I follow the old rule- models are good for extrapolating known data at varied points in time; however, with each update you get new variables that affect the equations- therefore, the solutions, therefore causing newer tracks. I will believe the modeled tracks when Dean is at least 3 days from impacting my area.

Until then- prepare and enjoy the genesis of what will be a good show and hope for the best - to all of us!