Just b/c a model goes from "storm to no storm" and vice-versa doesn't mean a model sucks. This is one reason why I think there are just too many products available to assess. In the medium range the only products that hold any weight whatsoever are mid/upper-level. Forget what the snow maps show, forget what the QPF maps show, and forget what the SLP tracks are. All those products...are useless and pointless in the medium range.
You have to understand why a model is wavering back-and-forth. In the case of this potential, there are several key pieces at hand
1) PV and associated confluence
2) Shortwave energy within the north stream
You can even expand this list and focus how the energy moving into the PAC NW is influencing the evolution of the pattern downstream
It is unrealistic to expect a model to be consistent or sniff these features and their exactly evolution nearly perfect in that range. There is too much chaos involved. That will never, ever, ever happen.
When looking into the medium range, all you want to see is how your key features are being played out by each model while using the current pattern regime to hopefully narrow down a list of potential solutions/outcomes.