EVT 2026-02-16 - Stage F Simple Threshold Model and Scale Guardrails

Session scope

Finalize simple-but-scalable execution policy for recovery thresholds and fallback behavior.

User-confirmed outcomes

1. Practice-time threshold model:
   • Option B selected.
   • Threshold is configured by skill.
2. Completion-ratio threshold model:
   • Option B selected.
   • Threshold is configured by exercise family (objective vs subjective).
3. C-tier fallback guardrails:
   • Option D selected.
   • Apply both:
     • allow tier C only when A/B inventory is below minimum,
     • block cross-program fallback even when same skill.
4. Recovery chain requirement:
   • Option A selected.
   • One qualifying similar exercise is enough for reset (no multi-step chain).
5. Product principle:
   • Keep logic simple enough for users and scalable enough for operation.

Trace

• Promoted into: DEC-0020
• Reflected in: STATE_Practice_Recommendation
• Reflected in: STATE_Progress_Metrics