Stan,
Ref(2) , Rank = {( Latest Skip - Min Skip) /(Max Skip - Min Skip) *100%}, Immediately you will notice this statistic at the numerator is incorrect, it relates to the presumption that the 'Latest Skip' in the skip bandwidth is representative of the occurences . This is incorrect as there is no such relationship, worse still it does not account for any repeats. Unfortunately this ordinal statistic is on all statistics pages, which I believe has not been properly concieved or tested. I do agree this statistic needs revamping and do hope we all get the opportunity to put it right.
Ref(3), I thought the problem was the calculation of 'No of possible occurency' for 'sum' was difficult, because EL calculates statistics for each sum value, and perhaps this is computationally intensive. My suggestion was simply to reorganise that statistics page similar to the other pages for sum values in ranges such as (0-20, 20-40,40-60,60-80,80-100,100-120,120-140,140-160,160-180,180-200,200-220,220-240,>), you are left with 12 rows of statistics as opposed to 220+ rows as shown currently. I assume the calculation of all the rest of the statistics are point parameters hence should be not be a problem.
In my way of thinking we need not take up any real estate, only three indicative statistics (Rank,Expectancy or Probability, Strength) are required to enable useful analysis of the filter and package statistics. I do hope we will have the opportunity to revamp these pages by general consensus?.