Week 7 Studio: Puzzles and red herrings

Designing a puzzle only for a clue that actually deserves one, and checking it against a partner cold.

5 April 2027

A hard puzzle isn’t automatically a good one — a puzzle is good when someone enjoys solving it and it was worth the table time it cost. That means using puzzles sparingly, and only where the clue behind it matters enough to justify slowing the group down.

Puzzles roughly sort into four types — find-the-object, spot-the-pattern, wordplay, and decoding — and none of them outranks the others; the right type follows from the clue, not from what feels impressive to design. Two tests decide whether a clue deserves a puzzle at all rather than a plain skill check:

• The “why bother” test. If a check would deliver the same information just as well, use the check — a puzzle only earns its place when solving it teaches players something a check can’t. Compare finding a badge loose in a drawer (a Spot Hidden roll) against finding it inside a four-digit combination safe behind a bookcase: physically the same object, but the second version tells players this badge matters before they’ve even read what it says.
• The multi-thread test. A single puzzle that fully blocks progress is a stuck point waiting to happen. Give players two or three independently-solvable puzzles pointing at the same piece of information instead — this is also, mechanically, most of what makes a scenario “open” rather than linear: multiple live threads a stuck group can set aside and come back to.

Before the session#

From your tagged clue map, pick at most one or two clues that are both important and interesting to gate behind a puzzle. Everything else stays a skill check — that’s not a lesser design choice, it’s the right one for most clues.

In the session#

Design the puzzle, then swap with a partner and have them attempt it with no help from you. Afterwards, ask them what they thought the puzzle was for — if they can’t connect it to what it reveals, the puzzle is testing the wrong thing.

Afterwards#

Revise based on what your partner got stuck on or missed entirely. A puzzle that only you can solve isn’t ready for a scenario that other people will run.