As a huge fan of research and interest in long-range forecasting, there needs to be more studying and analyzing patterns as a whole. When it comes to research, whether it be with severe weather or winter storms, etc. there is a focus to just look at the events which produced. Not to say this is a bad thing. It's what needs to be done to understand these systems and help with forecast awareness, but when you're ONLY looking at what produced in the outcome you're missing a whole lot else.
How many times do these "good patterns" really produce? Is it 70% of the time? 50% of the time? 5% of the time? I mean in reality, if the percent probability is very low, can it really be characterized as "good"?
But this is also why a pattern only means so much, at the end of the day it's how the pieces are moving, evolve, and interact within the pattern that are going to determine the outcome. And you can have the same pattern 50 times, how these move, evolve, and interact, are going to happen 50 different way.