Four 2026 TESS sector means show WASP-18 b transits 41 seconds before the printed linear clock

We test the printed WASP-18 b linear transit clock of Nediyedath et al. (2026) with public TESS data. We measure 205 complete transits in ten 120-second SPOC light curves. A transit template is fitted only to the six sectors used by the source. We then hold that template fixed. The four later sectors, 103 to 106, contain 78 selected transits. Their mean fitted time is 40.96 seconds before the printed clock, with a measurement standard error of 1.96 seconds. The source does not give its reference-time to period covariance. We allow any such covariance and keep its forecast error shared across sectors. The combined standard error is then 2.93 seconds, conditional on the fixed template and independent sector errors. Eight tested reduction changes give offsets from -42.10 to -40.29 seconds. An unknown error shared by all later sectors is not measured. An assumed shared error of 30 seconds reduces the offset to 1.36 standard errors. A linear clock fitted only to the older sectors predicts the later means with chi square 5.24 for four degrees of freedom. A linear clock also describes all ten sector means. The result identifies a discrepancy in the tested printed clock. It does not identify the cause or establish a new planet or orbital decay. The package includes original light curves, file hashes, source text, code, event selection, transit times, figures, and commands for a new calculation. This is a retrospective test on data that the source did not use. It shares the source telescope and archive pipeline.

WASP-18 b · TESS · transit times · transit clock · ephemeris · reproducible data