Four TESS sectors do not resolve a linear HAT-P-49 b transit width change from 2021 to 2024 under a sector error model

We test whether the HAT-P-49 b transit width changed in four public TESS sectors from 2021 to 2024. Coverage rules retain 33 transits with 120-second exposures. A stretched spherical reference profile gives a linear slope of -0.542 minutes per year. The 95% sector interval is -1.348 to +0.265 minutes per year. It assumes independent normal sector errors with a common variance about a linear mean. The equal 2024 mean is 1.900 minutes smaller than the 2021 mean. Its conditional event bootstrap interval excludes zero, but holds sector effects and the reference shape fixed. A separate sector spline gives a contrast of -1.692 minutes, with a conditional interval from -3.345 to +0.057 minutes. All tested sector slope intervals include zero. Grouping the visits into three observing years further increases the uncertainty. Detector and time effects are partly mixed. Artificial transits recover input slopes of zero and -1 minute per year under a measured-residual noise model. These data preserve a candidate narrowing but do not establish a linear physical change or a precession rate. The package includes unchanged FITS files, source hashes, code, event decisions, results, figures, numerical checks, and instructions for a new run.

HAT-P-49 b · TESS · transit width · transit duration variation · sector errors · orbital precession · reproducible astronomy