Are you trying to explain the Born rule, by producing a strictly real-valued superposition of universe-wavefunctions, which therefore looks more like an ordinary probability distribution?

ETA: I have a downvote, I don't know why. My very next observation would be that you would still have to square these real-valued amplitudes in order to get the probabilities, possibly followed by a discussion of the prospects for doing QM over algebraic fields other than R and C, and whether this changes anything.