One for the Future: measuring the mass transfer rate in the ULX M82 X-2 by using orbital period changes will take millenia

Bachetti et al. (2021) have recently claimed to measure the mass transfer rate in the pulsing ULX system M82 X-2 by following the change of its orbital period over 7 yr. We reiterate the known point that this method cannot give a reliable result (or even necessarily predict the correct sign of long-term period change) without a far longer baseline (here $\gg 1000$~yr) or for systems with a much higher long-term mass transfer rate ($\gg 10^{-4} \rm M_{\odot}/{\rm yr}$), if they exist. Applying the method of Bachetti et al. (2021) to measured orbital period derivatives predicts that the well-studied quiescent X-ray transients XTEJ1118+480, A0620-00 should currently instead have steady accretion discs and be bright X-ray sources, while Nova Muscae 1991 should be still brighter (a ULX). But all three sources are observed to be extremely faint. We conclude that there is no evidence to support the high mass transfer rate that Bachetti et al. (2021) find for M82 X-2 as it is deduced from period noise not related to the binary evolution.