PER alwasys gets a lot of stick, and I understand why, but I don't think it's that terrible. It does a pretty good job of predicting subjective evaluations of excellence even when PER is retroactively calculated and so couldn't have influenced those subjective evaluations (look at top seasons in PER vs. what people people consider the greatest seasons of all time or PER vs. MVP voting, though now it probably influences voting as well).
If you look at the statistical criticism that your link cites (well the ones that aren't paywalled), you see that there are a few statistical/empirical criticisms about who PER favors and who it doesn't. Basically, it favors bigs who rebound well, finish possessions, but not players who distribute. The former actions basically let your rack up "points" in PER without a lot of risk, whereas the latter is where you expose yourself to negative points. (like turnovers). Players like (prime) DeAndre Jordan are examples of players that PER overrates. DJ was a great piece, but I always thought PER slightly overrated him.
Nikola Jokic - with his high minutes and role as the offensive hub - just isn't the type of player that PER overrates. I think PER is like most other stats in that it can be useful, you just have to understand what it's doing and so its strengths and weaknesses.
I do think that SGA has a very legitimate MVP case. I think his defensive impact in this current Thunder team is slightly overrated. I think he's benefiting a lot in the counting stats essentially taking advantage of the disruption that the other perimeter defenders (Dort, Carusos, JDub) are creating, but he is ideal to do that with his anticipation, quickness and long arms. On/Off, though, the Thunder aren't really better or worse with him on or with him off (partially because they have Caruso coming off the bench). On the other hand, I think his offensive impact is underrated. People talk about the Thunder being a good team without SGA, but they're a good DEFENSIVE team without him. They're pretty mediocre without him offensively and very good with him. That is the jump that is allowing them to play like a super elite team and that's all down to his ability to get their offense humming. When he is on, their turnover rate goes down 2%, their free throw and offensive rebound rates goup 3, their effective field goal percentage goes up 4.4% and they score 14.6 points per 100 possessions more when he's on the floor. That's spectacular (99th percentile in the NBA) and when you're doing that to elevate a team to the levels the Thunder are at, that's a legit MVP case to me. Cleaning the Glass's expected wins added by on/off is 26 for Shai. That's monstrous, especially as he does have good players around him, but he's what makes things tick for the Thunder. I might vote for him if I had a vote and in most seasons it would be an absolute no brainer.
Except the Nuggets score 24 points more when Jokic is on the court, have a bigger bump in defense (partially because they are thinner backing him up), and his on/off numbers would lead to an 56 game difference in record.
So i don't know who I would vote for. Probably SGA if the record gap stays similar (percentage wise), but if the Nuggets tighten it up, Jokic's amazing efforts might tip me toward him.
