Radial velocity studies of accreting binary stars commonly use accretion disk emission lines to determine the radial velocity of the primary star and therefore the mass ratio. These emission line radial velocity curves are often shifted in phase from
the expected motion of the primary. These phase shifts cast doubt on the use of disk emission lines in the determination of mass ratios. We present a systematic study of phase shifts, using data from the literature to distinguish between possible explanations of the phase shift. We find that one widely adopted class of models is contradicted by observations (section 2). We present a generalized form of another class of models, which we call measurement offset models. We show that these models are quantitatively consistent with existing data (figures 2 and 3, and the discussion in section 4.4). We consider the implications of adopting measurement offset models, for both disk structure and determination of binary parameters. Specifically, we describe in section 6 how measurement offset models may be used improve determinations of the mass ratio based on disk emission lines. This could be a valuable new tool in determining masses of important astrophysical objects such as accreting neutron stars and black holes.
