Evony Battle Simulator

About

The Evony Battle Simulator models march-vs-march PvP combat using community-researched battle mechanics. Configure two armies, set buff percentages, and watch the engagement unfold step-by-step on a spatial battlefield.

Work in progress: The battle mechanics are based on the best community research available, but many facts are still unclear, missing, or approximated. If you have corrections, better data, or suggestions, please open an issue, join the discussion, or submit a pull request — all contributions are welcome.

Key Features

• Army Configuration — Set troop counts per type and tier (T1–T15) for attacker and defender, with ATK/DEF/HP buff percentages per troop type.
• Step / Round / Full Playback — Advance the battle one attack at a time, one full round, or run to completion, with adjustable speed.
• Spatial Battlefield — A 5,200-unit linear field with troop movement, range indicators, and attack arrows.
• Battle Log & Summary — Filterable log of every attack event, plus a side-by-side loss comparison panel.

Not Yet Modeled

These real-game mechanics are excluded from the current simulator:

• Generals (base stats, ascending stars)
• General Skills & Specialties
• Equipment on generals
• Research / Technology buffs
• Civilization & Alliance buffs
• Monarch Gear & Sub-city buffs
• Wounded / Killed ratio
• Wall / Trap mechanics (city attacks only)
• Rally mechanics (multi-march)

Links

• GitHub Repository — Source code, issues, and pull requests
• Discussions — Ask questions, share ideas, and discuss battle mechanics

Battle Mechanics Fact Sheet

Reference for the simulation rules. Sources: community research, in-game observation, and community sites listed below.

Troop Types

Four troop types, each with tiers T1 through T15. Each type fills a distinct battlefield role:

Speed and range are constant across all tiers for a given type. Only ATK, DEF, and HP scale with tier. Siege is the exception — its range increases in tier brackets (T1–T4: 1400, T5–T8: 1556, T9–T10: 1711, T11–T12: 1867, T13–T15: 2178).

Notable: Siege ATK = Siege HP at every tier. Ground HP is roughly 6× its ATK at every tier.

Sources: Theria Games — Troop Base Stats • Evony Guide Wiki — Troop Types

Phase Order (Attack Sequence)

Each round is divided into four phases. Troops attack sequentially by range (longest range fires first). When two types share the same range, the faster unit goes first:

Within each phase: The attacker strikes first, then the defender's surviving troops counter-attack. This gives the attacker a slight edge — they reduce the defender's count before the counterstrike.

Within a side (same type, multiple tiers): Higher tiers attack before lower tiers, dealing damage with better stats first.

Sources: EvonyTips — How Battles Work • Server 806 — Battle Mechanics

Battlefield & Movement

Linear battlefield, 5,200 units long. Attacker starts at position 0, defender at 5,200. Each troop type has its own independent position (8 positions total: 4 per side).

Each round, before attacking, each troop type evaluates movement:

• Check: Is any enemy unit within range from current position?
• Yes → Hold. Stay at current position and fire at the highest-priority target in range.
• No → Advance. Move forward by speed units toward the enemy. Fire if a target is now in range.

Key rules:

• Speed = distance per round. Mounted moves 600 units/round, Ground 350, Ranged 100, Siege 75.
• Direction is always forward. Troops never reverse or retreat.
• Hold triggers on any enemy in range, not just the priority target.
• Melee collision: When opposing troops would pass through each other, they stop at melee range (50 units apart).

Typical engagement timeline (both sides identical):

• Round ~4: Siege finds approaching Mounted in range (2178). Holds, fires.
• Round ~5: Mounted pairs lock in melee near midfield.
• Round ~8: Ground pairs lock in melee.
• Round ~17: Ranged finally reach engagement range (slow speed 100).

Ranged troops are late to the fight despite having the second-longest range — their low speed (100) dominates. Siege engages early due to massive range offsetting slow speed.

Sources: EvonyTips — How Battles Work • Theria Games — Layers Guide

Targeting Priority

Each troop type has a fixed priority chain determining who it attacks. If the #1 target is dead or out of range, it falls through to #2, then #3, etc.

Within a priority level (e.g. multiple tiers of the same type): the highest tier is targeted first, then the largest troop count.

Sources: Evony Answers — Ground • Ranged • Mounted • Siege

Damage Formula

• troopCount — surviving count of the attacking layer
• ATK — per-troop attack stat of the attacker (base × (1 + buff%))
• DEF — per-troop defense stat of the target (base × (1 + buff%))
• modifier — type matchup multiplier (see below)
• target_HP — per-troop HP of the target (base × (1 + buff%))

The ATK/(ATK+DEF) ratio ensures damage is always > 0 (no zero-damage edge cases), and high DEF reduces damage proportionally rather than by flat subtraction. The ratio is always between 0 and 1.

Buff percentages are additive: a 50% ATK buff on a T14 Mounted (base 6670 ATK) yields 6670 × 1.50 = 10,005 effective ATK.

Sources: Evony Guide Wiki — Battle Mechanics • Evony Guide Wiki — Buff/Debuff Guide

Damage Multipliers

Tier-dependent 4×4 coefficient matrices. The counter triangle gives bonuses and reverse penalties. Siege is offensively weak against everything. Two cells change at T11+.

T1–T10

T11–T15

Sources: Server 806 — Battle Mechanics • Top Games (Official) — Battle Mechanics

Battle End Condition

The battle ends when one side has zero troops remaining. The surviving side is declared the winner.

Whether field march-vs-march battles have a fixed round cap (community often cites 4 rounds for certain battle types) is unconfirmed for open-field PvP. This simulator runs until one side is eliminated.

Open Questions

Mechanics where community evidence is inconclusive or contradictory:

• Starting distance — Do all troop types start at the same distance from the enemy, or do they start at different positions based on their range? Visual evidence suggests troops may fan out by range, which would affect how quickly melee units engage.
  Ref: YouTube — Battle formation analysis (1:33)

Not Yet Modeled

These real-game mechanics are excluded from the current simulator:

• Generals (base stats, ascending stars, skills, specialties)
• Equipment on generals
• Research / Technology buffs
• Civilization & Alliance buffs
• Monarch Gear & Sub-city buffs
• Wounded / Killed ratio (all losses are treated as kills)
• Wall / Trap mechanics (city attacks only)
• Rally mechanics (multi-march)

References & Resources

• Theria Games — Evony Guides Collection (troop stats, layers, buffs, calculators)
• Theria Games — Troop Base Stats (complete T1–T15 stat tables)
• Evony Guide Wiki (generals, troops, PvP tips, boss mechanics, calculators)
• EvonyTips — How Battles Work (round structure, movement, attack sequence)
• Evony Answers — Battle Mechanics Forum (per-troop-type targeting & damage coefficients)
• Server 806 — Battle Mechanics (battlefield length, counter triangle, range/speed)
• Top Games (Official) — About Battle Mechanics (developer statement on counters)
• Evony TKR Guide — Troop Layers (practical layering for offense & defense)
• Evony Guide Wiki — Buff/Debuff Guide (how buffs modify the damage formula)

YouTube: Search for DerrickD (damage formula reverse-engineering), TheRiaGames (stat tables & layering), and Gamer Momment (PvP battle reports) for video explanations of these mechanics.

This simulator is open source: github.com/slowmock-a11y/evony-battle-simulator