

Thats what you do normally playing the game, this is just...acknowledging it and trying to get better at it.  I don't know why you'd be upset about "executing basic strategies" in a strategy game?

For me the fun of a strategy game is in coming up with a strategy, not executing it.  If all you're doing is executing someone else's well-worn map to victory you're not really playing a strategy game -- THEY'RE playing a strategy game and you're playing a middle management game.  

[sinij]

bhodi can you explain to me how timed colossi push works?

[Malakili]

For me the fun of a strategy game is in coming up with a strategy, not executing it.  If all you're doing is executing someone else's well-worn map to victory you're not really playing a strategy game -- THEY'RE playing a strategy game and you're playing a middle management game.  

Well, there is plenty of room for that, but you know...its also smart to build on other ideas that have worked in the past.  Its not like you just execute click for click someone elses strategy, but there is a reason a "13 gate, 17 core" and so forth, work.  It emerges from smart play (always be building workers, never get yourself supply blocked, etc).  Its not just some "oh, this smart player said do it so I do it"  It can be boiled down to supply numbers and such near the beginning just for the sake of simplicity, but the game is far more fluid than that.

[slog]

You can do lots of fun things in 1v1 if you practice them.  For example, Zerg baneling drops.  I played a couple games against the AI to get this down, and it really comes down to these steps.

1) Train the skills
2) Make 15 banelings ish.
3) Load Banelings in Overlord.
4) Shift queue the banelings to drop in the enemy base and queue your overlords to come home.

Step 4 was the part that took practice.  The game allows you to queue up everything you need to do up front.  I will typically, set up a route for them to use the edge of the map, then Shift-D and shift-click a bunch of times where I want them to drop (it's better, but not required if you do this last part for each individual overlord.)

I can then completely forget about the drop and concentrate on other things.  I imagine it must look like I am a multitasking master to my opponent sometimes. :)