Actually had time to go over some weather modeling today, caught up on the last week and 1 /2 of runs and I will venture to say, everyone is about to start seeing some very wild solutions. Yes, Wintry solutions. I wouldn't be surprised to start seeing some "historic" type stuff showing up..
Please don't model chase though. They're going to be back and forth in the atmospheric chaos that the modeling is alluding to. I'd use the ensembles for the overall pattern recognition/temperature (PV stuff) and use the operational runs for 500mb energy and ideas at this point.
The higher resolution operational runs are going to pickup on pieces of energy the lower resolution ensembles will just plain miss out on for their forecast. There's a lot of volatility about to float around and small changes to angles, strength, and speed will greatly affect the way these systems track and develop.