Choices, uncertainty, and risk
Give players alternatives they can reason about without removing every surprise.
An interesting decision depends on context. Alternatives should differ in ways the player can understand, while outcomes may remain uncertain. Decisions are one source of enjoyment alongside execution, expression, sensation, and story.
Concepts
- Tradeoffs and dominant strategies
- A tactic that is best regardless of the opponent or situation can collapse a decision. Distinguish that from an option that is strong in one context. Check how the options perform in actual play.
- Information
- Separate what the player knows, can infer, and cannot know. Hidden information can be the central pleasure of a game. Uncommunicated rules are a different kind of uncertainty.
- Risk and probability
- A choice can have good expected value and still go badly. Teach the difference between evaluating a decision using available evidence and judging it only by the eventual outcome.
- Counterplay
- A strong action becomes more interesting when another player can anticipate, avoid, interrupt, or exploit it. Counterplay does not require every matchup to be perfectly even.
In games
Into the Breach ↗
Visible enemy intentions help the player compare interventions. Preventing damage and eliminating an enemy need not be the same decision.
Among Us ↗
A vote is interesting because players must interpret uncertain evidence and claims. Removing uncertainty would remove much of the activity.
Common mistakes
- Calling every unlucky result unfair.
- Giving one option all the benefits without a situational cost.
- Treating a meaningful expressive choice as invalid just because it does not maximize a score.
Write a decision worth discussing
- Describe one situation and two or three available actions.
- List the information, benefit, cost, and uncertainty for each.
- Change the situation and see whether the preferred option changes.
Done when: You can explain why a reasonable player might choose either of two options.
Notes save in this browser.
Check your understanding
A 90% action fails. Does that alone show the decision or the game was unfair?
Show answer
No. Evaluate whether the probability was communicated, alternatives existed, and the result followed the stated rules. A likely outcome is not guaranteed.
Sources
- MDA: A Formal Approach to Game Design and Game Research ↗ Design framework · Robin Hunicke, Marc LeBlanc & Robert Zubek · 2004
- GameFlow: A Model for Evaluating Player Enjoyment in Games ↗ Evaluation model · Penelope Sweetser & Peta Wyeth · 2005