The kink south to the surface warm front is likely severely under-modeled. Cold Air damming and ageo flow along the Apps—this looks like a classic example down to the mid Atlantic.
At the margin this will likely have important implications and cause the surface low to travel southeast of I90. And how this feeds into the potential development of a mesolow near 40N, mitigating easterly fetch, we could still be looking at cold surprises down to I95. But this latter points seems far less likely at the moment, outside of persistence factors—climo, snowpack etc...
A surface low track halfway between I90 and I95 —or thereabouts—is my best guess at the moment...Hedging colder than guidance...