It was really the HRRR, Nam 3km, SREF(blend), and a couple of RAP runs that really nailed this one with the SREF really catching the extent of the snow to the Northwest. Credit goes to the Nam and HRRR for catching the warm nose and the undermining of dry air(nam3km) beginning over the southern plateau. HRRR 11z Thursday Kutchera snowfall was the ultimate and overall winner. Merit hasn’t been lost in the other models for short range, however. This was a really unique event with an especially difficult setup for the valley. Perhaps sometimes in the past you have seen the exact opposite happen: with the HRRR and Nam3km showing a drier and warmer profile than reality with precip and cold winning the battle. Perhaps the Nam 3km isn’t the most intelligent model overall, just in this case. It was just it’s nature to catch a warm frontal sector of air and that’s what happened. I think terrain over the plateau, SE Ky Mtns, and SW VA Virginia Mtns and the smoky mountains may have had something to do with this outcome. This also may have something to do with the idea of the storm's fast movement trumping the forcing, as MRX pointed out. If the entire southeast US was flat, I think most East TN wouldn’t have seen the 4-7 KY saw but instead may have seen a widespread 2”-3.5.” The smoky mountains funneled warm and dry air advection north which in turn allowed a broader surface low to propagate slightly further north and east than expected. The plateau and Cumberland Gap Mtns enhanced the temp and dry air gradient with, generally, those to the north of the SW VA, SE KY line seeing a fully saturated column with suitably cold temperatures which also enhanced rates and total accumulations.