Full wind recovery was rare within 72 hours after first offshore 30-kt losses in two basins in 1982-2023

We tested whether a future offshore filter changes full wind recovery rates after large wind losses. We used fixed IBTrACS v04r01 tracks in the North Atlantic and eastern North Pacific in 1982-2023. Each storm supplied its first eligible loss of at least 30 kt in 24 hours. Recovery means a return to the wind before the loss, within 72 hours and before coastal approach or exit from accepted storm status or basin. Only 2 of 257 episodes recovered (0.78%). In the subset with all future fixes offshore and in accepted status, 1 of 44 recovered (2.27%). The paired difference was 1.49 percentage points, with a 95% season-block interval of -1.12 to 6.23. These sparse recoveries do not resolve the filter effect. Four records had unknown follow-up before a known outcome. If all four recovered, the all-episode rate would be 2.33%. This bound does not include recovery after coastal or status exit. The result is a narrow track benchmark, not a test of ocean cause or operational risk. We provide source fields, hashes, episode labels, code, checks and a figure. Planned checks with 20-kt and 25-kt losses gave positive filter differences, so the result depends on loss size.

tropical cyclone · rapid weakening · wind recovery · IBTrACS · future selection · competing events · null result