Many thanks again to Grandfather and Limbic~Region! Both for the bug find and the sub-optimal examples.

Both examples are food for thought. Limbic-Region's distribution is interesting because it is a good example of where looking ahead only to the next largest deviation does not yield an optimal result. The smallest sized buckets can only be optimally filled by ignoring 6.0 even though it is smack on the mean (a deviation of 0). Grandfather's "nasty mark II" shows clearly that least(greatest?) common denominators (i.e. 2) and not just individual bucket size or frequency of bucket size affect the result.

Best, beth