On 8/22/11 I saved a screen shot of the forecast model tracks for Hurricane Irene. See below, and click on the image to enlarge.
It is interesting to compare this with the actual track of the storm as shown below.
Of the various forecast models, the NGFDL seemed to have a good fit as the storm approached North Carolina, while the GFS had a good fit during the later path of the storm. According to what I saw on TV, the model tracks were all pretty close together as the storm traveled up the east coast. Naturally, small changes in the track's position east or west made a big difference in storm surge at the local communities on the coast.
Note that 8/29/11 was the 6th anniversary of Hurricane Katrina's landing on the coasts of Louisiana and Mississippi.