"Edgar" wrote
Yes, indeed. It does not make sense to calculate something to many places
of
decimals when the input data is not in itself all that accurate. If you do
that you are just deluding yourself.

Exactly. And one of the first things to do is figure out whether more
accurate data is available, and whether or not it is worth whatever
investment is necessary to get it.
OTOH if you have enough data, the decision makes itself.


The best rule of thumb I can think of off the top of my head is the rule
of
12ths for calculating tidal heights. Works pretty well on the whole for
practical purposes but ignores barometric pressure effects and is not
valid
for ports such as the Solent in Uk that have special tidal effects.
I used it to good effect once in the port of Treguier in France. There is
up
to 30 feet tidal range there. The tide was starting to fall and several of
us were anchored lower down the river eyeing the last good place to anchor
nearer the town. It was clear all were wondering if they would find
themselves aground at low water. I used the 12ths rule and decided to go
for
it and it worked for me as at low water I had 1 foot below my keel.
Everyone
else had a long trip in the dinghy!


I guess it was too late to go back & buy a shallower draft boat...
maybe a centerboarder....

Gilligan wrote:
I have my own rules of thumb for bidding contracts. Once, with a group of
"rocket scientist" we were to bid on a project. They all jumped in
calculating the minuteau of the project - 5 of them spending days. I told
them they were crazy and did my estimation in 45 minutes. We differed in
total by about 1% and categorically were within 5% of each other. 5 man days
vs 45 minutes and got the same numbers. I was never invited to do proposals
again.

A mistake. They should have made you the chief estimator. I adopted
that title informally and thought it was a joke until a few years back
I had to conference with a guy at one of those big-name int'l firms
whose offficial title was "Chief Estimator." How!

signed- Injun Ear (formerly known as Eagle Eye)