The occam's razor argument against CI is:
1. Quantum mechanics talks about quantum states evolving according to Schrödinger's equation (or similar)
2. CI includes another form of evolution for quantum states (collapse)
3. Collapse, as used in CI, has no observable effect
4. Therefore, CI has unnecessarily multiplied entities, a violation of Occam's razor
As far as I can tell, once you get past the silly arguments, it's just a matter of gauge freedom -- and CI tries to insist that it's choice of gauge fixing is a physical truth. MWI simply studies what happens in a 'frame' where unitary evolution holds good.
See above^. I think it is discounted by being deterministic though is it not?