The other thing is that a lot of the forecast models we use don't always handle all of the physics scenarios and have to simplify some of the calculations. That especially comes into play when there's a complex system like this where just a little bit of a difference in any direction can change the outcome drastically. This storm we're talking about is still a few days out. From my experience forecasting, and admittedly it's been a few years since I've done this regularly, that 4-6 day window is notorious for being imprecise.
The data I am looking at does indeed look like this could be a significant ice storm generally along the mid-South. That time frame is only now starting to come into range of the shorter-range models (out 2-3 days). which generally have higher resolution and handle the complex physics better.
The other caveat, though, is that, well, this is the south we're talking about. The ground is inherently warm, and that below-freezing layer you need for freezing rain usually happens when the air has already been below freezing for at least a few days and the ground is frozen. We don't have that scenario here. So if there is an ice storm, it's more likely to be in the form of graupel or sleet pellets that freeze in the air.
