Cheating Bill G, yes the answer is obvious and trivial in QM and matches experimental results but they don't accept or understand QM
It can be described by QM but the reason is purely conservation of energy, the initial EM emission only has a discrete value of energy available. The problem is simple (it is the same as my RF harvester example) you reach a point very easily where you have more observers absorbing the energy than you have available at the emission point. With a single photon anything more than one observer creates the condition. If each observer really could get the same energy you would violate conservation of energy and make no mistake that is what is behind the strange behaviour.
Paul doesn't believe in QM and that is fine we don't need to convert him. I encourage and support him trying to find an alternative answer and anything short of denying the experimental results is fine with me. The experiments are dead simple and are repeatable in millions of ways and all give the same result, pretty much because noone has ever been able to beat conservation of energy laws.
The problem even comes up in cameras and eyes which should not work if light is a wave like Paul wanted. With a camera or eye looking at the night sky the wave should come in thru the lense be focused as a spread area on the view area and the whole area should see the light. You wouldn't be able to localize it to a point. Go to an optical site and they have to draw the light as a ray (a transverse wave) to make sense of the results, you cant draw it as a spherical emission wave. The real answer is the spherical wave does enter as Paul imagined, if the whole plate saw the photon and all absorbed it we again have the violation of conservation of energy. The shortest path on probability will see the photon (which matches the ray description) and that pixel triggers and so the result localizes to one point. We in fact have experiments where we fire single photons at a photographic plate and the result is indeed a scattering of the dots the highest probability at the shortest path. So a photographic image is never perfectly clean because it does involve an element of probability. Much harder to explain all that to a layman and so we cheat and turn light to a ray representation when discussing optics. Most layman never pick up there is hand waving involved in normal school level optics.
Prof Christopher Baird who Bill S linked goes the same path I would it is better not to lie and hand wave when discussing this stuff. He also like me stopped short of dragging QM into it because it isn't required for an understanding.
So none of that behaviour requires a QM description and Paul can have his spherical waves so long as he remembers the waves are subject to conservation of energy. Once the energy that was emitted has been absorbed however he wants to describe it the spherical wave must cease everywhere else because there is no energy left. If he does that he will derive the same answer as the experimental results.
For the record the cameras do not have to be the same distance that just changes the probabilities. The closer camera is more likely the further camera less likely but the probability change is steep. That result is also consistent with a single photon fired at a photographic plate.
