Current observations are insufficient to confidently associate the binary black hole merger GW190521 with AGN J124942.3 + 344929

Recently, Graham et al (2020 arXiv:2006.14122) identified ZTF19abanrhr as a candidate electromagnetic counterpart to the binary black hole merger GW190521. The authors argue that the observations are consistent with a kicked binary black hole interacting with the accretion disk of the active galactic nucleus AGN J124942.3 + 344929. If a real association (rather than happenstance), this has implications for the sources of LIGO/Virgo binary mergers, future prospects for electromagnetic counterparts, and measurements of the expansion rate of the Universe. In this work, we provide an analysis of the multi-messenger coincident-significance based on the localisation overlap and find that, under optimistic assumptions, the odds of a common source for GW190521 and ZTF19abanrhr range between 1 and 12. These odds are strongly dependent on the waveform model and, with current models, are unable to consistently capture both the effects of precession and orbital eccentricity. We consider this insufficient evidence to warrant confidently associating GW190521 with ZTF19abanrhr and hence caution against any astrophysical conclusions based on the association.