| Canonical · observed shadow audit | Funded sizing simulation · simulated | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Entry ET | Side | Entry | Stop | Stop dist | Qty | $ Risk | Result | Exec-adj | P/L | Hold | Exit | Sim qty | Sim risk | Sim P/L | Sim status |
| No true-live TRAIL observations yet. | |||||||||||||||
A2.10's frozen shadow implementation and its database migration are deployed and wired into the live webhook processor, but its true-live boundary is still the eternal-future placeholder.
Remaining step: after a real production deployment is independently verified — migration applied, real MNQ webhook bars reaching the runner, a real observation row written and inspected, restart/idempotency clean — replace A2_10_TRUE_LIVE_BOUNDARY_UNIX in lib/strategy-engine/research/a2Dot10Shadow/config.ts with that verified moment. Until then every signal is skipped by design and none counts toward the prospective sample.
Canonical sizing is the existing deterministic MNQ stop-distance engine (SHADOW_ONLY_STOP_DISTANCE_SIZING), max 3 contracts — unchanged. The amber funded sizing simulation bands re-express those same realized R outcomes at a provisional funded-account policy (funded-sim-v1-provisional); they are a what-if overlay, never observed P/L, and they do not affect any R, any shadow figure, or how a single trade was taken.