We present the analysis of a sample of twenty-four galaxy-galaxy strong gravitational lens systems with a background source and deflectors from the Illustris-1 simulation .
We create mock lensing observations with a data quality comparable to known samples such as the SLACS lenses , to study the degeneracy between the complex mass distribution of the lenses , subhaloes , the surface brightness distribution of the sources , and the time delays .
Using a novel inference framework based on Approximate Bayesian Computation , we find that for all the considered lens systems , an elliptical and cored power-law mass density distribution provides a good fit to the data .
However , due to the presence of unphysical cores in the simulated lenses , most reconstructions are affected by some form of the Source Position Transformation .
The latter leads to a systematic underestimation of the source sizes by 50 per cent on average , and an underestimation of the time delays by up to 53 per cent .
On the other hand , we find no degeneracy between complexity in the lensing potential and the inferred amount of substructure .
We recover an average total projected mass fraction in subhaloes of f _ { sub } < 1.7 - 2.0 \times 10 ^ { -3 } at the 68 per cent confidence level which is consistent with zero and in agreement with the fact that all subhaloes had been removed from the simulation .
Our work highlights the need for even higher-resolution simulations to quantify the lensing effect of more realistic galactic potentials better .
Finally , our results confirm that additional observational constraints may be required to break existing degeneracies .