It's called BPA, which percentage wise is the best potential for return, at least in the first 2-3 rounds. For the later rounds, you can try and draft for organizational or positional needs, but those picks generally have very little success translating to the NHL level. The other option past your first couple of picks is to go after boom or bust players, high potential but concerns over development, foot speed, wants to come over from Europe, etc.
He has tried to draft for the best player that's fallen to him in the draft and has been a little more curious beyond that. There have been some later round picks I have some faith in and others that appear to be outright fliers, but I think he's done an good job at the draft so far.
Besides, I'd like you to back up your claim that mainly undersized guys were drafted. Can you give me a full list and come up with better than 1/3rd that were undersized?
BPA is such an overused term. Every GM uses it and they usually say it after they drafted a player because what else are they going to say? Are they going to say that the player drafted 10 spots behind the player that they drafted was actually better?
YOU think that he's done a good job at the draft so far but in reality, he doesn't have a single player that he's drafted that was on the playoff roster last year. So, exactly how good could he be drafting?
As for a list of players, you can look up the list yourself, as I'm not going to do all of your homework for you but since Gillis has been with the Canucks from 2008 until now, he's had 5 drafts. With the average NHL player being over 6'1" tall and 205lbs, in the 5 drafts he's had, this is his breakdown.
2008 draft - 5 players drafted with 4 of those players being under the league average
2009 draft - 7 players drafted with 5 of those players being under the league average
2010 draft - 5 players drafted with 3 of those players being under the league average
2011 draft - 8 players drafted with 4 of those players being under the league average
2012 draft - 5 players drafted with 4 of those players being under the league average
I will admit that some of these players might be taller than the league average and might fill out and get heavier but right now are under the average weight. But, since you didn't think that I could come up with 1/3 or 33% of the players who were undersized, this above list actually shows that the actual number is 2/3 or 66%. So, in answer to your question, yes I obviously can come up with more than 33% and I can actually double it.