Tom Letcher wrote:
For many years I have been using the formula
(most recent value-oldest value)/oldest value
[....]
I am considering switching to a growth measure
which averages the year on year growth

A simple arithemtic average of each year's growth
rate gives a potentially misleading number, depending
on how you intend to use it.

Consider the following end-of-year values in series:
$1000, $1500, $1000. The yearly growth rates are
+50% and -33%, and the arithmetic average is +8.33%.
But if you used 8.33% as the "average annual growth
rate", you would expect to have $1174 after 2 years,
not $1000.

Some financial institutions do like to report 8.33%.
It certainly looks better than 0% ;-). But it is
misleading.