Week 10: Implication over description

Week 10 — why horror comes from what's withheld, and how escalation needs warning shots before it can be lethal.

26 April 2027

The most common way a scenario undercuts its own horror is over-describing the threat. This lecture is about writing less, on purpose, and about pacing danger so it reads as escalation rather than an arbitrary spike.

Two more design habits worth carrying past this week, even though neither gets its own lecture slot: a puzzle only earns its place ahead of a plain skill check when solving it teaches players something the check can’t — everything else should stay a check, however tempting it is to dress up a find with a lock and a key. And every NPC in a draft should survive one test before making the final cut: a single sentence stating what they let a player do that the scenario couldn’t do without them. An NPC who fails that test, however good their backstory, is padding; the ones who pass it earn exactly one vivid, playable trait rather than a paragraph of backstory nobody at the table will ever see.

Puzzles generally fall into a few kinds: object-hunting puzzles, pattern puzzles, wordplay puzzles, and decoding puzzles. A “hard” puzzle isn’t automatically a “good” one — a puzzle exists to be solved, and a scenario isn’t ranked by how difficult it is, only by whether the players got some enjoyment out of it. Running two or more puzzles in parallel that lead toward the same answer often works well: when players get stuck on one, they can solve another first and work backward from its result into the one they couldn’t crack. For NPCs specifically: before placing one in a scenario, be clear about the role they’re actually meant to play — guiding the players to the next scene, delivering a hidden clue, or acting as a pawn sent by the mastermind to mislead them — and write that down explicitly. A memorable NPC comes from one distinctive trait, not from a name and a paragraph of backstory; and NPCs talking to each other at length while the players just watch should be avoided — running a session is a story you build together with your players, not one you perform for yourself.

Outline#

• describing effect (what a character hears, smells, senses) instead of appearance
• knowing more about your antagonist than you tell the players, and treating that as the point rather than a gap to fill
• escalation: a threat should give visible warnings before it’s capable of killing a character, so danger is legible rather than sudden
• two named frameworks for that pacing: Blase’s five-stage escalation (unease, confirmation, encroachment, confrontation, aftermath) and Petersen’s “Creepy Stuff Rule” — three warnings (hint, clear danger, the monster itself) before a threat can actually kill
• this is also prep for playtesting in week 11 — a scenario that over-explains on the page tends to over-explain at the table
• a puzzle earns its place only when solving it teaches players something a plain skill check can’t — everything else should just be a check
• giving every NPC a one-sentence reason to exist and exactly one playable trait, and cutting the rest rather than padding the cast for flavour