It could be that Twitter is not really in your repertoire and that might be why you're not seeing the discussions between the Pension Plan Puppets crew, Mirtle, etc. on Corsi. If that's the case, then I suppose you'd miss these conversations that have been going on all season.
But sure, we can skip there.
I think you can overcome poor possession with a) great goaltending b) pure skill up front / a team that can capitalize on it's transition game and odd man rushes and c) good special teams. Or at least, that's how the Leafs have been doing it for 115 games now.
I only follow Twitter for Richard Sherman.
I certainly think that there's something to the goaltending aspect; if we believe that save percentage is a true measure of goaltending skill (to some degree) then there shouldn't be a reason why that aspect of PDO should regress entirely (I do 99% of my analytics on the goaltending end, so if I'm missing a fundamental aspect of PDO, someone please speak up). There *is* a significant amount of variation in save percentage year-to-year (as with any binomial distribution), but there's an underlying probability driving it.
I assume that when you say skill up front, you're referring to the opposite side of the coin (if you have players who consistently find ways to exceed a league-average shooting percentage, then the same argument would hold as what I describe above for goaltending).
As far as special teams are concerned, is PDO inclusive of special teams (or is it even strength only)?
Beyond that, it's still perplexing that a team that's consistently outshot can post a good record, since ultimately (goals for) = (shots taken) * (shooting percentage), and (goals against) = (shots received) * (1 - save percentage). It's clearly something to be studied further.
On the other hand, if we perfectly understood hockey at this point, then where would the fun be in that?