Context. Given that mergers are often invoked to explain many exotic phenomena in massive star evolution, understanding the evolutionary phase directly preceding a merger, the overcontact phase, is of crucial importance. Despite this, large uncertainties exist in our understanding of the evolution of massive overcontact binaries. Aims. We aim to provide robust observational constraints on the future dynamical evolution of massive overcontact systems by measuring the rate at which the periods change for a sample of six such objects. Furthermore, we aim to investigate whether the periods of unequal-mass systems show higher rates of change than their equal mass counterparts, as theoretical models predict. Methods. Using archival photometric data from various ground- and space-based missions covering up to ∼40 years, we measure the periods of each system over several smaller time spans. We then fit a linear regression through the measured periods to determine the rate at which the period is changing over the entire data set. Results. We find that all of the stars in our sample have very small period changes and that there does not seem to be a correlation with the mass ratio. This implies that the orbital periods for these systems are stable on the nuclear timescale, and that the unequal-mass systems may not equalize as expected. Conclusions. When comparing our results with population synthesis distributions, we find large discrepancies between the expected mass ratios and period stabilities. We find that these discrepancies can be mitigated to a degree by removing systems with shorter initial periods, suggesting that the observed sample of overcontact systems may originate from binary systems with longer initial orbital periods.