Nav wrote:
For a limited range of outcomes optimization of outcome is always best.

In theory, yes.

In practice, when the stuff hits the fan, there is no time to draw up a
flow chart and calculate a risk/benefit chart.

You should know this -it's a fundamental tenet of good engineering
practice isn't it?

I also know a lot of very smart engineers who make poor sailors. They
tend to over-analiyze everything and either take no action or try to
change their action in mid-course. Doctors are worse. The way to skipper
a boat is to be able quickly conceive a plan and follow it through,
whether it is "best" or not.

In theory, there's no difference between practice and theory.
In practice, there is.

DSK