Ron Rosenfeld wrote:

I cannot reproduce your problem with your posted data. I suspect the
two scores are NOT exactly the same on your worksheet.

1. How are the scores calculated? If these are calculated rather
than simple entries of integers, it is likely that the two 19 point
scores are NOT exactly the same.

2. What do you get if you do an equality between the two cells (e.g.
in some cell enter the formula =S19=S20 assuming S19 and S20 are the
two cells where you have these 19 point scores). If you get FALSE,
then they are not the same and the RANK function would rank them
differently.

Yes, the two 19-point scores are in S19 and S20. Entering =S19=S20 in a cell
returns TRUE. <chuckle What now?

3. The variability that you are seeing in the least significant
digit is inherent in spreadsheets that comply with the IEEE standards
and use double precision math. There have been numerous discussions
on these workgroups about this issue and the matter comes up at least
once a week.

Thanks for what must be an FAQ. I will simply round the percentages to a few
decimals.

4. Rounding is one method of working with these issues.