Jack,
I tend to agree with your trusted expert opinion, though I struggle with Sutherland's definition of "realism". He says:
"My research is concerned with the interpretation of quantum theory and such issues as Bell's non-locality, the quantum measurement problem and the nature of interference. In particular, I am interested in mathematical extensions to both quantum mechanics and quantum field theory that reinstate realism explicitly.
"By realism here I mean the assumption that an underlying physical reality exists in the absence of measurement. Surprisingly, maintaining this assumption in conjunction with quantum mechanics leads (via Bell's theorem plus certain reasonable assumptions) to either a clash with special relativity or to the existence of backwards-in-time effects. Since the experimental success of relativity and the theory's attractiveness in my eyes make me wary of abandoning it too hastily (even at a 'hidden' level), my research is focused on possible models involving backwards causation. In particular, I have constructed a time-symmetric formalism in which events are determined by both initial and final boundary conditions. A second model of this type is presently being formulated." http://sydney.edu.au/time/people/sutherland.htm
I don't think an "underlying physical reality" is necessary to assure local realism. I agree with Sutherland's time-symmetric determinism, with the proviso that the entropy generated by past and future events is identical. Entropy hides its origin (information from the future is as likely as information from the past), so entropy is a good candidate for the 'hidden variable'. Initial and final boundary conditions then have a common source, eliminating the boundary between classical and quantum physics.