Player rating for tournaments across all installments of Heroes.

No system is immune to this. It seems more like you're just engaging in conversation for the sake of it.
And you haven't provided a solid argument in favor of your system. Nor have I seen any drawbacks in mine.

Well, read the thread where rating systems are discussed.

I repeat, the maps are fundamentally different, and it's not entirely correct to tie the results to the leader. It's like considering the gap between competitors in a marathon and a 100-meter sprint. It's absurd.

And when you apply this to a tournament, Radik and Dmitry won't show up. Then Perpolken takes first place, and IronAxe takes second, and I can still get a decent rating EVEN if I'm twice as far behind Perpolken in terms of results (and it doesn't matter if his result is the maximum or not; if it's not the maximum, then it's a complete disaster!). Because, according to your system, the result and its connection to it are meaningless! But the place taken is everything!
And we both took some pretty good places. I'm tired of explaining why this flaw with the PLACEMENT is worse than the flaws in a system that ties results to the leader's result. We're not at the Olympics, and there are no qualifying rounds, so even at BestOfBest, the attendance and quality of players are unpredictable. In this case, it's still more logical to rely on the leader's result than on the places taken, given that players of, say, the middle tier might not show up. And if players of the highest tier don't show up, then the flaw will be apparent in both systems. But with a system that ties results to the leader's result, the problem is already half-solved.
Let's consider your example with the marathon and the 100-meter sprint. On a large map like BestOfBest, you take 30 days, and I take 60. You get a rating of 100, and I get 50. It doesn't matter what my place is; we're comparing results relative to the leader. On a small map, you finish in 3 days, and I finish in 6. You get 100, and I get 50 again. What's wrong with that? And what's wrong with it, in your opinion? I may not have fully grasped what Perpolken proposed. But in my system, where results are compared to the leader's result, the data will be exactly as I described in the example. Read the thread about rating discussions; there are probably 1000 posts there. And they still haven't come to any conclusion.
P.S. As for the example: Let's say you don't show up for the tournament, and the leader gets a result of 50 days on BestOfBest. Then, of course, he will get 100, and I will get more for 60 days. But who will be to blame? Exactly. Dmitry A.S.! He will be to blame in any case IF HE DOESN'T SHOW UP FOR THE TOURNAMENT, whether it's a system based on placement or a system that ties results to the leader's result. And also for starting a biathlon-Heroes-style rating system without reading the thread about rating discussions. Everyone gets a rifle and goes skiing!