1# Verification checks21. Clean re-run: deleted results.json/table1.csv/figure1.png/exclusions.csv, re-ran analysis.py; results.json byte-identical (cmp) — PASS.32. Input SHA-256 equals README value — PASS.43. n_input_rows = 2498 = wc -l minus header — PASS.54. Table1 n_days sum = 2498 = n_input_rows — PASS.65. Baseline cross-check: mean change −0.042851 bp equals 100·(last−first)/(n−1) telescoping identity — PASS.76. Exclusion ledger: 2498 − 1 − 2497 = 0 rows in estimand windows; n_analysed = null consistent — PASS.87. n_treated = 0 consistent with all dates < 2020-03-04 and post_cut never 1 — PASS.98. Conclusion code premise_invalid with feasibility infeasible and all inestimable keys null — PASS.109. CI-contains-estimate / SE>0 checks: not applicable (no estimate) — N/A.11