On Starcraft and strategy

Brit has a good post up about a tricky design decision that goes into computer strategy games: the question of attack concentration. Pardon the lengthy quote, but here’s the section that particularly caught my attention:

If grouping units together increases their power, it means there is an incentive to group units together and a disincentive to split units apart. This fact affects gameplay in a major way. In games where there is a strong incentive for grouping, the progression of the game is rather predictable: expand until you encounter the enemy, maybe engage in a few skirmishes to capture objectives that the enemy hasn’t defended well (because he just arrived), build-up a large military, engage in one large battle which effectively determines the winner (the game may not end there, but the battle effectively determines the eventual winner), and then play out the foregone conclusion. The predictability of it is somewhat boring, and I get a little bored of the “military build-up” phase. I’ve seen a lot of games of Starcraft and Warcraft where this happens.

Brit here is pointing out something that any veteran Starcraft player (or any player of similar strategy games) has undoubtedly noticed: games between reasonably skilled players tend to follow the same basic pattern. Because of the way the game’s combat and other systems work, most games feature a relatively quiet, and often quite lengthy, period of military build-up followed by a massive, apocalyptic battle in which each side throws every conceivable unit into the fray. This massive battle either “breaks” one of the opponents, or (quite often) it ends with both sides’ armies effectively destroyed, prompting a second period of military build-up during which each side races to restore its fighting power before the enemy does. That’s a bit of an oversimplification, but it’s how games usually go.
This isn’t likely to interest most of you, but this got me thinking about the way that my Starcraft games against humans play out. Here’s what a typical game for me looks like:

  • Phase 1: Establishing a presence (5 minutes): Each side scrambles to get a functional base. A few defensive structures and units are built, usually just enough to safeguard the fledgling base from a sneaky early-game rush attack from the enemy.
  • Phase 2: Early build-up and expansion (5-10 minutes): Both players start building more advanced structures and begin to assemble an army. Scouts are dispatched around the map to hunt for mineral deposits. A few basic “recon” battles may occur as each side tries to get a glimpse of the other side’s army composition and general location.
  • Phase 3: Skirmishes and expansion (15 minutes): Both sides make moves to claim any mineral deposits that haven’t yet been secured. Lots of skirmishing between medium-sized forces can happen, as both sides try to win “quick and easy victories” over enemy expansion bases that aren’t yet well-defended.
    Almost invariably, during this phase, the “pivot point” of the map becomes clear: the strategic location over which almost all future battles will be fought. Often this is a rich mineral deposit located in the center of the map, which promises to provide a decisive strategic edge to whoever can claim final control over it.

  • Phase 4: Clash of the titans (10 minutes): Enough time has passed now that both sides have built up substantial armies, probably including one or two advanced unit types. Typically, a strategic stand-off settles in while each side carefully (but hastily) prepares for a huge offensive.
    Somebody (usually the person who hasn’t been able to claim the pivot point, and thus feels pressure to reverse the strategic situation before it’s too late) pulls the trigger and launches a massive attack on the pivot point. The other player pulls in all forces to the defense and the battle is joined. Clouds of Terran battlecruisers and siege tanks, Protoss scouts and carriers, and Zerg hydralisks and guardians pour across the map.
    This phase typically ends when both sides annihilate each other’s forces. Usually, somebody emerges from the uber-battle in better shape than the other player, but rarely with enough surviving force to go the final mile and win the game.

  • Phase 5: Frantic rebuilding (5-10 minutes): Both players retreat any survivors and immediately set to rebuilding their bases and armies as fast as possible. At this point, whoever can get even a medium-sized force onto the battlefield first usually has a big advantage.
    Both sides often try “probing” attacks against the enemy’s main base, bypassing the pivot point, in the hopes that a decisive (even if under-strength) attack on a distracted enemy’s weak point will win the game quickly. (This usually doesn’t work.)

  • Phase 6: Armageddon (10 minutes): Somebody decides that their army is sufficiently rebuilt and launches a major attack. The other player responds by pulling in all available units to stop it. Because mineral supplies are running low, this fight usually decides the game. Often both sides annihilate each other again, but afterwards one side finds that it no longer has the economic ability to replace its losses. Although this player may have plenty of static defenses left on the field, it’s clearly just a matter of time before the other player slowly rebuilds and creeps inexorably across the map.
    At this point, somebody usually surrenders rather than watch the enemy roll across the map uncontested.

Of course, the fun part of the game often comes in deliberately disrupting this schedule to throw off an enemy who’s expecting the game to play out in about this fashion. So many years after the game’s release, you’d think that it’s not possible to be surprised by enemy tactics and strategies… but almost every time I play, my opponent pulls off something new and interesting (and alarming).
That said, I’ve got a Starcraft game date set for later this week, and I wouldn’t want to reveal all of my strategies. Thanks, Brit, for giving me an excuse to muse on one of my all-time favorite games!

Facebooktwitterredditpinterestlinkedinmailby feather

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.