http://cstpr.colorado.edu/prometheus/?p=3436 Wed, 29 Jul 2009 22:36:51 -0600 http://wordpress.org/?v=2.9.1 hourly 1 http://cstpr.colorado.edu/prometheus/?p=3436&cpage=1#comment-970 Harold Brooks Fri, 25 Mar 2005 14:47:04 +0000 http://sciencepolicy.colorado.edu/prometheusreborn/?p=3436#comment-970 That's a terribly disappointing decision. For a number of years, I've hoped that they'd make a number of changes to their forecast products and this would have been the minimum. I'd also like them to stop using the historical forecast errors as the only input into the probability ellipses and actually allow for case-to-case differences in the confidence. There's a list of other changes I'd make, but adding the historical error statistics to the single line seemed like a minimal change, especially in this graphical age. There's no reason they can't put a line on the probabilities to indicate the best guess track. I like the last line in the article: "Encouraging people to pay less attention to the skinny line is one of themes of the conference." I guess it isn't a theme of the NWS. That’s a terribly disappointing decision. For a number of years, I’ve hoped that they’d make a number of changes to their forecast products and this would have been the minimum. I’d also like them to stop using the historical forecast errors as the only input into the probability ellipses and actually allow for case-to-case differences in the confidence. There’s a list of other changes I’d make, but adding the historical error statistics to the single line seemed like a minimal change, especially in this graphical age. There’s no reason they can’t put a line on the probabilities to indicate the best guess track. I like the last line in the article: “Encouraging people to pay less attention to the skinny line is one of themes of the conference.” I guess it isn’t a theme of the NWS.

]]>