AoE II Civilization Win Rate Simulator
Observe how ELO matchmaking keeps civilizations hovering near 50% win rate, regardless of how strong or weak they are.
Simulation Parameters
- ELO player drift assigned matching a typical skill distribution curve found in competitive games.
- In ELO matchmaking, players match against opponents within 100 ELO whenever possible.
- Each simulated match randomly assigns civilizations to both players; mirrors are allowed.
- Random matchmaking ignores ELO pairings and relies solely on civilization strength for outcomes.
- The civ strength spread control exaggerates or flattens balance differences before each simulation.
- Visualization updates stream in batches so progress feels alive without instant jumps.
Civilization Performance
Strength is the theoretical balance point; win rate is measured from simulated results.
| Civilization | Strength | Expected Win Rate | Simulated Win Rate | Difference | Matches |
|---|
Simulation Insights
Dive into civs that notably over- or under-performed versus their theoretical strength.
Author’s conclusions
Win rates by civilization do not reflect their strength. i.e. if a civ is 70% stronger on average than all other civs, it will only have a slightly higher than 50% win rate, instead of 70% (expected).
This does not demonstrate that civs in aoe de are imbalanced. It only demonstrates that if civs are imbalanced, it will not reflect properly in the win rates for civilizations.
Therefore, win rates should not be used to determine civ strength or for civ balancing purposes, since even small win rate differences could be hiding or understating massive civ imbalances.