Tweet
Discussion Regarding Fairness of Re-Rolling
Original Title Force Changed by Pussy Staff: MMR is Trash and Staff has been Doing Wrong From the Start - Recommended by Hookecho validated by working link and more evidence on how fucking sad this staff is; they can't even handle slight criticism without pissing themselves and trying to damage control for priitk or whoever the fuck is left in charge, which i think is no one.
TLDR: GPT agrees that staff is silly for moderating MMR. GPT says that introducing biased rerolling and game canceling creates more problems/bias in the statistics than it solves.
... staff doesn't understand statistics so, in TW fashion I'll have chatGPT explain it here.
PROMPT:
We have a game where 5 players vs 5 players. The bot randomizes the teams based on past performance, but occasionally someone will do extra worse or leave. I personally don't think there is any reason to moderate these games and try to cancel them when the moderators are inconsistent and often biased in their actions. Does statistical chance not typically assume that if someone will leave off one persons team eventually they will do it on another persons team? This would mean non-moderation should eventually show equal statistics over a large sample size.
ChatGPT: 4o
You're correct in suggesting that, over a large sample size, random events like players leaving should theoretically balance out. In the long run, the law of large numbers implies that each player will experience roughly equal conditions, assuming the randomization is fair and the sample size is large enough.
Here's a breakdown of the key points:
In conclusion, if the teams are genuinely randomized and the sample size of games is large enough, non-moderation should result in equal statistics over time. The key is ensuring that the randomization process is unbiased and that there is a sufficiently large number of games to allow the law of large numbers to take effect.
PROMPT 2:
Furthermore, does allowing staff to at the whim of the best player re-roll the randomization process to create new teams add a factor of unfairness? Or would this be okay if you wanted to continue to have a very fair game?
ChatGPT
Allowing staff or the best player to re-roll the randomization process introduces a significant factor of potential unfairness. Here are several reasons why this practice could disrupt the fairness of the game:
To maintain a fair game, it is crucial that the randomization process remains unbiased and consistent. Allowing re-rolls at the discretion of staff or the best player compromises these principles and introduces elements of unfairness. If fairness is a priority, the randomization process should be respected without intervention, except in clearly defined and agreed-upon exceptional circumstances.
TLDR: GPT agrees that staff is silly for moderating MMR. GPT says that introducing biased rerolling and game canceling creates more problems/bias in the statistics than it solves.
... staff doesn't understand statistics so, in TW fashion I'll have chatGPT explain it here.
PROMPT:
We have a game where 5 players vs 5 players. The bot randomizes the teams based on past performance, but occasionally someone will do extra worse or leave. I personally don't think there is any reason to moderate these games and try to cancel them when the moderators are inconsistent and often biased in their actions. Does statistical chance not typically assume that if someone will leave off one persons team eventually they will do it on another persons team? This would mean non-moderation should eventually show equal statistics over a large sample size.
ChatGPT: 4o
You're correct in suggesting that, over a large sample size, random events like players leaving should theoretically balance out. In the long run, the law of large numbers implies that each player will experience roughly equal conditions, assuming the randomization is fair and the sample size is large enough.
Here's a breakdown of the key points:
In conclusion, if the teams are genuinely randomized and the sample size of games is large enough, non-moderation should result in equal statistics over time. The key is ensuring that the randomization process is unbiased and that there is a sufficiently large number of games to allow the law of large numbers to take effect.
PROMPT 2:
Furthermore, does allowing staff to at the whim of the best player re-roll the randomization process to create new teams add a factor of unfairness? Or would this be okay if you wanted to continue to have a very fair game?
ChatGPT
Allowing staff or the best player to re-roll the randomization process introduces a significant factor of potential unfairness. Here are several reasons why this practice could disrupt the fairness of the game:
To maintain a fair game, it is crucial that the randomization process remains unbiased and consistent. Allowing re-rolls at the discretion of staff or the best player compromises these principles and introduces elements of unfairness. If fairness is a priority, the randomization process should be respected without intervention, except in clearly defined and agreed-upon exceptional circumstances.
- Randomization and Fairness: If the teams are truly randomized based on past performance, each player has an equal chance of being affected by a teammate leaving. This should balance out over many games.
- Law of Large Numbers: Over a large number of games, statistical fluctuations (like a player leaving) should average out. This means that any disadvantage caused by a player leaving should be equally distributed among all players over time.
- Moderation and Bias: If moderation is inconsistent and biased, it could introduce more unfairness than it prevents. Inconsistent moderation could skew the balance rather than maintaining it.
- Sample Size: The larger the sample size, the more likely it is that these random events will even out. In the short term, there might be perceived unfairness, but over a long period, the distribution should become more uniform.
- Bias and Influence: If re-rolling is done at the discretion of the best player or staff, it introduces personal bias. The decision to re-roll could be influenced by personal preferences, friendships, or perceived advantages, rather than maintaining an objective, fair process.
- Consistency and Predictability: For a system to be fair, it must be consistent. Randomization that can be overridden introduces inconsistency. Players cannot predict or trust that the process will be fair, which can undermine the integrity of the game.
- Impact on Team Dynamics: Frequent re-rolling can disrupt team dynamics. Players may feel less invested in their teams if they know that teams can be changed at any moment, leading to less cohesive and collaborative gameplay.
- Psychological Effect: Knowing that re-rolling is an option might affect how players approach the game. They might rely on the possibility of a re-roll rather than striving to improve or adapt to their current team composition.
- Long-term Balance: Over the long term, the law of large numbers will balance out random events, but this balance is disrupted if re-rolling introduces non-random changes. This can skew the statistics and fairness of the game.
Comment