We model the fastest moving ( v _ { tot } > 300 { km } { s } ^ { -1 } ) local ( D \lesssim 3 kpc ) halo stars using cosmological simulations and 6-dimensional Gaia data .
Our approach is to use our knowledge of the assembly history and phase-space distribution of halo stars to constrain the form of the high velocity tail of the stellar halo .
Using simple analytical models and cosmological simulations , we find that the shape of the high velocity tail is strongly dependent on the velocity anisotropy and number density profile of the halo stars — highly eccentric orbits and/or shallow density profiles have more extended high velocity tails .
The halo stars in the solar vicinity are known to have a strongly radial velocity anisotropy , and it has recently been shown the origin of these highly eccentric orbits is the early accretion of a massive ( M _ { star } \sim 10 ^ { 9 } M _ { \odot } ) dwarf satellite .
We use this knowledge to construct a prior on the shape of the high velocity tail .
Moreover , we use the simulations to define an appropriate outer boundary of 2 r _ { 200 } , beyond which stars can escape .
After applying our methodology to the Gaia data , we find a local ( r _ { 0 } = 8.3 kpc ) escape speed of v _ { esc } ( r _ { 0 } ) = 528 ^ { +24 } _ { -25 } { km } { s } ^ { -1 } .
We use our measurement of the escape velocity to estimate the total Milky Way mass , and dark halo concentration : M _ { 200 , tot } = 1.00 ^ { +0.31 } _ { -0.24 } \times 10 ^ { 12 } M _ { \odot } , c _ { 200 } = 10.9 ^ { +4.4 } _ { -3.3 } .
Our estimated mass agrees with recent results in the literature that seem to be converging on a Milky Way mass of M _ { 200 , tot } \sim 10 ^ { 12 } M _ { \odot } .