The way I think about it is that you start with a tie matching Skjei one good (his best) season with Yandle’s. Then you look at other factors as the “tie breakers”. To me these factors were Yandle’s use by AV on one hand and the fact that Skjei’s other seasons were mostly underwhelming (unlike Yandle’s). That’s why I think myself and a lot of other voted for Yandle even though he played significantly less in terms of sheer number of games.
I just think if we consider use a tiebreaker, we'd have to factor in Skjei's use as well. I also think it becomes hard to compare underwhelming to non-existent.
Let's say their best season is a virtual tie. Skjei didn't have more good games in the additional 170 he's played than Yandle did in the additional 21 he played? I just don't see that.
So for me, that's where the sample size for Yandle is just too darn small for me to consider the whole "peak" approach.
Essentially, their best was a tie, and the rest goes to Skjei by virtue of playing nearly two additional seasons worth of games.
And interestingly enough, kind of lost in all this, is that Skjei is actually on pace for a 13 goal, 32 point season right now. So while no one really seems to believe he's the long-term future as a first pair defenseman, I think we're kind of blowing past that he hasn't exactly been without value either. It wasn't all concentrated in his rookie season.