I'm well aware that Kane has made a career out of being a statistical outlier when it comes to his own production. It's true that prediction models for production don't really work with him, simply because he's constantly scoring more than his possession, shot and chance generation would suggest. That's great for him, and makes him a dangerous weapon, as he can consistently score against the grain.
But I'm arguing that Kane's underlying numbers suggest that he's not going to score as much. We're discussing a fixed window of performance, not a projection of future performance based on predictive measures. And even if I were making a projection, I wouldn't argue that Kane's individual production would suffer significantly due to him being underwater. He's proven he can overcome that.
I'm talking about on-ice impact as far as shot and chance generation (and ultimately goal-scoring) both ways. It's great that Kane can continue to generate even when he's underwater, but it doesn't change the fact that the team is consistently under-water when he's on the ice. That means more shots and chances against, and inevitably that means more goals against, m especially against elite players that can make the lop-sided shots and chances in their favor count more often.
We Blackhawks fans know this, we know that's why Kane was deployed on a 2nd line almost exclusively in the offensive zone against weaker competition, while the 1st line took on the top competition and 4th line (really 3rd) took on the defensive zone duty. That left Kane to get the most out of his offensive upside, and kept the guys that would have the best chance to use the lop-sided shots and chances the other way against him off the ice.
As far as career vs career, I'd obviously rather have Kane, predominantly for the playoff performances he's provided. But between these two, fixed, somewhat arbitrary time periods, I'd rather take the guy that doesn't need to be insulated as much because he drives play better, and only loses 10 points of raw production in the process.