The official 5am forecast discussion for hurricane Ivan contains this quote: “If I did not have a previous forecast to maintain some continuity with. I would have shifted the track even further to the left.”

Professionalism assures that these guys are striving to output the very best forecast they possibly can. I assume they know that you get better forecasts if you don’t start from scratch every day. Better may mean more effective, or it might mean more accurate; I wonder.

But do realize how broad the distribution of predictions really is at this point. They may say they are predicting land fall near the border between Alabama and Mississippi; but really they aren’t. That statement is just the highly condensed version of a set of predictions. This enumerates their real prediction. For each location along the gulf coast it has their current probability that the hurricane will make landfall close to it. None of them rise above 16%. At one end of the coast line the chance that Key West get’s it is 9%; while at the other end all the way over in in Galveston Texas it’s 6%. New Orleans is 14% only 2% lower than the center of their forecast landfall probably curve.

That means that shifting the center a bit left or right would only change the threat level for these locations by one or two percentage points.

Leave a Reply

Your email address will not be published. Required fields are marked *