Massive stars are typically in binary systems and end their lives in supernova explosions that have dramatic consequences on the binary itself. In this talk, I will discuss the observable consequences of these explosions, in particular for gravitational wave observations. Specifically, the vast majority (>75%) of massive binaries are disrupted by the first supernova they experience: therefore X-ray binaries and gravitational wave progenitor are the exception rather than the rule in massive binary evolution. However, this creates the opportunity to use a common binary evolution byproduct (the "widowed" stellar companion ejected from the binary) to probe the explosion physics, compact object formation, and supernova natal kicks. Using Gaia data on these unbound companions might provide observational constraints on these key processes for gravitational waves. At the more massive end, stars become unstable because of the pair-production instability. Pulsational mass loss due to this process has a large and as-yet poorly explored effect on gravitational wave observables such as the predicted mass, eccentricity, and spins distribution of massive black holes. Observational confirmation of the long theorized pair-production instability have so far been elusive in the electromagnetic spectrum, but they might come with the next ~10-100 gravitational wave detections.