> But this has nothing to do with the determinism of physical models.
It seems rather confusing to state it has nothing to do with determinism of physical models. More accurately, it does have to do with determinism of physical models if you assume a physicalist perspective, but it might not, though then you have to resort to much more involved and comprehensive models of what consciousness and will are (like changing both the future and past).
This is similar with the notion of time. A typical perception is that only now exists. Yet according to physics there is no now. All our physical models based on experience imply that the universe is 4-dimensional static something. There is no now and all points across the time dimension have same properties just as points across space.
One needs metaphysics to try to explain this discrepancy between perception and very successful physical models.
I'm not sure what you mean by this so I'm also not sure how to address it, but it does seem reasonable to assume free will simply does not exist exactly because phenomena is either deterministic or stochastic, not some third option which would allow free will. This view is informed by physics.
> There is no now and all points across the time dimension have same properties just as points across space.
This is a much more interesting problem and one that has kept me up many times.
As for the problem of now, for me it is similar to the problem of free will. Starting from Parmenides and Buddha one way to resolve this was to declare that the perception of now and movement is an illusion similar to the notion of free will. And as with free will, that will be compatible with physical models and the opposite cannot be expressed within physical models.
How do you know, that there is no non-local influence, that makes your predictions "from the patch alone" incorrect? I don't think this can be easily excluded as a possibility.
We can assume that there is no non-local influence and try to make progress from there, but we might be wrong about it, which is what the article is getting at, if I understand it correctly.
Only if you take the experimenter and his decision as part of the local patch, and take the decision to be determined by the same state which also determines the experiment's outcome, which is essentially what superdeterminism is, no?
Experimental observations of violations of Bell equations tell that no, one cannot tell the evolution of the local patch from the patch alone. Standard interpretation of quantum mechanics and physical super-determinism are just different ways to explain this.
In particular the standard quantum mechanics assumes that things are still local, but the wave function is not observable in principle so we can only talk about statistical properties. Super-determinisme assumes that things are not local and tries to explain how.
In philosophy determinism is essentially the opposite of free will. It implies that what people perceive as personal free will is an illusion. But this has nothing to do with the determinism of physical models. In particular, free will is compatible with physical determinism of what one perceives as an external world. One possible explanation of how this is possible is that the act of free will changes both future and past. So it looks like the future state reflecting the choice of will is deterministically follows from the past. It is just the past is different from what would be if the choice would be different. Stanford encyclopedia entry on free will has more splendid explanations.