Preceding slow motion gave no clear next-day wind forecast gain in two cyclone basins in 2002-2024

We tested whether preceding slow motion improved next-day wind forecasts in selected Atlantic and eastern Pacific tropical cyclone windows. We used fixed IBTrACS v04r01 data and fit each 2002-2024 test season with earlier seasons from 1982. The baseline used initial wind, position, month and prior wind gain. Adding a below-3-m/s indicator reduced eastern Pacific mean squared error by 0.329%, with a conditional 95% three-season block interval of -0.019% to 0.732%. Atlantic skill was -0.129%, with an interval of -0.744% to 0.599%. The main tests used 3,771 Pacific fixes from 313 storms and 1,309 Atlantic fixes from 150 storms. Both main intervals included zero. A relaxed 100-km land bound gave a small positive Pacific interval. Other model and scoring checks left gains uncertain. The result gives a chronological forecast benchmark after an earlier retrospective motion study. It does not show no motion effect. The earlier study used the same years to choose the threshold, so this is not independent validation. Final best tracks, future-based selection of complete tropical windows, simple models and 23 test seasons limit inference. No ocean cause or operational skill is established. Source hashes, exact derived inputs, every forecast, code, a numerical audit and a figure are included.

tropical cyclone · forecast skill · translation speed · chronological test · IBTrACS · null result