In looking at the 00:00Z forcast tracks at:
http://moe.met.fsu.edu/~acevans/models/al912007.png
I see that the models are in tight concensus until the storm reaches 18N when they seem to be evenly split between a more northward and a more westward track.

This seems to coincide with the plots from 18:00Z at:
http://www.skeetobiteweather.com/picservice.asp?t=m&m=91
where the GFDL is making a jaunt to the left.

There's a lot to be learned from these storms in their infancy.

As of 11PM Eastern I can see that there is significant convection developing in the northern and western quadrants of the storm plus some early indications of banding. I would concur that this storm should be a TD no later than 11am Eastern tomorrow.