Exploiting eight years of magnetic field data from the Cruster mission, we employ an automated magnetopause crossing detection routine to determine the magnetopause location over varying magnetic latitude and local time. For a period spanning nearly one solar cycle we build a database of 2709 magnetopause crossings and compare these locations to the magnetopause models of Petrinec and Russell [1996], Shue et al. [1998], Dmitriev and Suvorova [2000] and Lin et al. [2010]. We compare our detected locations with the predicted locations for a variety of solar wind conditions and positions on the magnetopause. We find that, on average, the Petrinec and Russell [1996] and Shue et al. [1998] models overestimate the radial distance to the magnetopause by ∼1 R_E (9%) whilst the Dmitriev and Suvorova [2000] and Lin et al. [2010] models underestimates it by 0.5 R_ERE (4.5%) and 0.25 R_ERE (2.3%) respectively. Some varying degree of control on the differences between the predicted and encountered locations, by the solar wind and location parameters, are found.