Re: The filtering is done BEFORE the Bells test
You are missing the fact that Bell's theorem describes a difference between the filter output based on your scenario and the filter output based on nonlocal entanglement. Guess which model matches the actual laboratory results.
Or, are you just pointing out the definition of an observation as an interaction of the observer with the observed, under the impression that this is something that the average quantum physicist has missed?
Sorry buddy, I am signing off now before this gets even more painful.