> that this could be observational error This could absolutely be observation error. Messups in science happen all the time. You never take a single data poi

> that this could be observational error This could absolutely be observation error. Messups in science happen all the time. You never take a single data point at face value without question, in any field. The spectrum might be polluted by another more distant object, the line matching might have been faked by some other intervening medium, the modelling for the lensing galaxy could be wrong. I agree, it looks good. But it could totally be a mistake.