The geometric challenge of testing gravity with wide binaries

Wide binaries provide promising laboratories for testing general relativity (GR) in the low-acceleration regime. Recent observational studies have found that the difference in the proper motions and/or radial velocities of the components of nearby wide binaries appear larger than predicted by Kepler's laws, indicating a potential breakdown of GR at low accelerations. These studies have not accounted for projection effects owing to the different position of the two stars on the celestial sphere. I show that two stars in a wide binary with identical 3D space velocities often have significantly different proper motions and radial velocities purely due to projection effects. I construct a sample of simulated binaries that follow Kepler's laws and have similar phase-space distributions to the observed samples of nearby binaries. Beyond separations of similar to 0.1 pc, direct comparison of the components' proper motions would suggest strong tensions with GR, even though the simulated binaries follow Kepler's laws by construction. The magnitude of the apparent disagreement is similar to that found observationally, suggesting that the apparent tension between observations and GR may largely be due to projection effects. I discuss prospects for constraining gravity at low accelerations with wide binaries. Robust tests of GR are possible with current data but require measurements of 3D velocities. Further work is also needed to model contamination from unbound moving groups and unrecognized hierarchical triples.