Loading field map
Every StarCraft player faces a relentless trade-off during combat: how to make each unit deal as much damage as possible while taking as little as possible, all under the pressure of real-time decision-making. This is the domain of micro and unit control, a subfield distinct from macro (economy and production) and build orders (opening sequences). Micro is about the moment-to-moment manipulation of individual units or small groups—moving, targeting, using abilities—to gain an edge in fights. Over two decades of competitive play, five major frameworks have emerged, each offering a different answer to the question of how to control units effectively.
In the early years of StarCraft, players discovered that the game’s default unit behavior—auto-attacking the nearest enemy, clumping together—was often suboptimal. The Foundational Micro School codified a set of basic techniques that became the vocabulary of all later micro. Focus fire (ordering multiple units to attack the same target to eliminate it quickly), kiting (moving a ranged unit backward while attacking to keep melee enemies at a distance), and stutter-stepping (alternating move and attack commands to maintain damage output while repositioning) were among the core skills. This school also emphasized using unit abilities such as the Dragoon’s ranged attack or the Vulture’s mines. The Foundational School’s contribution was to show that overriding the engine’s default behavior could dramatically improve combat outcomes. These techniques remain the baseline expectation for any competitive player today.
The Korean Mechanical School did not replace the Foundational School; it absorbed and scaled its techniques. Korean players, particularly in the professional scene of Brood War, reframed micro as a trainable physical and cognitive skill rather than just tactical knowledge. They pushed the limits of actions per minute (APM), control group usage, and multitasking. A player like Lee “Flash” Young-ho could individually control multiple groups of units simultaneously, executing precise focus fire and spellcasting across the map. This school treated micro as a speed-intensive craft: the faster and more accurately you could issue commands, the more value you extracted from each unit. The Korean School coexisted with the later Deathball School during the overlap period (2008–2015); Korean mechanical skill was often applied to deathball compositions, but the school’s emphasis on individual unit control sometimes conflicted with the army-level thinking of the Deathball School.
As StarCraft II arrived and the scale of battles grew, a new school shifted focus from individual units to managing massed armies. The Deathball and Composition Micro School argued that the most important micro decisions happen before the fight: choosing the right unit composition, positioning the army as a cohesive formation, and using area-of-effect abilities (like the Colossus’s thermal lance or the High Templar’s Psionic Storm) to wipe out clumped enemies. This school narrowed micro to army-level control—a-move with strategic pre-fight positioning, rather than constant individual unit commands. It coexisted with the Korean School, but its assumptions differed: the Deathball School valued composition synergy and formation over raw mechanical speed. In practice, players needed both: Korean mechanics to execute the deathball’s positioning and spellcasting, and composition awareness to build the right army.
The Splitting and Positioning School emerged as a direct reaction to the Deathball School’s vulnerability. Massed armies were extremely susceptible to area-of-effect damage—Banelings, Psionic Storm, Siege Tank shots—that could wipe out entire clumps in seconds. The core technique of this school was unit splitting: rapidly dividing a group of units (especially Marines) into smaller groups to minimize splash damage. This revived the importance of individual unit spacing within armies, a concern the Deathball School had downplayed. The Splitting School built on Foundational techniques (stutter-stepping, focus fire) but added a new principle: preservation through spacing. It coexisted with the Deathball School; top players learned to both build a strong composition and split their units during engagements. The Splitting School remains essential in matchups like Terran vs. Zerg, where Marine splitting against Banelings is a decisive skill.
The most recent framework breaks from the intuition-based approaches of earlier schools. The Engine-Driven Micro Optimization School uses data analysis, replay mining, and machine learning to evaluate micro techniques objectively. Rather than relying on a player’s feel or a coach’s advice, this school tests micro decisions by simulating thousands of engagements or analyzing professional replays for optimal patterns. For example, it has validated the optimal Marine split patterns taught by the Splitting School, showing that certain spread angles minimize Baneling damage more effectively than others. It has also refined kiting angles for Stalkers against Roaches, revealing that a 45-degree retreat path yields higher damage output than a straight line. This school provides an infrastructure for training: tools like Sparky (a StarCraft II micro trainer) and AI-driven analysis allow players to practice specific micro scenarios with immediate feedback. The Engine-Driven School is leading today because it offers objective benchmarks that can confirm, refine, or challenge the techniques of all prior schools. It does not replace them; rather, it provides a methodological backbone for understanding why certain micro works and how to improve it.
Today, all five frameworks remain active in competitive play. The Foundational School’s techniques are the basic vocabulary that every player learns. The Korean School’s emphasis on speed and multitasking is still valued, but it is no longer sufficient on its own—players must also understand composition and spacing. The Deathball School’s composition awareness is essential in the late game, especially in StarCraft II’s large-scale battles. The Splitting School’s spacing principles are crucial in specific matchups, particularly Terran vs. Zerg. The Engine-Driven School increasingly informs training regimens and strategic analysis. The main disagreement among these frameworks is between craft-based intuition (the Korean School’s belief that micro is a skill honed through practice and feel) and data-driven optimization (the Engine-Driven School’s insistence that objective measurement should guide technique). In practice, the two approaches complement each other: top players use data to identify optimal patterns and then drill them until they become second nature. The ongoing tension between human execution and analytical refinement continues to drive the evolution of micro and unit control.