How recommendations are made
Pokémon GO battle simulation methodology
Go Battle Forge uses deterministic, scenario-based modeling to compare teams consistently across every listed opponent combination. Results are planning estimates—not guarantees of an identical live battle.
Current lineup scenarios
Shadow Battle Lab expands each Rocket opponent's lead, second, and closer options into exact three-Pokémon lineup scenarios. A displayed win rate is the share of those modeled scenarios completed successfully by that team.
Stats, types, and moves
The model uses Pokémon species stats, combat power multipliers, type effectiveness, fast-move timing, charged-move energy and damage, Shadow modifiers, and eligible Mega or Primal forms. Recommended moves are shown directly on each team card, including Elite TM notices where applicable.
Rocket battle behavior
Leader and Giovanni recommendations account for shield pressure and the post-switch or post-charged-move delay used in Rocket battle planning. Team order is evaluated by role: opening pressure, shield support, and closer safety. Settings let trainers prioritize survival or speed and compare direct leads with an opening swap.
Ranking and replacements
Teams are ranked first by exact scenario outcomes, then by the selected survival or speed priorities. Choosing “I don't have this” removes that exact form and state—such as Shadow Kyogre without removing regular Kyogre—and reranks the same evaluated team pool.
Data updates and limitations
Rocket rotations and battle data are refreshed through an automated validation pipeline. Pokémon GO can change lineups, moves, stats, and battle behavior; real outcomes also depend on tapping, switching, lag, Pokémon level, and IVs. Please report suspected differences through the contact link on our About page.