16th. So right square in the middle of the pack. Top 5 teams are Toronto, Dallas, Carolina, Boston, and Philly, in that order. Seems like an almost random selection, no? When you look at the list, it's a very haphazard distribution of good and bad teams. Minnesota is 23rd, Vancouver is 9th. It certainly would seem that FO% has little relation to how much a team wins and losses.
People like to say that, but even if you have a guy like Giroux who is 60%, that's still a 40% chance to lose, which is pretty high. Go less than that, to a guy who's only 55%, and that's a 45% chance to lose. And you only get to flip that coin once, it doesn't "even out". You either win a draw, or you lose it, no do-overs where eventually the better man wins out. So a slight edge in key situations really doesn't matter all that much, a bad bounce or a lucky break for the other guy will cost even the very elite faceoff guys fairly often.
It seems counterintuitive, but faceoffs are glorified coin tosses, that's all. They should be treated as such when building teams. Coaches already treat them exactly this way, they have set plays written up for whether the team wins or loses the draw. Of course, you would rather win it clean, but how a team reacts to losing a draw is far more important than just the win or loss.