I have a list of 500 items that I have ranked across 10 separate categories
using the STDEV function for each individual category. I then SUM the
standard deviations of the 10 categories to come up with a "total value" for
each item. I'm trying to rank the items 1 to 500 based on their total values.

Instead of evaluating each of the items against the average of the entire
population, I'd like to have Excel determine the top 100 items based on
"total value" and rank all 500 items against the averages of that subset of
the top 100.

In other words, I'd like for Excel to cherry pick the top 100 items and then
evaluate the entire population of 500 against that subset - instead of
against the entire population. Is this possible? Thanks.