Linear baseline fits give a four-minute decrease in V453 Cyg eclipse spacing from 2019 to 2024

We measure the primary-to-secondary eclipse spacing of V453 Cyg in nine public 120-second TESS SPOC light curves. A shared symmetric profile and a linear local brightness baseline give 98 event times. Equal-sector mean spacings are 1.8896617 days in 2019 and 1.8869113 days in 2024. The change is -3.96 minutes. An orbit-block bootstrap gives a conditional standard deviation of 0.41 minutes and a central 95% interval from -4.60 to -3.00 minutes. Seven fits with linear local baselines give changes from -5.72 to -3.78 minutes. Independent matching-depth measurements also give decreases with linear baselines. With quadratic outer baselines, the same depth test gives changes from -13.98 to +9.58 minutes across flux and depth choices. Thus the method choice exceeds the sampling error and can change the sign. A noiseless synthetic test on the real timestamps gives less than 0.02 seconds of false endpoint change. One assumed pulse and noise signal gives about 36 seconds of bias. It is not an error bound for the real star. Several published apsidal rates can describe the reference trend after eccentricity and orientation are refitted. We do not claim an improved physical rate or a planet. The contribution is a nine-sector timing dataset and a measured limit on its physical use. We preserve original FITS inputs, hashes, all source code, event and sector tables, bootstrap draws, adverse controls, figures, and offline reproduction instructions.

V453 Cyg · TIC 90349611 · TESS · eclipsing binaries · apsidal motion · eclipse timing · stellar pulsations · baseline sensitivity · reproducible astronomy