Strong gravitational lenses are potentially important probes of cosmology
that can be used to constrain the Hubble constant (H0).
Recently, Xu et. al. (2016) fit radial density profiles to strong lensing
galaxies found in the Illustris simulation.
By assuming power-law profiles, they formally break the mass-sheet
degeneracy and find that scatters and biases of, at best, 5-10% on the inferred
value of H0 exist if they apply a careful selection criterion to the lenses.
However, these uncertainties only take into account rudimentary lensing
information, such as the Einstein radius.
We use realistic, N-body simulations from the EAGLE project, which take
into account the effect of baryons, to simulate mock lenses.
By combining lensing information with stellar kinematics of the lensing
galaxy, we assess whether these uncertainties can be brought down to the
sub-percent level in order to be competitive cosmologically and whether
upcoming large-scale surveys, combined with follow-up observations, will
be able to provide the necessary constraints.