Looks like the GFS is too shallow by at least 4 hPa but that's not unusual. The GFS T+3 did not quite match the WV imagery at 15 UTC. Haven't looked at the ECMWF. The GFS handled the Zeus storm, coming out of the same area, earlier this year quite well at short range, but not a similar one in 2015(?). One area where small errors (ie at the mesoscale) might impact is on any snow forecast for the SE tomorrow.
19 UTC- cloud hook formed, likely sting jet in the next few hours but only over the sea. Cold conveyor belt will produce the strongest winds over land.,