Simply put, this is not a clear-cut pattern we're dealing with as there are a lot of moving parts in the atmosphere. On the chart above, look for #1, #2, and #3 and we'll parse the differences between the two models. The GFS is on the left half of the graphic, the Euro (EC) is on the right.
1) The "50/50" low near Newfoundland. Conventional weather wisdom says you need a 50/50 low in place to help steer the storm on a more optimal track for snow and/or rain. However, the low needs to be in the right spot, at the right strength, and not too strong or too weak, too far east or west to prevent the storm from taking a suppressed track (too strong, too far south) or bring rain to the East Coast (too weak). The Euro has been showing a stronger 50/50 low over the Gulf of St. Lawrence (blue shading), which is hardly the right spot for a mid level low to be in. The low, optimally, should be near/over Newfoundland if possible. The GFS has this low in a better position; however, it is too weak. Because the Euro is farther south/west with this feature, as well as stronger, the result is a storm track that is farther south of our region.
2) Northern energy. The GFS is more aggressive with digging a trough down through the Midwest on Saturday (see the greens back over the Great Lakes). This trough helps provide the fuel for the storm and provides energy for it to intensify as it treks across the country. The Euro is weaker, less amplified with this northern energy (yellow shading, not green), and as a result there is not a full capture of energy between the two streams of energy and the model merely scoots the low eastward.
3) The resulting ridge in the Atlantic. For every action, there's an opposite reaction. Since the trough in the Midwest on the GFS computer model is stronger, the resulting ridge in the Western Atlantic is "pumped up" a bit. This allows for a channel for the storm to track between, giving you the rain and snow the GFS is predicting. The Euro, because of the lack of troughiness in the Midwest, is less amplified in the Atlantic as a result. This means a flat storm track and a lack of a drive up the East Coast for the low. More importantly, the ridge can't be pumped up in the Atlantic much because of the stronger 50/50 in the Gulf of St. Lawrence in the Euro depiction. This is adding to the suppression tactic that the model is showing.
This low that forms the storm system that may hit on Sunday is in Southern California today and will begin to trudge east across the country's southern half. Now that the upper level feature is on Shore, the next piece to fall in place will be the northern energy. That piece is over Western Canada. The energy that results in the 50/50 is currently over the Southern US and will be the rain event that we get tomorrow afternoon and evening.
That's pretty much why you have a 500 mile difference four days out on a storm system that will track through the southern half of the country. Even if the GFS is right, it's not all snow...especially along the coastline. There is no real pump of Canadian chill coming in and we would have to rely on the storm bombing out, drawing down colder air from aloft, to get a snowy solution. This storm might not be strong enough to pull off the trick for the coast. Inland, especially farther north and west of I-95, is in a better spot IF the GFS from today's midday run is right.
Hopefully in 24-36 hours we'll have some consensus on what all comes together between now and then.