Risks (strategic register)

# Risks (strategic register)

- **S1 · The reading is heard as a grade.** Whatever the copy says, a person may take "held the line at the
  hearing" as a score. *Read:* world-language only, uncertainty always shown, no numbers in the reading
  surface, the reading proposes the next hour rather than closing the last.
- **S2 · Anchors drift or show.** Route-invariant forks reused across hours are what link the scale; if
  generation moves them they measure nothing, and if a player recognises them they measure recognition.
  *Read:* fit statistics on every anchor; the generating model as a facet; anchors disguised by setting.
- **S3 · Emotion without change.** A journey can be moving and move nothing. *Read:* the proof programme
  reports "no clear change yet" honestly; the product is still an experience worth its subscription, and
  says so at that rung ([12](12-the-practice-monopoly.md#the-proof-programme)).
- **S4 · The proof takes longer than the runway.** A pre-registered transfer study with three-month
  retention is a calendar, not a sprint. *Read:* start the record with the first paying journey; sell the
  hour on what it is while the study runs; publish either way.
- **S5 · The test world is learned, or trusted more than it should be.** A benchmark that can be
  memorised is a corpus; one that overstates what it proves is a liability. *Read:* kept history is the
  contamination defence and the transfer boundary is the honesty defence
  ([13](13-consequential-test-worlds.md)).

Raw Markdown source