As in if you agree something is volatile, can it be correlated accurately? Instead of putting Bergeron as an example, maybe look at the more average NHLers and see how their +/- could be year after year, or whether their first half +/- performance could be carried over to their second half +/- performances.
I think you're confusing the word correlate and the word predict. Correlate just means there is a relationship. i.e. Two variables are not completely independent. It doesn't really make sense to say "can it be correlated accurately". The closest thing to do would be to ask if it is a strong correlation, but I never made any claim that it was. In fact, I did the opposit, because I said it's highly volatile so predictions are less accurate. That's literally the definition of volatility ("liability to change rapidly and unpredictably").
Regardless of the strength of the correlation or predictability though, the definition of "pace" is a rate of progress/performance/delivery. So yes, the rate at which he has accumulated +/- so far this season is indisputably at a +50 pace over an 82 game season - that's a fact, not an opinion. Saying someone is doing something at a certain pace doesn't mean that you think they will finish at that point, it just means that's the rate they've gone at so far.
All that said, what I do believe and was implying, is that W. Karlsson has a really freaking good +/- so far this year, and that should have value because it has some correlation to future +/- (albeit not as strong a correlation as points). Obviously getting a player who is +40 is better than getting a player who is -40, all other things being equal.