I recently understood Bell’s theorem (the inequality and the QM calculation), with the help of you guys. But something still bothers me: assuming for the moment that Bell’s inequalities were NOT violated by experiment, how we would we understand the dependence of the varying correlations on the angular separation between axes of measurement? How would the hidden variables produce non-correlation at 90 degrees and 50% correlation at 45 degrees? In the QM framework, we just say that the operators for orthogonal measurements don’t commute, and we accept the cosine formula. But what do we say about the hidden variable model? Doesn’t the noncorrelation at 90 degrees already demand something like the “noncommuting operators” explanation?