How does disregarding realism but maintaining locality explain the Bell Experiment? If we conduct a simple Bell experiment, we can show that "hidden attributes" are inconsistent with the probabilisitic distribution of results that we get in an Alice/Bob type game played with quantumly entangled bits. This implies that we must disregard either realism or locality, and most sources that I have consulted disregard realism. However, how, then, does this explain what is going on in in the Bell experiment? If we disregard locality, we can say that ok, maybe the two entangled bits in the quantum version of the Bell experiment communicate non-locally with each other and this is how they are always in sync (this doesn't need to violate relativity since perhaps they communicate without transporting energy between themselves). This explains our results. But what does disregarding realism give us? How does this help explain the Quantum Bell experiment? Even if the bits don't exist in any state until they are observed, they're still going to have to coordinate things with each other, right ? (well obviously not since we accept locality but disregard realism, but I cannot see how this is the case) 
 A: Quantum mechanics is a theory which is local but not realistic, and it explains the Bell experiment. 
A: Disregarding realism is the key concept of Copenhagen interpretation of QM. Following Copenhagen interpretation, Bell experiment shows that if we wish keep realism in QM, we need to admit non-local, or faster-than-light, transmission of hidden variables, which contradicts special relativity theory.
Disregarding realism does not explain Bell experiment but allows to discuss it without contradictions.
A: Marcel Mazur,
"we must disregard either realism or locality, and most sources that I have consulted disregard realism"
This is false. The EPR argument, more exactly Einstein's reality criterion proves that QM must be either incomplete or non-local. So, the only way to retain locality is to have deterministic hidden-variables (realism).
Bell's theorem is a refinement of EPR. The theorem is based on two assumptions (locality and statistical independence). So, the only way one could keep QM local is to go for a hidden variable theory that violates independence (the so-called superdeterministic theories).
"Even if the bits don't exist in any state until they are observed, they're still going to have to coordinate things with each other, right ?"
Right! You will not see a local, non-realistic explanation of a Bell test.
