Player rating for tournaments across all installments of Heroes.

My last post from work was interrupted, and I lost my internet connection. I was editing my last message like this:
-----------------------------
They both competed against each other in both the easy and difficult maps and received corresponding ratings that no one and nothing can influence because only their pure results in days are taken into account, which depended solely on their skill in playing Heroes, regardless of the time constraints, since there is no need to focus on secondary parameters. And in your example, you say that Player A played in Tournament 1 and did not play in Tournament 2, while Player B played in Tournament 2 and did not play in Tournament 1. Nothing will help ANY SYSTEM without a map difficulty coefficient. Because we assume that there is also Tournament 3, where neither A nor B played, but a newbie played and, with a result of 1000 days on a super-easy map, receives 100 points according to my first system, just like players A and B in their Tournaments 1 and 2. The philosophy of THIS rating system is: who is to blame for the fact that the newbie received ONE HUNDRED? Answer: players A and B. One of them should have participated in the tournament. Since they didn't, accept the newbie into your group and then be frustrated by his rating :)

And if you take a system with a Virtual component, then in this case, only Player B will suffer slightly. The philosophy of THIS rating system is: why did Player B suffer? Because he missed Tournament N1. Well, Player B will now have to deal with Player A and wait for him to miss a tournament with an easy map and get his revenge. Although, in reality, look at the real tournaments that have taken place on the forum. The results of the leaders on the maps do not exceed a 2-week period.

The philosophy of both rating systems is that you shouldn't miss tournaments :) I'm at a crossroads now, but for now, I'm thinking of sticking with my FIRST system (without the Virtual component). Because I'm leaning towards the conclusion that there will definitely be someone from the more or less decent and strong players who will show a normal result (even if not ideal) in the tournament. And I think that a system with a Virtual component will be worse because the probability of different maps of different difficulty being present in different tournaments is high. And then you need to come up with a map difficulty coefficient and just give up :)

Added 14 minutes ago

Hey, buddy... This was said in another thread, and it was about the clan rating, where the question was that, in order to increase the clan rating, a team can easily discuss the walkthrough of a tournament before it ends... And therefore, the very idea of calculating the clan rating may be at risk...

Hey, buddy... And with your system with an oscillator, this team can also lower the rating of the "right" guys :) What are you not saying? :)

:smile19: Deleted :smile19: Kyle: off-topic!

words about the oscillator? Just to once again show the "advantages" of your calculation method? You clearly doubt it, since you tirelessly advertise it... Iron, don't change the subject and don't try to mislead people...

And you don't avoid direct questions either. I don't need advertising. But I showed the flaws of your oscillator, I'm showing them, and I will continue to show them, no matter how much you don't want it.

If you couldn't find a relationship between the parameters, that doesn't mean it doesn't exist. These parameters are related to each other, and not taking them into account is at least incorrect. The question is how? That's what we're working on. You are suggesting such a scheme: "I'm afraid that I won't understand your method and you will deceive me, so let me calculate it myself, and everything will be simple and there will be almost no mistakes"... These are your problems to believe or not to believe people who are doing calculations. And don't put pressure with the number of posts and words per unit of time. 90% of your statements are aimed at discrediting alternative calculation methods, while there is simply nothing left to say about the essence...

This is about the safety of your system :) I understood everything in your system, and it's flawed :)

Your idea is clear and will be considered. In addition, there are others. There is no need to reject them for far-fetched reasons. Please answer the question directly, and do not throw words around as you usually do.

You tell yourself that, :smile19: Deleted :smile19: I answer all questions promptly. "Your idea is clear and will be considered." Three times ha-ha. Don't feed me breakfast. You want to gloss over the answer to a direct question :) You perfectly understand that if the system takes into account the place, then the method of artificially lowering the rating by a group of players that I described, you will not be able to eliminate in any way :) And in this system, the player will not feel that his rating for the tournament is protected. :smile19: Deleted :smile19:

Added 13 minutes ago

IF all participants always participate in all tournaments, then yes. But firstly, this is practically impossible, and secondly, in this case, you can use your first system - there will be no problem.

I have already come to this conclusion - it's just the end of the month, and I ran out of internet at work at the time when I was writing the post in response to yours. I mean, I proposed the Virtual component, this system eliminated one flaw, but introduced another, which I didn't notice right away. But if you add some kind of calculation of coefficients to this idea (this is for the smart guys), then maybe something would work out. But they are stuck in searching for a balance, although their systems are already flawed just because they take into account the player's place. Plus, there is a loophole in the protection of player ratings.

we assume that there is Tournament 4, where Player A and B do not participate, but a newbie participates and completes a super-easy map in 1 day and receives 10000 points. And for the first tournaments, where Player A and B completed a super-difficult map in 10000 days, they receive 1 point. The result is unsatisfactory.

I agree. The solution is to try not to miss tournaments, which is unrealistic. That's why I came to the conclusion that I won't change anything in my system for now. It already works well enough. The flaw of a strong player not being in the tournament is a small probability. But even if that happens, it's better for the newbie to get a hundred, just like the leader on another map, than to make him an order of magnitude higher :)

P.S. In your first system, the same rating was given for all tournaments, and in the second, a higher rating may be given for an easy tournament than for a difficult one. Draw your own conclusions.

I did. I didn't say that I would definitely use this system. I presented it as another idea, thinking that it might improve something. It's not a fact that it doesn't have further development. I'm just not a master at coming up with formulas with 9 levels :)