Materials: Graph paper or the Sandbox widget; A simple economy game (coins / tokens); Analog fallback: paper ledger + pencil chart instead of Sandbox
Artifact: Hypothesis sheet + one revised economy dial
Kills: Playtesting is asking friends if they liked it.
Feeling is not magic. It is structure: loops that reinforce or balance what players do.
Pick a familiar game. Name:
Draw it as boxes and arrows. Label positive vs balancing edges.
Build a 10-turn economy with one income dial and one cost dial.
| Turn | Income | Cost | Bank |
|---|---|---|---|
| 1 | |||
| … |
Change one dial. Predict the curve before you fill it. Then fill it.
Analog fallback: same table on paper; chart the bank column by hand.
Write a hypothesis you could lose:
If I raise cost by 2, players will stall by turn 6.
Run a 10-minute play. Separate said from did. Kill one loved feature if the evidence demands it.
Name one cognitive-load or input barrier in your prototype. Treat it as design material, not compliance paperwork.
Misconception killed: “Playtesting is asking friends if they liked it.” → It’s an experiment you can lose.