I don't think you can do better than brute force search for cryptographic randomness, but it would be interesting to see if this does well on personal heuristic

I don't think you can do better than brute force search for cryptographic randomness, but it would be interesting to see if this does well on personal heuristics people use. Eg the XKCD approach.