
Protect the information economy
Town wins when separate clues can be combined. A clean will, early result post and accurate visitor log are valuable even when they do not catch evil immediately. Evil factions win when those records are missing, delayed or contradictory.
Choose actions that resolve a live question
Every ability should have a reason. Investigate a contested claim, protect a player whose survival produces future information, block an action that threatens Town, or kill a target whose removal changes parity. “I had to target someone” is usually a sign the action lacked a decision model.
Spend irreversible power later than reversible pressure
Questions, claims and trials create information; executions and shots remove players. Use the cheaper information tools first when time allows. A Jailor or Vigilante that waits for two independent clues will usually outperform one that converts the first suspicion into a kill.
Track who benefits from tempo
Fast voting helps the faction that already knows more. Mafia starts with private teammates; Town starts uncertain. When evidence is incomplete, evil benefits from rushing. When a mechanical contradiction is clear, Town benefits from acting before the liar can rebuild the story.
Review losses by decision, not result
A correct process can produce a bad outcome because the game contains uncertainty. After a match, ask whether the information available at the time justified the action. That habit improves future play much more than labeling every losing decision “bad.”
A winning-process example
Town has two conflicting investigative claims. Instead of immediately eliminating one, a Lookout can watch the more likely kill target while a Tavern Keeper blocks one claimant. The next morning may create two independent observations. If the game state still allows a safe day, gathering that information is stronger than gambling on a 50/50 vote.
Common mistakes
- Optimizing for being personally correct instead of for faction win probability.
- Using lethal abilities before cheaper information can resolve the same question.
- Rushing because the chat is loud rather than because the mechanics are solved.
- Reviewing outcomes without reconstructing what was knowable at the time.