Posts from Calculating the rating for clans!

Greetings, gentlemen,
Since Tournaments are now actively being held on our portal, and various Clans also exist on the portal, I propose introducing a so-called Clan Rating.
Consequently, it needs to be calculated according to certain rules. I suggest that we discuss and outline these rules in this thread.

Let's start with my proposal:
I propose introducing three ratings for each Clan:

1) Cumulative Rating:

AllRating_new = AllRating_old + (P1 + P2 +...Pn)
where: AllRating_new - the rating after the Tournament; AllRating_old - the rating before the Tournament; P1, P2,...Pn - the rating of the players participating in the Tournament.
In other words, the larger the Clan and the more often it participates in Tournaments, the higher its rating.

2) Average Rating:

MeanRating = AllRating_new/N
where: MeanRating - the average rating; N - the number of Clan members nominated in this version of the game (thank you, AlexeyPank).
In other words, having "inactive members" in the Clan is not encouraged.

3) Clan Activity:

ActivityRating = (ActivityTournament * 100) / AllTournament
Where ActivityRating - Clan activity (measured in percent); ActivityTournament - the number of Tournaments in which the Clan participated; AllTournament - the number of Tournaments held.
It's quite simple here...

I would be happy to hear your opinion on this system and/or see your system of rules.

Sincerely.

Unfortunately, I don’t remember who exactly, but someone on the portal already expressed an opinion about this, and I absolutely agree... The clan rating should be calculated based on several strong players... for example, the top 5... This will show the real strength of the clan (because it is these players who participate in tournaments). This way, the clan won't suffer from "dead souls"... And it will give an opportunity for newcomers to play without worrying about the rating... A cumulative rating is not a good option at all... strong players in a small clan will have a lower rating than those who play much worse but are in a huge number... it's just not right.

In principle, the idea is correct. However, not all clans play all versions of the game, so how should we calculate this? And counting all players, even if they don't play this particular version, doesn't seem right. It would be better to simply sum up the rating of the clan members. IMHO.

Unfortunately, I don't remember exactly who said it, but someone on the portal already expressed an opinion about this, and I absolutely agree with them... The clan rating should be calculated based on several strong players... for example, 5 of them. This will show the real strength of the clan (because these are the players who participate in Tournaments). This way, the clan won't suffer from "dead souls"... And it will give beginners a chance to play without worrying about the rating...

As far as I understand, you mean this. It’s a good system. But I don't like it precisely because there can be "dead souls" in the clan. In my version, this is not encouraged...
However, Natasha's proposed system can be considered an alternative. Moreover, not only to the clan rating, but also to the "tournament" one, along with the "current" one.

And a cumulative one is definitely not an option... Strong players in a small clan will have a lower rating than those playing much worse, but in large numbers.... something isn't right about that....

Yes, but then their average rating will be VERY high!

In principle, the idea is correct. Only here, not all clans play all versions of Heroes, so how should we calculate? And calculating based on all players if they basically don't play this version is not quite right. It is better to simply sum up the ratings of the clan members. IMHO.

You are absolutely right! Then we need to make a correction to the formula for the average rating... N - the number of clan members nominated in this version of Heroes. I’ll fix it in the first post. Thank you.

P.S. I would really like to know what VDV_forever, IronAxe, AmberSoler and all clan leaders think about this...

I agree with CrasH; we can consider the clan rating as the sum of the ratings of the best players in the clan. I think 5 people is the best number to provide an objective assessment of the clan, especially since most clans on the resource have 5 or more members... (not counting the "Soul Catchers" clan, which has only 3 members left after Alexei left). Another question is whether to create a clan rating for all versions. If so, what should be summed up? Should it be the players' ratings taken from the overall rating across all versions, or a single rating based on one version (presumably the best rating for a player)?

I agree with CrasH, we can consider the rating as the sum of the ratings of the best players in the clan. I think 5 people is the best number that can give an objective assessment of the clan.

I agree with the statement that we can sum up the rating of a limited number of players in the clan. But I disagree with the number "5" because it was clearly taken out of thin air... If we are going to limit the number, I suggest the number "4" because that is the minimum number of people needed in a clan to officially register it on HW. Perhaps we should conduct a poll on this matter...

...another question is whether to create a clan rating for all versions? If so, what should be summed up? The ratings of players taken from the overall rating in all versions? Or one rating for a specific version (probably the best one for the player)?

I believe that it is not necessary to create a general rating for all versions! because these are different games... the essence of the game is the same! But the games are different! Therefore, the rating should be separate for each version.

Sincerely.

I agree with the statement that it is possible to summarize the rating of a limited number of players in a clan. But I disagree with the number "5," as it was clearly pulled out of thin air... If we are going to limit the number, I suggest the number "4," because that is the minimum number of people needed in a clan for it to be officially registered on HW. Perhaps we should conduct a poll on this matter...

Sincerely.

Yes... the number 5 was indeed pulled out of thin air... It was just said as an example....

I agree with the statement that it is possible to sum the Player Rating of a limited number of players in a clan. But I do not agree with the number "5", as it was clearly pulled out of thin air... If we are to limit the number, I suggest the number "4", because that is exactly how many people must be in a clan for it to be officially registered on HW. We should probably hold a poll regarding this...


Respectfully.

Just before writing this post, I looked through the representative threads of all clans, and everyone has more than 5 members..... well if you want to count by four, then count by four, it's not critical...... the mechanism is what matters.....
And another snag: let's assume we sum the results of 4 players to determine the clan rating, for example in 5.... and I am not sure that every clan will have 4 people playing at a tournament level in 5; the same situation applies to other versions.

In other words, it is necessary to decide: should the clan rating be considered as a single entity (in which case all versions are taken into account) or as the sum of the players' ratings (in which case it is calculated separately for each version)?
And in addition, the formula for calculating individual ratings is also a controversial issue.

In principle, it is necessary to decide: should the clan rating be considered as a single entity (in which case all versions are taken into account) or as the sum of the players' ratings (in which case it is calculated separately for each version)?
And besides, the formula for calculating individual ratings is also a controversial point.

I repeat: "I believe that it is not necessary to create a General rating for all versions!" That is, for each version of the game, there will be a separate "cumulative" and "average" rating...
The individual rating (i.e., the rating for a tournament) is calculated using the IronAxe or AmberSoler system.

And again, there's a snag: let's say we sum up the results of 4 players to find out the clan rating, for example, in the 5th game.... and I'm not sure that all clans will have 4 people playing at a tournament level in the 5th game; the same situation applies to other versions.

That is exactly why I didn't want to introduce any restrictions...
Firstly: there is no problem that you mentioned.
Secondly: all members of the clan contribute to the cumulative rating.
And finally: the condition for the average rating does not encourage the presence of "dead souls".

Sincerely.

Good day to you all, ladies and gentlemen. After the preliminary calculation of clan ratings, I decided that we need to introduce a time limit for accumulating clan ratings, i.e., specify the rating for a certain period, such as a year. Why is this necessary? Let me explain: Let's say we have a clan called "Winners," and this clan has been leading for a long time and has significantly outpaced other clans in terms of cumulative and average rating. Then, at some point, this clan "dies," meaning its members stop participating in tournaments on our portal. However, due to their huge accumulated rating, they can remain in the top positions of the "hit parade" for a long time. To avoid this, I propose introducing a time limit, i.e., the cumulative rating starts being calculated from January 1st and ends on December 31st. And at the end of the year, we can summarize the results, like: "Best Clan of 20xx," "Best Player of 20xx," "Most Active Clan of 20xx," etc. I look forward to your comments. Sincerely.

I'm waiting for your comments.

What's there to comment on? :) You started quoting "dialogues between Iron and Amber"... :D

All these questions were discussed somewhere on the second page of the conversation about the rating calculation method... (now there are more than ten pages of text there). Moreover, everything is explained there much more thoroughly, with examples, options, and conclusions... Sometimes the rating is cumulative (it ages, it is not reset), sometimes it is cyclical (it does not age, it is reset after a period). The number of participants in the rating is the same for all Clans. This is mandatory; otherwise, we will see not the strength of the Clan, but its size. Who needs this information?

If you have a goal to repeat the discussion of the rating again, but only within the framework of Clan calculations, then I'm out... :) There is nothing new here anymore, and all the issues you encounter and will encounter have already been calculated, collected, and broken down... :) Aren't you wasting your time? Or are you reading slower than you write? :D

I am strongly against having a fixed number of participants in the ranking, how are you going to compare clans with 18 members and 4 members???????????
For example, let's take the situation with the 2v2 mode. In a clan with 18 members, there will most likely be 4 people who play 2v2 at an average level, while in a clan with 4 members, there is only 1 player in this version, but he is strong and plays at a higher level... so, should a clan with a small number of players not be ranked in the 2v2 mode just because they don't have enough participants in this version????????
Then such a ranking will be very subjective and worthless if a champion cannot bring a good coefficient to the clan because he is the only

The point is that the main idea of our "stone-crushing" competition with Iron, which spans 10 pages, is this: there is no single rating that takes all factors into account. There can't be, and there won't be. The participants in the game (the "rating" game, as an option) must agree among themselves which criteria will be significant and which will not. And calculate based on their agreements. Agree that height is very important for a basketball player, as it gives an advantage in the game. Therefore, hypothetically, you can calculate the rating of basketball teams based on the average height of the players on the team. The "taller" the team, the lower the coefficient they receive for a win (because it was easy for them to win)... :)

I'm exaggerating, although there is some reason in this. But the basketball association does not accept such amendments - here, height doesn't matter anyway - the main thing is to throw the ball into the basket more often. And let the team take care of selecting players. But the main thing is: there are always an equal number of participants on the court! Do you see the analogy? Strength and height are as they are, but the number is strictly defined.

That's how it should be with us. No one cares how many members of a Clan play in which version. The rating is only for those Clans that have a qualifying minimum number of players needed for the calculation (this is something they can agree on - you can not even set a minimum, if one player participates, the Clan rating is calculated based on him). But the maximum must be strictly limited... For example, three participants... Otherwise, there will be a huge imbalance... :)

Who needs a rating that is determined by the number of players in a Clan? Will the awareness of your gaming strength warm you up just because there are many of you? :) It won't warm me up, for example. By the way, I anticipate possible comments about the average level of all players in the Clan - this is also not a solution, because Clans will start getting rid of (or not accepting, as an option) potentially weak players. This will not lead to anything good. But a qualifying trio will be an incentive for any player in the Clan and will eliminate many possible complications in calculations and relationships. This is just a suggestion.

I am categorically against a fixed number of participants in the rating, how are you going to compare Clans with 18 people and 4?????

That's what I'm talking about. Based on this comment, there is nothing to compare at all. Where there are more players, they are stronger. Just list the Clans by number, starting with the largest in terms of membership - and there you have a ready-made rating... Why complicate things with calculations? That's why I suggest setting a limit. The three strongest - count. The rest - on the bench. The more players, the more chances to get a good qualifying trio in each tournament. But - without a guarantee that this will allow small Clans to compete on equal terms. Otherwise, is this even a rating?

to AmberSoler: we need to think about this... And why exactly three? Why not four or five? Or maybe two?
The advantages of this system are obvious (and very good), but there are also disadvantages... By the way, you say:

Rating - only for those clans that have a qualifying minimum number of players required for the calculation

I don't agree with this; it should be: "Rating - only for those clans that have a qualifying number of players required for the calculation," i.e., a specific number of players for all clans.