HeroesWorld

Posts from Alternative rating.

Tournament Committee
5.0 (1 rating)
Open
Do you need an alternative rating?
46 voters
Single choice
Sign in to participate in this poll.
Poll results will be visible after you vote or when the poll closes.
In reply to bRitvo4kA
User Avatar
Offline invisible
Registered:
Posts: 754
#46
Auto-translated
Stikkies +1

Britva, you're currently playing for a rating using a formula that SS-D came up with while hungover)))(just kidding). No international ratings have been conducted using this formula. Someone proposed their own system: if it doesn't work, then it won't be used. And everyone can be smart, including me)). I also don't know English, so what?)) Not everyone is so advanced :smile09:
Я ламер и играть не умею, нах!:smile33:
0
⚖️
0
In reply to Wicc
User Avatar
Offline Alex_R
Registered:
Posts: 1,739
#47
Auto-translated
Wicc
The only question is the number of these points.
I suggest that for tournaments, instead of giving a fixed number of points only to the prize-winning places, we simply introduce a coefficient for tournament games, which, by the way, will also not be constant, but will be formed based on objective reasons, for example, a prototype of a "pool/conflux."
The quantitative and qualitative composition of the players registered for the tournament will directly affect the size of the coefficient.
For example, if the average rating of all participants is more than 1600, then the coefficient is 1.5; if it is more than 1700, then it is 2.0 (the tournament is awarded the status of a master tournament), and so on. And if, for example, it is less than 1600, then the coefficient = 1.1 and (the tournament is awarded the status of a fan tournament).
The numbers are taken from the air (need to be discussed).
P.S. IMHO, the advantages:
+ this provides more incentive to attract players to tournaments
+ the reward is extended to all participants, not just the winners.
1
⚖️
0
In reply to Alex_R
User Avatar
Offline Дубровин Павел
Registered:
Posts: 1,832
#48
Auto-translated
Alex_R
I suggest that for tournaments, instead of awarding a fixed number of points only to the prize-winning places, we simply introduce a coefficient for tournament games, which, by the way, would also not be constant, but would be formed based on objective reasons, for example, a prototype of a "pool/conflux".
The magnitude of the coefficient will directly depend on the quantitative and qualitative composition of the players registered for the tournament.
For example, if the average rating of all participants is more than 1600, then the coefficient is 1.5; if it is more than 1700, then it is 2.0 (the tournament is awarded the status of a master-level tournament), and so on. And if, for example, it is less than 1600, then the coefficient = 1.1 and (the tournament is awarded the status of a fan-level tournament).
The numbers are taken from the air (need to be discussed).
P.S. IMHO, the advantages:
+ This provides more incentive to attract players to tournaments.
+ The reward is extended to all participants, not just the winners.


Yes, indeed, it's logical!
I'm in favor.
С истиной где-то рядом...





Лето 2010.
1
⚖️
0
In reply to invisible
User Avatar
Offline bRitvo4kA
Registered:
Posts: 1,127
#49
Auto-translated
invisible
Breezy, you're currently playing for ranking using a formula that SS-D came up with while hungover.
Then explain why you'd want to change something good for something worse? :smile38: If you're going to switch, it should be to something normal, proven, and worthy of attention and respect. :smile42:
В дикім полі
0
⚖️
0
User Avatar
Offline SpiriTT
Registered:
Posts: 876
#50
Auto-translated
Everything is great and wonderful, and if it's important, I, for example, am for it. One question: for example, after 2 months, it reaches the point where, for winning against me, the so-called "TOP" player receives 1 point. What's the point for them in playing against me then? Everyone is talking about this, that you can't increase your rating by playing against weak players. But what if this weak player wants to learn how to play, improve their skills, etc.? Won't it turn out that the TOP players will only play against TOP players, and the weak players will play against weak players? Then it becomes a kind of analogue of a neighboring development – the one about the top league (or about 2 leagues). Just some thoughts, I said what came to mind.
ну это вообще всем аргамездец!© Kash
без авто:smile20:
Человек - игрушка Бога. Этому - то и надо следовать. Надо жить играя.
Платон.
0
⚖️
0
In reply to SpiriTT
User Avatar
Offline Alex_R
Registered:
Posts: 1,739
#51
Auto-translated
SpiriTT
It has come to the point where, for winning against me, for example, the so-called "TOP" player will receive only 1 point. What's the point for them in playing against me then?
To reach such a point, you need to sink to the very bottom of the table, losing to everyone in a row, even not to the TOP players, and lose a couple of games to this particular TOP player.
In principle, it's a fair point, such a situation is quite real, although many play for rating, but without considering how many points they can gain from a particular opponent.
Of course, I'm sure that there will be different players and TOP players who will gladly play with any opponent for any number of points (even for 1), but you can also suggest a protective mechanism, for example, a player who is challenged to a duel must accept the challenge at least once.
Of course, you can always refuse, but for objective reasons.
P.S. I will say right away that such a restriction does not appeal to me personally, therefore I suggest compiling a list of players who will play with any opponent (regardless of the benefit in case of victory), and refuse only due to temporary and other technical incompatibility, or purely personal reasons.
Players who agree to this can be marked in the ranking table in some way, for example, by adding a column where you can put a checkmark "I play with everyone", or highlight them with a special font.
If the idea is suitable, I'm on the list!:)
0
⚖️
0
In reply to Alex_R
User Avatar
Offline Lider1
Registered:
Posts: 3,358
#52
Auto-translated
Alex_R
...a player who is challenged to a duel must accept the challenge at least once...

There is no need for any penalty system. This isn't a job. Besides, if a person receives 10 requests to play in a day, and is obligated to play with each one, there will be legal disputes, not games.
Сейчас балуюсь этим https://vcmi.eu/download/
За то что нейтралы не бегают за единичкой, для меня это уже стоит того
0
⚖️
0
User Avatar
Offline StickyFingaz
Registered:
Posts: 1,499
#53
Auto-translated
If you're going to switch to something, it should be something reliable, worthwhile, and deserving of attention and respect.
Dude, can you answer my questions about this rating? I'm just not getting it right now – what makes this rating better, and how do you manage it?
The most accurate rating can be obtained based on tournaments in which players of roughly equal strength compete.
Oops, inconsistency #1.
The Elo rating system is based on the assumption that the strength of each chess player can be represented as a probabilistic variable, following a normal distribution.
Well, I "got lucky" – I have a background in this, and I can understand something from these words. But there are other people here too. Can you explain it yourself? Because I suspect you don't really understand it...
The calculation of a specific player's rating based on the results of a tournament is based on comparing the number of points they scored with the expected number of points, predicted based on their rating.
How will you predict a player's likely position in a tournament, how many games they lost, how many they won, etc.? There will be so many arguments about this...
If, at the end of the tournament, the number of points scored is greater than the predicted value, then the rating of that player increases.
Another inconsistency... 95% of our games are not tournaments. What about them?
Everything is clearly explained in the links.
I just quoted the only Russian link, and you can't answer my questions. You just inserted something complicated, like you were trying to avoid answering, assuming that people will just swallow it. You could even find something in Chinese... No, answer for what you say =)

Well, about 20 people have joined. When vbn appears on the forum, I'll discuss the table with him. And to you, dear forum members, a question – what should we do with tournaments? In principle, awarding victory points with coefficients for a separate tournament is a good idea, but points should be deducted for a loss without coefficients =)

To get to that point, you need to sink to the bottom of the table, losing to everyone, even not the top players.
Such players can learn something even from players with a rating of 16++. Personally, for me, this was another incentive to improve my skills in order to play with even stronger opponents.
0
⚖️
0
User Avatar
Offline Lider1
Registered:
Posts: 3,358
#54
Auto-translated
Based on the Elo rating, it is calculated on the hero esports portal, and if someone really likes it...
Сейчас балуюсь этим https://vcmi.eu/download/
За то что нейтралы не бегают за единичкой, для меня это уже стоит того
0
⚖️
0
User Avatar
Offline bRitvo4kA
Registered:
Posts: 1,127
#55
Auto-translated
What’s so hard to understand? You don’t need to understand the formulas; I provided a link to a calculator that calculates using the ELO formulas. It will take the data from seasonal games and tournaments. The initial rating for each player is 1600. :smile02: Let StickyFingaz input the results and compile them into a rating table and publish it for everyone. In general, I’m in favor of :smile20: this rating system; it will more objectively determine skill. Plus, each player can check :smile30: their points, which will help avoid calculation errors :smile35:.
В дикім полі
0
⚖️
0
User Avatar
Offline Loi
Registered:
Posts: 1,084
#56
Auto-translated
the strength of each chess player can be represented as a probabilistic variable, following a normal distribution
Apparently, this means that a chess player is most likely to play at their expected level (Mean, so to speak), and the probability that they will play significantly above (or below) their level decreases rapidly towards zero. Something like that...
Стратегия - искусство возможного.

Первое правило форума :
"Хвалу и клевету приемли равнодушно
И не оспаривай глупца" (с)

Новичкам: ВНИМАТЕЛЬНО перечитайте ВЕСЬ первый пост по нижеприведенной ссылке и 90% ваших вопросов отпадут( на остальные 10% с удовольствием отвечу):
Правила онлайн игр
0
⚖️
0
User Avatar
Offline StickyFingaz
Registered:
Posts: 1,499
#57
Auto-translated
You don't need to understand the formulas, I just provided a link to a calculator.
Ah, so this isn't Elo anymore; it doesn't take all its features into account. Well, it's just more convenient, because you get many more points for winning against a weak opponent. You need to stop relying on easy wins...
Instead, give me some advice on how to handle games from tournaments.
1
⚖️
0
User Avatar
Offline bRitvo4kA
Registered:
Posts: 1,127
#58
Auto-translated
Treat tournament games the same way as seasonal ones, put them all together.
В дикім полі
0
⚖️
0
In reply to StickyFingaz
User Avatar
Offline Дубровин Павел
Registered:
Posts: 1,832
#59
Auto-translated
StickyFingaz

And to you, esteemed forum members, a question – what should be done with tournaments? In principle, awarding victory points with coefficients for a specific tournament is a good idea, but points should be deducted for losses without coefficients.

Yes, Slava, I agree with this point!
С истиной где-то рядом...





Лето 2010.
0
⚖️
0
User Avatar
Offline StickyFingaz
Registered:
Posts: 1,499
#60
Auto-translated
Well, opinions are divided. The more additional coefficients, the less objective the rating becomes (the idea of an alternative rating is to calculate pure skill). On the other hand, tournament victories should be rewarded. What should we do? 1) Games with coefficients should be included in the old rating; they won't be in the alternative one. 2) Conduct a vote. 3) ... I am ready to accept any of the options; it all depends on the community.
0
⚖️
0