The Walrus Blog

Rock, Paper, Starcrafts

From hand-to-hand combat to advanced battle strategy, it’s all in the metagame

From hand-to-hand combat to advanced battle strategy, victory is all in the metagame

Illustration by David RusakDavid Rusak

VIZZINI sits at a covered table, on which rests a bottle of wine and two goblets; BUTTERCUP, his blindfolded captive, is to his left. The MAN IN BLACK, seated opposite, has just added a fatal poison to one of the cups — unseen by VIZZINI, who now must choose which drink to consume. The survivor of this contest will claim BUTTERCUP as his prize.

MAN IN BLACK: All right. Where is the poison? The battle of wits has begun. It ends when you decide and we both drink, and find out who is right… and who is dead.

VIZZINI: But it’s so simple. All I have to do is divine from what I know of you. Are you the sort of man who would put the poison into his own goblet, or his enemy’s? Now, a clever man would put the poison into his own goblet, because he would know that only a great fool would reach for what he was given. I am not a great fool, so I can clearly not choose the wine in front of you. But you must have known I was not a great fool. You would have counted on it, so I can clearly not choose the wine in front of me.

MAN IN BLACK: You’ve made your decision then?

VIZZINI: Not remotely. Because iocaine comes from Australia, as everyone knows. And Australia is entirely peopled with criminals. And criminals are used to having people not trust them, as you are not trusted by me. So I can clearly not choose the wine in front of you.

MAN IN BLACK: Truly, you have a dizzying intellect.

VIZZINI: Wait till I get going! Where was I?

My game begins, as it always does, with a central building and a handful of workers. With them gathering minerals and more workers on the way, I scout out my enemy’s location — he is playing as the gooey, alien Zerg — while developing my base and building an army. After a few minutes, my first wave of troops — a small squad of rocket-firing heavy infantry called marauders — is ready to go and, fearing the loss of the initiative, I venture out to invade my opponent. As I reach his base, a stream of zerglings arrive, already upgraded to run at high speed, and instantly swarm my fighters, trapping and then clawing them apart before dashing off to demolish my base. Moments later, they tear through my protective wall, the couple of new soldiers I’ve managed to pump out behind it, and, finally, my precious resource gatherers. With a dejected “gg,” I surrender and go straight to searching for a new match, eager to try new approaches against the next random opponent.

StarCraft II came out this July and, like its predecessor StarCraft (the most popular real-time strategy game of all time, released in 1998 and still played in global tournaments by star players who earn six-figure salaries), features battles to the digital death among three races: Terran, Protoss, and Zerg, each with its own unique selection of units, buildings, and technologies. Was I a fool for choosing those slow and swarmable marauders? Not necessarily. If my opponent had fielded (as many do) the Zerg’s tougher, acid-spitting roaches instead, my armour-piercing rockets would have devastated them. I could have anticipated his zerglings and made hellions, light vehicles with flamethrowers that roast whole rows of the critters alive, but these in turn would have stood no chance against roaches.

One of the most trumpeted virtues of the original SC, and the goal of its still-developing sequel, is near-perfect game balance. For every strategy, there must be an effective counter; each race must be able to succeed or fail against any of the others; each unit must be vulnerable to some other. If any one turns out to be dominant, the onus is on Blizzard, the maker of the games, to “nerf” (i.e., worsen) the offending element in the next update. Since every move in SC ideally has a countering response, the success of your strategy always hangs on that chosen by your opponent. In this respect, it is essentially an elaborate version of a still more classic game: Rock, Paper, Scissors.

That may sound like a bad thing. RPS is very simple: there is no window dressing, just three equivalent moves, and the optimal strategy, according to game theory, is to randomly play each of rock, paper, and scissors one third of the time. However, game theory deals only with “ideally rational” players; as has been noted on this blog before, the involvement of real human beings changes everything. When we play a game like RPS, we impart a messy landscape of feelings and expectations to it that have little to do with the game’s pure, symmetrical form. If my opponent’s scissors beat my paper last round, will she be caught off guard when I use paper again? Or will she expect this sneakiness and counter with scissors, necessitating rock? Rock is often characterized as aggressive, scissors crafty, and paper subtle or refined; these characters are wholly baseless, but, like the guess based on previous rounds, provide an arbitrary starting point for strategizing (“I need to get aggressive; should I throw rock? Or scissors to counter the paper that counters rock? Or…”). In a strategic situation such as this, the two players’ expectations of one another’s expectations feed back infinitely, like a pair of facing mirrors.

Humans are both notoriously terrible at randomizing (especially under pressure) and excellent at detecting patterns (like those underlying our crappy randomization). So, as any skilled poker player knows, you can learn to intuit the order that underlies even your opponent’s most deliberately unpredictable behaviours. Throwing with perfect randomness would always guarantee you a 50-percent chance of winning or losing at RPS; but if you can predict your opponent’s actions, you can do much better than that.

The gaming world has a term for this crucial well of information which lies not within the mechanics of a game but in the people who play it: the metagame. “Metagame” refers to the tendencies of the whole community (e.g., “The Zerg metagame is all mass roaches now, so I always make some marauders”), but also to using knowledge of individual players (e.g. “He pumps zerglings a lot, so I’m going to metagame and go straight for hellions”). The metagame of SC (and other competitive strategy games, from the tectonic shifting of chess to the frenetically refreshed Magic: The Gathering) is constantly changing and developing in a sort of game-wide Brownian motion, as new strategies are invented to one-up the previously dominant ones. This is made all the more interesting by the many levels on which RPS-like dynamics exist in SC. A player who builds up both army and economy at an average rate will be able to repel and then beat a “rushing” player who makes the quickest possible army at the expense of long-term production; a player who immediately focuses on building up her economy will be able to field a larger army later on than a player with a balanced strategy; and that economy-centric approach is in turn easily taken out with a rush attack. As described above, different unit types also counter and are countered by one another in an analogous way. And each of the nine race matchups one can play (as Terran against Terran, Zerg, or Protoss, etc.) demands a different approach according to the different races’ unique strengths and weaknesses. The immense sophistication of the metagame among top players is evident in this history of the strategies developed for just one of those matchups — Terran vs. Protoss — in the original SC. The picture forming is of a game that consists of many nested and iterated RPS-like battles, the nuances of which can be felt keenly by watching the back-and-forth exchange of a well-matched expert game.

Strategy is only part of the formula, though; SC is still a computer game, and the vast majority of learning a novice player faces has to do with gaining a command of its particular nuts and bolts. One popular measure of sheer game skill is a player’s “actions per minute” (APM); casual players reportedly perform somewhere between 50-100 APM, whereas pros can range into the freakishly high 300-plus area — which has nothing to do with strategic thinking and everything to do with being able to effortlessly handle the constant micro- and macro-management required by SC. Similarly, studies in psychology have shown that expert chess players’ exceptional ability to remember board positions is not due to generally superior memory skills, but rather to familiarity with typical chess positions. When faced with a board in which pieces are placed completely randomly, chess masters fare no better than novices at remembering where they were.

While encyclopedic knowledge of the tech trees and build orders of SC provides zilch by way of transfer to other pursuits, the strategic intuitions of strategy game players doubtless are transferable — if the success of competitive SC and Magic players in poker tournaments is any way to judge. What these games all have in common is their strategic core: the metagame of psychological second-, third-, and nth-guessing that underlies any situation in which your choices depend on your opponent’s and vice versa. Strategic insight is insight into every such situation, which is precisely why it has been popular to search books like The Art of War (explicitly about war) or The Book of Five Rings (explicitly about martial combat) for wisdom in areas of life from games to business to intimate relationships.

Which brings us back to Rock, Paper, Scissors: a pure — perhaps perfect — instantiation of the strategic encounter, virtually untainted by considerations of game-specific skill (admittedly, some manual dexterity required). So, the next time you throw down Rock, Paper, or Scissors for the front seat or the last bit of food, just remember: though the game you are playing seems trivial, the metagame it taps into is deep, ancient, and universal. And if your opponent plays a lot of StarCraft, you may want to flip a coin instead.

* Editor’s Note: An earlier version of this post asserted that games of StarCraft 2 begin with four workers, when in fact this was only the case in the first SC.

Tags • , , , ,
Posted in The Haulout  • 

  • Jack

    Don’t try to write anything “smart” anymore, please, do world a favor. d a favor

  • exiLe


  • Dover

    This article betrays the authors poor knowledge of the competative StarCraft scene.

  • Patrick

    This is all so wrong.

    In RPS, scissors beats paper no matter what. In SC2, hellions don’t automatically beat zerglings. The better player almost always wins, regardless of opening builds.

    Also, good SC2 players won’t blindly mass one unit type. They’ll scout what their opponent is doing, and respond appropriately. When denied scouting, they’ll generally favour a unit mix that does well against anything. In RPS, every decision is made completely blindly.

    The author of this article has no idea what he is talking about.

  • jim

    Terrible article, doesn’t consider half of the strategy in the game and ignores key elements of decision making. The author also seems confused about what game he’s playing because there are no helions in brood war and you don’t start with 4 workers in starcraft 2.

  • UndrWrlDream

    Sorry but i’ve got to say, you are totally wrong sir.

  • Fail

    First try to learn the game, then write a blog about it.
    Btw, go write a blog.

  • David Rusak

    Sorry, everyone: I forgot that the worker count had changed when I was writing this.

    Patrick, I’m aware that what I’ve written here is (by necessity) a gloss on the actual depth of SC strategy — but even in a high-level game with mixed armies, using certain units to counter others is a core part of the game, is it not? Hence “many nested and iterated RPS-like battles”. I’m not saying you may as well just play RPS instead.

    (And, for the record, Exile, I’m placed in gold, not bronze. Gold! That’s like first place, right, guys?)

  • jim

    I really like that you replied to the angry nerds my friend. There is a saying in the community that once you are in diamond league you are just begginning to learn the game.

Canada & its place in the world. Published by
the non-profit charitable Walrus Foundation
The Walrus SoapBox
The Walrus Laughs
Walrus TV
Archived Blog Posts
  • August 2012
  • July 2012
  • June 2012
  • May 2012
  • April 2012
  • March 2012
  • February 2012
  • January 2012
  • December 2011
  • November 2011
  • October 2011
  • September 2011
  • August 2011
  • July 2011
  • June 2011
  • May 2011
  • April 2011
  • March 2011
  • February 2011
  • January 2011
  • December 2010
  • November 2010
  • October 2010
  • September 2010
  • August 2010
  • July 2010
  • June 2010
  • May 2010
  • April 2010
  • March 2010
  • February 2010
  • January 2010
  • December 2009
  • November 2009
  • October 2009
  • September 2009
  • August 2009
  • July 2009
  • June 2009
  • May 2009
  • April 2009
  • March 2009
  • February 2009
  • January 2009
  • December 2008
  • November 2008
  • October 2008
  • September 2008
  • August 2008
  • July 2008
  • June 2008
  • May 2008
  • April 2008
  • March 2008
  • February 2008
  • January 2008
  • December 2007
  • November 2007
  • October 2007
  • September 2007
  • August 2007
  • July 2007
  • June 2007
  • May 2007
  • April 2007
  • March 2007
  • February 2007
  • January 2007