I'm not a physicist, but I am a statistician (with an expertise in experimental design). This is a little OT from your direct question, but it might help explain why there is so much disagreement and constant discussion on the impact of pickup position, types of wood, etc., etc.
There are linear, direct effects of things (called 'main' effects in the experimental design/statistical analysis literature) and non-linear effects (one specific type on non-linear effect that weighs on your issue is called an 'interaction'). An interaction occurs when the level of one variable (e.g., wood) has a differential impact on an outcome (e.g., tone) based on the level of another variable (e.g., string material and voicing).
Many times there are significant 'main effects' (i.e., maple boards tend to sound a little brighter than rosewood boards), and also significant interactions (maple boards with pickups that have a more upper mid content seem differentially brighter than rosewood boards versus that same comparison with darker voiced pickups that don't amplify those frequencies, etc.).
As you can see, this quickly gets complicated, and in the example above we are only talking about the simplest of non-linear effects (the two way interaction). The current state of statistical analysis really bogs down above the level of three or four way interactions (e.g., pickup x body wood x string brand x headstock mass, etc.). The problem is, to really understand how this stuff all works, we are probably talking about 10+ way (and probably much higher level) interaction s(body density x headstock mass x neck material x neck mass x string voicing x pickup voicing x etc., etc.). If you take the variation across individual pieces of wood into account, then we are into interactions of a MUCH higher level.
As a result, when you try to understand a non-linear, interactive situation by only talking about linear main effects (body wood, fretboard material, etc.), the results are confusing and seem to not make sense (i.e., some say rw boards are brighter than maple, some say maple is brighter than rw... put in the context of these non-linear interactions, everyone is probably right). This is why it is so hard to predict the weather.... where there are probably 100 way interactions that are 'significant'.
One way to think of the difference between 'science and art' is this non-linearity. Once you get to higher level multi-way interactions (i.e., what makes an instrument sound good?, why do some paintings or sculptures 'work' and look beautiful while others don't?, etc.), it's more gueswork and inspiration... i.e., art! So, IMO, that's why most great luthier's are considered 'artists' and not 'scientists or engineers'. It's also why some of the theories about why some things sound the way they do espoused by some luthiers (even the famous ones) seem silly to some of us!