Yea, the larger sample size of a full season seems to smooth things out a lot.
What we are talking about here is what all the eggheads who do player analysis for a living are trying to figure out as well. It isn't perfect but they have made progress.
Sticking with the shots, just look at the players on each team and look at their shot supression numbers and compare them relative one another while factoring in usage.
That should give a decent snapshot into how good a player is defensively. All of the xGA and high danger stuff are just trying to figure out how to separate the weak shots from the tough ones really. Bergeron, for example, is near the top of all these lists all the time.