One thing that concerns me is inland flooding potential. In the case of the 18Z GFS run, this is in MD, VA, NJ, and southern PA. The GFS 18Z run is predicting landfall on the NJ coast, moving inland, then sitting for nearly 36 hours nearly stationary somewhere in either northern MD or southern PA, sucking in huge amounts of tropical moisture and dumping it. I don't trust the total precip predictions since they usually underperform tropical systems; these predictions are showing 3-5" of rain for this region. For 36 hours under the center of a tropical or subtropical cyclone, that seems low.