Pull data for probabilities of hash changing so we can determine the best bucket list set up

[jedmccaleb]

Part of #118, part of selecting a set of appropriate number of buckets and overflow/merge/rehash threshold sizes is just plain arithmetic on the proposed data structure. Do this and document the reasoning based on size estimates, order and granularity of the structure, etc. Possibly do some simulations of different scenarios.