Fixed scene
The test removes economy, build order, and scouting. Only two armies, starting positions, a map, and one micro-control policy remain.
HYPOTHESIS VISUALIZATION
Micro is tested with paired fights, not belief
Battle Lab runs fixed armies in a baseline-vs-managed format: the same seed, map, side, formation, and composition. That makes it clear what a new tactic actually added and what was just noise.
EXPERIMENT DESIGN
How hypothesis testing works
Paired comparison
Every scenario runs a baseline and a managed variant. The delta is measured under identical conditions, not between random matches.
Uplift metrics
The system measures micro uplift, remaining army value, damage delta, duration delta, wins, losses, and technical failures.
Mechanics review
The visualization helps separate useful target scoring, Blink, Guardian Shield, or lift behavior from tactics that look nice but lose.
CURRENT TRACKS
Which hypotheses run through the lab
Defensive Blink, Colossus screen, Immortal armored target focus, Guardian Shield timing, Phoenix lift reservations, Void Ray target locks, firepower target scoring, and lethal squads pass through the same scenarios so progress stays measurable.
VISUAL LAYER
Scenarios are readable at a glance
Each card shows the paired fights, outcome, value delta, compositions, map, and run status. You can quickly see where micro improved and where a new idea is needed.
BOT IMPACT
From Battle Lab to NexusBot
Policies that produce stable uplift in the lab move through fixture matches next, and only then become part of the tournament bot's general logic.