If you need 'caveats' to prove your case, then you might want to re think why you believe your guy to be better in the first place.
This is VERY true. With Gretzky, his number #1 position across the board is always 'reality'; with Lemieux, it's always 'yeah, but what if?"
Gretzky was so far ahead of his class, that he was in a class of his own throughout his prime years...and also his ENTIRE playing years (when looked at collectively). It's only natural in sports to want to compare the best with other athletes - or better yet, find a comparable equal for the sake of good old fashioned debate. There's no way any strong argument could be made between '3rd place' and lower players (like Yzerman, Sakic, Jagr, etc), because the distance - and CONSISTENCY year in, year out - between them in numbers and records was simply not there.
Along comes Mario and he's an anomaly who was much, much better than the '3rd place' players, so therefore, a legitimate rivalry is born with Gretzky. Many times Lemieux was even showing signs he was just as great or greater than Gretzky at a specific given moment in a game/season. In retrospect though, when all was said and done and you're comparing everything each player has done in their career, Mario had a shorter and less dominating spread of greatness, when looking at it from a numbers game. And the 'numbers game' is usually how comparisons work in sports to rank, defend, compare players.
After that, adjustments, assumptions, and altered views come into play. And I don't mind that at all. They are fun to look at.
Over the years, listening in and reading about who is better, one thing always stands out: Gretzky believers sit pretty and don't have to fudge or alter anything to state their case, while Lemieux believers give reasons/excuses as to why Lemieux never achieved the things Gretzky did, and ask Gretzky believers (and anyone, really) to look at stats in an altered way, outside the standard.
Just let Wayne be 'great', and Mario be 'magnificent', and be comfortable with the pure, statistical obvious.