The idea of following science and not models is something that good mets do. The new guy at Millersville took the time to explain the why behind it - good read in words that all can understand:
It's no secret that there's been a lot "hype" circulating over the potential for an early-season snowstorm in parts of the mid-Atlantic and Northeast early next week. In situations like these, it's very important to consider what's most likely to happen from a climatological perspective. Late-November climatology tells us that it's rare for a major snowfall to impact the Lower Susquehanna Valley and I-95 corridor from New York to Washington, D.C., and large-scale features in next week's weather pattern aren't in the right places for this rule to be broken. To understand why this is the case, it's critical to discuss the North Atlantic Oscillation (NAO) index and its role in promoting East Coast snowstorms.
The NAO is calculated based on the pressure difference between Iceland and the Azores Islands. A neutral or +NAO is often more common, given that high pressure often resides over the Azores with low pressure over Iceland and Greenland. However, a -NAO features a reversal of this pattern (high pressure over Greenland/Iceland and low pressure near the Azores). One of the most important players for mid-Atlantic and Northeast snowstorms is for this "blocking high pressure system" to develop, resulting in a -NAO. However, the presence of a -NAO in and of itself doesn't necessarily mean that a Northeast US snowstorm is a "lock." It is very important to look at where the -NAO is based.. is the center of the blocking high over Iceland/eastern Greenland (east-based -NAO) or to the west of Greenland (west-based -NAO)? An east-based -NAO does little to promote big, East-Coast snowstorms since low pressure systems won't feel the influence of the blocking high until they reach northern New England and the Canadian Maritimes. East-based -NAO patterns often result in storms tracking from the Midwest into the eastern Great Lakes or interior Northeast, leaving the mid-Atlantic and much of southern New England on the warm and rainy side of the system. A west-based -NAO, on the other hand, is much more conducive for big storms since systems feel its influence sooner and tend to track farther south across the U.S, slowing down upon arrival to the mid-Atlantic coast.
In this case, the NAO is currently positive but will slide into its negative phase by the early to middle part of next week. However, it'll be an east-based -NAO, and the blocking high won't reach its maximum intensity until Wednesday or Thursday. So, why didn't I jump on the "hype train" earlier in the week when some computer models were showing a major nor'easter with flooding rain and heavy snow in parts of the Northeast? First, as I already mentioned, climatology simply doesn't support it. More importantly, these computer models were showing the same NAO evolution earlier this week that they're still showing currently.. but were somehow suggesting a historic storm for the Northeast. I always look at the large-scale pattern and ask myself if what the models are showing is reasonable in relation to the state of that pattern. Instead of assuming the models are right, I ask myself where they may be going wrong. In this case, the "historic nor'easter" idea simply didn't align with the state of the pattern in the North Atlantic, let alone the pattern in the western U.S. and Pacific Ocean (which isn't at all conducive for early-season snow).