Another obvious reason to put the squeeze on vol

I forgot in my last post on geometric returns (Some Thoughts On the Trade-offs Of Strategy Switching or Aggressive Tactical Allocation and the Virtues of Sometimes Taking a Tax Hit) about the most obvious thing about being aware of the longer term effects of geometric returns: choosing between two strategies that have more-or-less similar return profiles but radically different vol profiles.  One strategy might always be expected to win and it might not be the higher return strategy.  For example, using the same very-rudimentary back of the envelope excel sim (1000 random returns per period, 20 periods of  geometric compounding, all 1000 paths averaged to an expected value, and then the whole run x times) and two vaguely similar strategies: a) 7% return, 20% std dev, and b) 6% return and 10% std dev it could look like this:

Lower return wins. And it wins within a time-scale (on this run anyway) that can make sense with respect to a retiree's planning horizon.

Now, the individual paths could be insanely different (see notes in the referenced post) and each of the measly 100 times I did this all looked different, too... but 7% won or tied only twice at the 20-period expected value level out of a hundred times I ran it .  That may have more to do with my sketchy excel modeling but it is, probably, indicative of general expectations.  This would play out well on a mean variance map too but this is another analytical way of thinking about it.