This multi-model comparison above shows the GFS model in the top left, the GFS Ensembles in the top right, the ECMWF in the bottom left, and the UKMET in the bottom right. Because this is out of the UKMET's forecast range, there is no forecast shown. These maps are of the 500mb layer, commonly used to identify areas of high and low pressure. The GFS/ECMWF/GEFS are in good agreement on a system coming up the East Coast. Exact track is uncertain. The ECMWF delivers another worst-case scenario storm system to the Northeast, pounding the region and putting a big setback to recovery efforts of Superstorm Sandy. The GFS takes the system more offshore, while the GEFS doesn't appear to be in good agreement, shown by the lack of deeper coloring in the top right image.
I glanced at the ECMWF ensembles, and they are just offshore. It appears that the ECMWF model is the only one actually onshore, and this hurts its forecast credibility in my eyes. Another thing putting down such a solution is a neutral NAO going positive, which can deflect storm systems more out to sea than on to land.
At the moment, I believe the positive NAO and GEFS are right in that the system should be offshore, but not too far offshore. I do think that the ECMWF needs to be watched, as it could easily win this forecasting contest if I see more conclusive evidence from the GFS/GEFS.