On Aug 30, 9:47*pm, "Martin Rowley"
wrote:
"Graham Easterling" wrote ...

I may be missing something, but it's still not clear to me which WEB
forecast you are using as the base. The forecast maps are updated at
different times to the text, and sometimes don't agree.

... the Met Office point-specific web forecasts, example he-

http://www.metoffice.gov.uk/weather/...orecast_weathe...

Are you using Thurday's forecast for Monday, if so which, as they
changed. Or are you using a more recent forecast?

... if you follow the link in the original (and follow-up) post, then
for each forecast issue, it defines when the forecast was issued at
the top of each block. So, for Monday, I'm looking at all
site-specific & regional text forecasts for that day (where available)
from Tuesday last week onwards.

Martin.

--
Martin Rowley
West Moors, East Dorset (UK): 17m (56ft) amsl
Lat: 50.82N * Long: 01.88W
NGR: SU 082 023

I'm following this with interest Martin. I'll bet a lot are. Anyone
who tries to gain an objective appreciation of MetO forecasts deserves
that. At 5 days, I'd expect the MetO to be good, but how exactly how
would we define "good". NOAA are very specific about their success
criteria in their 5-day percentage outcome accuracy assessments of the
models, but what you're doing is different and is just what should be
factored in to any evaluation of a forecast; what the people on the
ground think. It really is crucial to traders on the south coast, as
you say. To judge the accuracy of forecasts, it woukld be really good
to do consumer surveys on crucial days like this. I think the MetO
could get some seriously good press out of it, actually. I know you
are not really going down that line, but your approach is very
customer biased. I like that.