Excellent post!
Maybe the data will prove me wrong here but I am not the snow climo encyclopedia like Will is, but I would think that more often that not, our winters which are either average in terms of snowfall or maybe a standard deviation higher are composed of numerous small-to-medium sized events. If you want to talk about the upper echelon of winters (your top 3) that's when you need to get multiple large events along with numerous medium sized.
But your point about the deltas...this can't be overstated and quite frankly, isn't discussed enough. Too often things are looked at in a static state...the NAO is this, the PNA is that, here is what the D7-10 500mb mean shows...the static state doesn't matter, it's the delta...it's what is going on between time point A, time point B, time point C, ..., etc. that ultimately matters. It's understanding how these deltas are all contributing to the shape and make of the pattern and then incorporating smaller-scale and even mesoscale features which influence the evolution on the shorter time scale.