#100998: "The selection of gurus by the number of games played generates a representativeness bias"
Que s'est-il passé ? Merci de choisir ci-dessous
Que s'est-il passé ? Merci de choisir ci-dessous
Veuillez vérifier s'il existe déjà un rapport sur le même sujet
Si oui, veuillez VOTER pour ce rapport. Les rapports ayant le plus de votes auront la PRIORITÉ !
# | Status | Votes | Game | Type | Title | Last update |
---|
Description détaillée
• Merci de copier-coller le message d'erreur qui s'affiche à l'écran, s'il y a lieu.
-• Veuillez expliquer ce que vous vouliez faire, ce que vous avez fait et ce qu'il s'est passé
These are Gurus who select the game configuration for arena seasons - Great idea !
Gurus are selected because they play one specific game a lot on BGA plateform - of course !
When a game offers different configurations for the number of players, games with few players are faster and therefore more numerous than those with many players. - Obvious !
Also the probability of becoming a Guru is greater if a player prefer to play short games than long games, because the selection is based on the number of games completed during a period.
Finally, over seasons, the games configurations (chosen by Gurus) will be shorter and shorter, with fewer players, even if the games concerned are richer in configuration with more players.
I think this can be considered as a problem/bug, because some good players (who like big setups) might lose interest in seasons, as we can see in some comments.
May be a more representative selection of Gurus should be established
Proposal : In a game with possible configurations for 2, 3 or 4 players, select 1/3 of the Gurus on their greatest number of games completed at 2 players, 1/3 on the greatest number of games completed at 3 players and 1/3 on the greatest number of games completed at 4 (taking into account duplicate players).
Thanks for reading me.
• Quel est votre navigateur ?
Google Chrome v117
• SVP copiez / collez ici le texte affiché en anglais au lieu de votre langue. Si vous avez une capture d'écran de ce bug (pensez-y !), vous pouvez utiliser Imgur.com pour la télécharger et copier-coller le lien ici.
-• Ce texte est-il disponible dans le système de traduction ? Si oui, a-t-il été traduit depuis plus de 24 heures ?
These are Gurus who select the game configuration for arena seasons - Great idea !
Gurus are selected because they play one specific game a lot on BGA plateform - of course !
When a game offers different configurations for the number of players, games with few players are faster and therefore more numerous than those with many players. - Obvious !
Also the probability of becoming a Guru is greater if a player prefer to play short games than long games, because the selection is based on the number of games completed during a period.
Finally, over seasons, the games configurations (chosen by Gurus) will be shorter and shorter, with fewer players, even if the games concerned are richer in configuration with more players.
I think this can be considered as a problem/bug, because some good players (who like big setups) might lose interest in seasons, as we can see in some comments.
May be a more representative selection of Gurus should be established
Proposal : In a game with possible configurations for 2, 3 or 4 players, select 1/3 of the Gurus on their greatest number of games completed at 2 players, 1/3 on the greatest number of games completed at 3 players and 1/3 on the greatest number of games completed at 4 (taking into account duplicate players).
Thanks for reading me.
• Quel est votre navigateur ?
Google Chrome v117
• Merci d'expliquer votre suggestion de manière précise et concise, de façon à ce qu'il soit aussi simple que possible de comprendre ce que vous voulez dire.
-• Quel est votre navigateur ?
Google Chrome v117
• Qu'était-il affiché à l'écran lorsque vous avez été bloqué·e (Écran vierge ? Une partie de l'interface de jeu ? Un message d'erreur ?)
-• Quel est votre navigateur ?
Google Chrome v117
• Quelle partie des règles n'a pas été respectée par l'adaptation BGA
-• La violation de règle est-elle visible dans le replay de la partie ? Si oui, à quel numéro de coup ?
These are Gurus who select the game configuration for arena seasons - Great idea !
Gurus are selected because they play one specific game a lot on BGA plateform - of course !
When a game offers different configurations for the number of players, games with few players are faster and therefore more numerous than those with many players. - Obvious !
Also the probability of becoming a Guru is greater if a player prefer to play short games than long games, because the selection is based on the number of games completed during a period.
Finally, over seasons, the games configurations (chosen by Gurus) will be shorter and shorter, with fewer players, even if the games concerned are richer in configuration with more players.
I think this can be considered as a problem/bug, because some good players (who like big setups) might lose interest in seasons, as we can see in some comments.
May be a more representative selection of Gurus should be established
Proposal : In a game with possible configurations for 2, 3 or 4 players, select 1/3 of the Gurus on their greatest number of games completed at 2 players, 1/3 on the greatest number of games completed at 3 players and 1/3 on the greatest number of games completed at 4 (taking into account duplicate players).
Thanks for reading me.
• Quel est votre navigateur ?
Google Chrome v117
• Quelle action de jeu vouliez-vous faire ?
-• Qu'essayez-vous de faire pour déclencher cette action de jeu ?
These are Gurus who select the game configuration for arena seasons - Great idea !
Gurus are selected because they play one specific game a lot on BGA plateform - of course !
When a game offers different configurations for the number of players, games with few players are faster and therefore more numerous than those with many players. - Obvious !
Also the probability of becoming a Guru is greater if a player prefer to play short games than long games, because the selection is based on the number of games completed during a period.
Finally, over seasons, the games configurations (chosen by Gurus) will be shorter and shorter, with fewer players, even if the games concerned are richer in configuration with more players.
I think this can be considered as a problem/bug, because some good players (who like big setups) might lose interest in seasons, as we can see in some comments.
May be a more representative selection of Gurus should be established
Proposal : In a game with possible configurations for 2, 3 or 4 players, select 1/3 of the Gurus on their greatest number of games completed at 2 players, 1/3 on the greatest number of games completed at 3 players and 1/3 on the greatest number of games completed at 4 (taking into account duplicate players).
Thanks for reading me.
• Que se passe-t-il lorsque vous essayez de faire cela (message d'erreur, message de la barre d'état du jeu, ...) ?
• Quel est votre navigateur ?
Google Chrome v117
• À quelle étape de la partie le problème est-il apparu (quel était le texte d'instruction du jeu) ?
-• Que se passe-t-il lorsque vous essayez de faire une action de jeu (message d'erreur, message de la barre d'état du jeu, ...) ?
These are Gurus who select the game configuration for arena seasons - Great idea !
Gurus are selected because they play one specific game a lot on BGA plateform - of course !
When a game offers different configurations for the number of players, games with few players are faster and therefore more numerous than those with many players. - Obvious !
Also the probability of becoming a Guru is greater if a player prefer to play short games than long games, because the selection is based on the number of games completed during a period.
Finally, over seasons, the games configurations (chosen by Gurus) will be shorter and shorter, with fewer players, even if the games concerned are richer in configuration with more players.
I think this can be considered as a problem/bug, because some good players (who like big setups) might lose interest in seasons, as we can see in some comments.
May be a more representative selection of Gurus should be established
Proposal : In a game with possible configurations for 2, 3 or 4 players, select 1/3 of the Gurus on their greatest number of games completed at 2 players, 1/3 on the greatest number of games completed at 3 players and 1/3 on the greatest number of games completed at 4 (taking into account duplicate players).
Thanks for reading me.
• Quel est votre navigateur ?
Google Chrome v117
• Veuillez décrire le problème d'affichage. Si vous avez une capture d'écran de ce bug (pensez-y !), vous pouvez utiliser Imgur.com pour la télécharger et copier-coller le lien ici.
-• Quel est votre navigateur ?
Google Chrome v117
• SVP copiez / collez ici le texte affiché en anglais au lieu de votre langue. Si vous avez une capture d'écran de ce bug (pensez-y !), vous pouvez utiliser Imgur.com pour la télécharger et copier-coller le lien ici.
-• Ce texte est-il disponible dans le système de traduction ? Si oui, a-t-il été traduit depuis plus de 24 heures ?
These are Gurus who select the game configuration for arena seasons - Great idea !
Gurus are selected because they play one specific game a lot on BGA plateform - of course !
When a game offers different configurations for the number of players, games with few players are faster and therefore more numerous than those with many players. - Obvious !
Also the probability of becoming a Guru is greater if a player prefer to play short games than long games, because the selection is based on the number of games completed during a period.
Finally, over seasons, the games configurations (chosen by Gurus) will be shorter and shorter, with fewer players, even if the games concerned are richer in configuration with more players.
I think this can be considered as a problem/bug, because some good players (who like big setups) might lose interest in seasons, as we can see in some comments.
May be a more representative selection of Gurus should be established
Proposal : In a game with possible configurations for 2, 3 or 4 players, select 1/3 of the Gurus on their greatest number of games completed at 2 players, 1/3 on the greatest number of games completed at 3 players and 1/3 on the greatest number of games completed at 4 (taking into account duplicate players).
Thanks for reading me.
• Quel est votre navigateur ?
Google Chrome v117
• Merci d'expliquer votre suggestion de manière précise et concise, de façon à ce qu'il soit aussi simple que possible de comprendre ce que vous voulez dire.
-• Quel est votre navigateur ?
Google Chrome v117
Historique des rapports de bug
You seem to misunderstand how gurus work.
You can be a guru without playing in the arena so people playing the game in various configurations CAN be represented.
But it is true that many experienced players only play or mostly play in Arena mode and they might pass for a season if they don't like the configuration which creates an unwelcome form of inertia.
Perhaps games with more players should count for a bit more as they take more time ?
I don't think that forcing other players counts to be represented would be a good idea.
-------------------
You aren't more likely to become a guru in a short game because the number of games required to play to become a guru depends on how many times the other players play the game.
I have once been a guru for a game (Expedition Northwest) for playing a single game in that season and I didn't even have 100 ELO.
----------------
It is also true that making gurus only those who play the game a lot creates a bias.
I think that there should be a certain randomness to it to create representativity :
- on top of the players selected for having played the most, some players should be selected randomly at some specific time before the end of the season. There would be a minimum requirement like having finished 2 games within the season, excluding "friendly" games.
All in all, I don't think it is a very important issue but it is an issue.
Ajouter à ce rapport
- Un autre n° de table / n° de coup
- La touche F5 a-t-elle résolu le problème ?
- Le problème est-il apparu plusieurs fois ? À chaque fois ? Aléatoirement ?
- Si vous avez une capture d'écran de ce bug (pensez-y !), vous pouvez utiliser Imgur.com pour la télécharger et copier-coller le lien ici.