
================ OBSERVED_RECONSTRUCTION_CORRECTION | SESSION 20260907_020601 ================
Correction provenance: V9.1.8 supersedes the prior observed-source reconstruction block only; original live capture, ledger, A-E shuffle results and prior text remain unchanged.
Reason: observed 247 no-peek timing/card consumption is honoured; an explicitly documented pre-Hand-1 starting-balance correction is treated as a start correction, not as a hand-settlement adjustment; legitimate Casual bankroll/table exit counts as a completed counterfactual journey.

================ OBSERVED-SOURCE FROZEN vs CASUAL RECONSTRUCTION | SESSION 20260907_020601 ================
V9.1.5 observed-source boundary: exact captured chronology is replayed in recorded order with observed 247 no-peek timing; every visibly observed shuffle boundary resets to the corresponding captured segment. No card is invented. A-E synthetic shuffle methodology is unchanged.
Frozen replay gate: hands 9 | final £10.00 | W/L/P 2/6/1 | exposure £180.00 | peak £115.00 (hand 1) | trough £10.00 (hand 9) | maxDD £105.00 (hand 9) | PASS
Frozen observed provenance (authoritative live capture): Total 11 | Textbook/Frozen-Aligned 11 | Research Overrides 0 | Primary H11v10 Overrides 0
Casual decision provenance: N/A
Directly Comparable Action Decisions: N/A | Matched N/A | Different N/A | Alignment N/A
RECONSTRUCTION STATUS: NOT SOURCE-VERIFIED | Frozen gate PASS | Casual completion FAIL | Arithmetic PASS
Publication rule: do not infer or transplant missing Casual/comparison values from another session.
================ END OBSERVED-SOURCE RECONSTRUCTION | SESSION 20260907_020601 ================
OBSERVED_RECONSTRUCTION_CORRECTION_STATUS | session 20260907_020601 | COMPLETE
================ END OBSERVED_RECONSTRUCTION_CORRECTION | SESSION 20260907_020601 ================
