Comment by throwawaymaths

then she is being misleading, at best. To date no observed galaxies (versus galaxy clusters) have had the arrangement (as is in this image) wherein the galaxy is so perfectly in line with a background galaxy and close enough to estimate the rotation curves (and the background galaxy is of the right disposition to know if dark matter halos extend beyond the galaxy and estimate by how much space). Without both those factors it's really difficult to do a proper correlation of dark matter distribution around galaxies and the observed light bending.