I think this part of the discussion has broken down into pedantic nonsense, and so I'll just say that I agree with what you're saying. Rarely does any one player carry their team to the playoffs, and even if that were the case it wouldn't be as important for this discussion as we're not talking about the Hart.
Yeah, agreed. Lets move on.
I use Corsi Rel QOC because it unquestionably tells you whether or not a player is put against the better possession players on the other team. The other types of QOC are either of such minuscule difference to be almost meaningless (TOI based) or weighted in such a way they benefit players on bad possession teams (Raw Corsi QOC).
I tend to prefer TOI based because most teams because the best players on a team almost always get more toi than the weakest ones. The same cannot be said about the best players on a team having the best Corsi rel.
To illustrate, here's a list of the top 10 forward competition according to TOI, and according to Corsi Rel. (40+ games)
TOI | Corsi Rel
JOHNTAVARES | JOETHORNTON
RYANNUGENT-HOPKINS | PATRICEBERGERON
JORDANEBERLE | TOMASTATAR
TAYLORHALL | PAVELDATSYUK
CLAUDEGIROUX | JOEPAVELSKI
ERICSTAAL | ERICSTAAL
NICKLASBACKSTROM | BRADMARCHAND
ALEXOVECHKIN | GABRIELLANDESKOG
KYLEOKPOSO | JAKUBVORACEK
PATRICKKANE | MIKERIBEIRO
SEANMONAHAN | NATHANMACKINNON
MATTDUCHENE | BRENDANGALLAGHER
MIKERIBEIRO | JAMESNEAL
TYLERENNIS | MARTINERAT
SIDNEYCROSBY | BRANDONDUBINSKY
PAVELDATSYUK | MICHAELRAFFL
HENRIKZETTERBERG | FILIPFORSBERG
SHANEDOAN | CLAUDEGIROUX
TYLERSEGUIN | HENRIKSEDIN
STEVENSTAMKOS | SCOTTHARTNELL
RYANJOHANSEN | NAZEMKADRI
JAROMIRJAGR | SAMGAGNER
MARKSCHEIFELE | DARRENHELM
JOEPAVELSKI | CHRISKUNITZ
GABRIELLANDESKOG | JAROMIRJAGR
ZACHPARISE | JIRIHUDLER
T.J.OSHIE | SIDNEYCROSBY
RYANGETZLAF | CRAIGSMITH
EVGENIMALKIN | MATHIEUPERREAULT
JAKUBVORACEK | JONATHANTOEWS
Neither is perfect, but in my mind at least, the Corsi rel list has more anomolous entries. The Rel Corsi list continues to get more and more wonky as you go down, with guys like JVR and Kessel below Clarkson on Tor, Couture and Marleau behind Nieto on SJ, Zetterberg and Nyqvist behind Sheahan Ablekader, and Helm on Det.
Zone starts have a similar problem to Raw Corsi QOC, and that's why in my analysis I like to talk more about how a player fits into their teams' structure than just give raw numbers. Granted, you can get a better idea of what's going on by using relative zone starts. So, saying Doughty starts in the offensive zone about 2% less than his average teammate, while Karlsson starts there about 5% more than his average teammate tells you a lot more than just ZSO% can. LA spent a lot more time in the offensive zone this season than Ottawa did, after all.
Rel zone starts are good for giving an idea of how the coach prefers to deploy players, but does nothing to tell you how "tough" his starts are, but more importantly is the actual effect zone starts have is often misunderstood (idk if that's the case for you specifically).
I don't know if you clicked the link in my post, essentially, the advantage or disadvantage gained by an offensive of defensive zone start has been shown to last about 10 secs. On top of that, there is only a significant advantage if your team wins the draw in the OZ, and conversely, a disadvantage in the DZ if you lose the draw.
So when people point to a difference of 2% points in OZ starts, they are arguing that a player has an advantage for roughly 4 mins over a full season assuming 1200 OZ and DZ starts combined for each player, and 50% on ice faceoff %. That's right, a 2% difference in OZ start % would be a difference of 48 starts (1200 x .02 =24 more OZ shifts, and 24 less DZ, total of 48 delta) with one player having an advantage, only roughly half of which would be a significant advatage where his team won the faceoff. So is 24 starts worth worrying about particularly when you realize the advantage doesn't even last the whole shift? Even in far more extreme cases where there's a 10% gap in OZ%, we're talking about roughly 20 mins of advantage over a season.
The point here was not at all that Doughty > Karlsson because of the D-men they play with. That would be a pretty silly comparison to make. It was that due to unforeseen circumstances, Doughty played minutes far above his own (already high) average this year.
I'm not trying to take away from the good year that Doughty had, I just a lot of the rational behind why it was impressive is faulty logic. Zone starts and QOC are too often overstated. I do however think that the increased workload is relevant, as is the lack of depth on the blueline with Voynov's situation, but I'm not convinced that should be enough to propel him into Norris contention.