We report the first plausible optical electromagnetic counterpart to a (candidate) binary black hole merger. Detected by the Zwicky Transient Facility, the electromagnetic flare is consistent with expectations for a kicked binary black hole merger in the accretion disk of an active galactic nucleus [B. McKernan, K. E. S. Ford, I. Bartos et al., Astrophys. J. Lett. 884, L50 (2019)] and is unlikely [<O(0.01%))] due to intrinsic variability of this source. The lack of color evolution implies that it is not a supernova and instead is strongly suggestive of a constant temperature shock. Other false-positive events, such as microlensing or a tidal disruption event, are ruled out or constrained to be <O(0.1%). If the flare is associated with S190521g, we find plausible values of total mass MBBH∼100M⊙, kick velocity vk∼200kms−1 at θ∼60° in a disk with aspect ratio H/a∼0.01 (i.e., disk height H at radius a) and gas density ρ∼10−10g cm−3. The merger could have occurred at a disk migration trap (a∼700rg; rg≡GMSMBH/c2, where MSMBH is the mass of the active galactic nucleus supermassive black hole). The combination of parameters implies a significant spin for at least one of the black holes in S190521g. The timing of our spectroscopy prevents useful constraints on broad-line asymmetry due to an off-center flare. We predict a repeat flare in this source due to a reencountering with the disk in ∼1.6yr(MSMBH/108M⊙)(a/103rg)3/2.