In article .com,
says...
The data I used are the best available to me.
The data are from the NOAA web page given at this URL:
http://www.aoml.noaa.gov/hrd/hurdat/...arandStorm.htm
The number of storms were counted on the yearly
maps provided.

If you have better data for Atlantic Basin huricane numbers,
Mr. Bombast, please post them here. Realize that your
data would have to significantly differ from those i used
in order to change the 1 in 10,000 event calculation.
The data I used are below:


[Data deleted]

How are you getting 1 in 10,000 years?

Using the more reliable period (1944-2004) from the data you posted, I
plotted p(x=X) vs X, where X is the number of named storms. When I fit
log(p[x=X]) vs X with a linear regression, I get a return period of ~
162 years for X=27.

Harold
--
Harold Brooks
hebrooks87 hotmail.com