The NAM is both much drier (only about 1" total precip, which is ~10" if all snow) and has significant % of that precip falling as sleet, due to the upper level low going further NW than anyone expected, given the strong, cold high in place, resulting in much of the areas showing 12-18" of snow by the NWS only getting 4-8" of snow and an inch or so of sleet (equiv to 3" of snow roughly). As I said this morning, I did think Mt. Holly was too bullish on total snowfall and NWS-NYC was likely a better call with 8-14" of snow for NYC/95 corridor.
Whether this is an outlier or the way the other models will go is an unknown right now, but that scream you heard was mass weenie suicide on the weather boards. Need to see if this is "real" or not vs. the other models, i.e., whether there's some heretofore unseen parameter leading to all models trending this way or if it's just some flaw in the data input or model physics with the NAM. We'll know in a couple of hours. Just like one shouldn't overreact to a huge jump up in snowfall for one model, one shouldn't overreact to one model slashing snowfall. Also, at this point, the NAM is often pretty good, so even if it's an outlier it can't be discounted.