Banned unfairly from MMR

[Beast]

Listen I get there's bans that are justified, but in my case I don't believe this to be the case. A match started, I let an mmr op know that "hey sorry man I got to go sub me please" I thought that by at least giving notice asap, that this wouldn't be an issue. I went afk shorlty after briefly letting Kengal know, I come back,later look up the message history, and dwopple decided that I needed a multi day ban for it. Like seriously dude what the fuck? Show your fellow twdt champion a little more respect. It's bullshit, and you know it.

[paradise!]

Wait.....

You did " !p ". Then when the game started you said you had to go and then went afk in the middle of the game? You think letting an MMR op know somehow makes it ok?

the man is absolutely delulu. ima pray for you big boy. pray i dont see you in the trenches. Im comin for that big head of yours bud. Don't ever post some dumb shit like this on my forums again. Honestly I'm pissed now. Way to fkn go. Don't come around here again. Idiot.

[Klean-X]

Listen, I was the first person to call Dwopple out on his bullshit trash staffing years ago. Does anyone listen? No.

These fucking tards on staff don't understand statistics so, in TW fashion I'll have chatGPT explain it to you fools here. Here is my dissertation on why trying to staff MMR thinking it's going to help create more accurate statistics is wrong. I'll also add that I've taken advanced M.S. level statistical courses.

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:

1. 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.
2. 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.
3. 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.
4. 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.

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.

[Klean-X]

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:

1. 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.
2. 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.
3. 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.
4. 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.
5. 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.

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.





4o

[Hookecho]

Not sure if you posted to the wrong thread or what but Beast wasn't talking about re-rolls in this thread. I happen to agree with you that the re-rolls thing seems unfair. If you want to bring that up, I would suggest a new thread. A thread where you're not trying to take a jab at staff and instead trying to collaborate constructively.

Beast, we have an entire appeals portal (https://trenchwars.org/support) to appeal bans which is where these kinds of things should be directed. If you have a complaint against a member of staff, sending an email to twcomplaints 'at' gmail.com is the easiest way to get in touch with the sysops.