Private Topics Discussion (Now with Mish Mash)

[Cheery Dog]

When I scroll down the View Active Topics page, I saw Neighborhood Kilo (http://forum.mafiascum.net/viewtopic.ph ... 3#p5838713) on the list.
I was wondering why it is included as an active topic (which excluding marathon weekends, are just non-game threads)

[Mr. Flay]

It's because Completed PTs weren't excluded from Active Topics until just now.

[RichardGHP]

Also, I think "manipulation" is an overly strong word. It is certainly not appropriate to describe what I do. I often reroll setups a few times anyway to get what I view as the best possible distribution. Again, this is not based on mathematical reasoning, just my thought process. For the same reason you might shake a pair dice for longer than usual or shuffle a deck of cards more thoroughly than usual, if you want to make sure that the result is "more random" (don't yell at me).

Just jumping in on this:
What you are describing is absolutely manipulating otherwise random results. It doesn't matter how much you shake the dice, the chance of any given outcome never changes. What you're saying is "oh I got three 2's, not random enough try again". Your dice and card examples are to assure randomness before getting a result. The modding practice you're describing is changing things

after

a result.

I already told you I know randomness is assured from the outset. In my mind, however, it is possible to have different levels of randomness. It is how I work. Don't like it, don't join my games. Also, you're misquoting my example, but since I am phoneposting I will let that slide.

[SleepyKrew]

I already told you I know randomness is assured from the outset. In my mind, however, it is possible to have different levels of randomness.

Altering a random result because it doesn't look random enough actually reduces the randomness.

It is how I work. Don't like it, don't join my games.

It's shitty practice. Don't do it, and then you won't have to worry about your players trying to outguess you on whether this game is the game you went against your convention to screw with people.

Also, you're misquoting my example, but since I am phoneposting I will let that slide.

When you get the chance, please tell me what I misquoted.

[Tierce]

In my mind

This fragment right here--this is not how randomness works.

Furthermore, you're excluding random options because they... don't look random? That doesn't make any sense and you're reducing the possibilities of a setup the players would be looking for. As SleepyKrew said, it's a poor practice and you're trying to shed blame from it, which just looks silly when you're the one doing it.

[RichardGHP]

Actually, I no longer care enough to keep debating this. I have defended my position as best I can and it's not going to change anytime soon, so the way I see it, we're just going around in circles.

Edit: In case I phrased it poorly earlier, my view is that while all random distributions are mathematically equal, some outcomes are more desirable than others. I get that the "done thing" is to stick with the first roll you get until the end, but I am not afraid to reroll if I get what I view as an undesirable outcome, and I stand by that absolutely.

[SleepyKrew]

some outcomes are more desirable than others.

I am not afraid to reroll if I get what I view as an undesirable outcome

kajdkdjskdjfnfkdowkdmdkxnsoxnsksjsndkxnajxoxnskxkd
I think you just gave my stats teacher a hernia

I want to keep telling you why you are wrong and it is bad but you have made it clear that you have no intention to listen and I don't think anyone else will be taking up the mantle for your position so I'll just go back to


Just
One final thing
You're mistaking arbitrariness for randomness

[Jake from State Farm]

but I am not afraid to reroll if I get what I view as an undesirable outcome,

what would be an example of an undesirable outcome?

[Faraday]

Rolling dice to prove you randomised seems a good idea, not like it puts much additional effort on the mod + being able to show the players is something I've always wanted to do but was too lazy to do zoraster's method.

[mykonian]

Actually, I no longer care enough to keep debating this. I have defended my position as best I can and it's not going to change anytime soon, so the way I see it, we're just going around in circles.

Going in circles? It's more all of MD telling you that what you are doing is wrong in every way you look at it. At this point you had two options: fix your shit or run for mommy.

[zoraster]

come on, that's not necessary. At the very least, at least RGHP is being honest, which is instructive.

[Jake from State Farm]

in the past on another site I have done similar things to richard, an important role goes to someone who is a VI or something you want to re-roll it.

Right or wrong it is his right to set his games up however he wants so like he said if you don't like the way he does it just don't join his games. I wouldn't be surprised if more people actually did this but most wouldn't have the courage to speak up.

[mykonian]

come on, that's not necessary. At the very least, at least RGHP is being honest, which is instructive.

that's just one step.

[RichardGHP]

Just to be clear, any rerolling is done

BEFORE

signups. I would never change the assignments after that, so it's not like I'm playing favourites, if Jake from State Farm gave that impression.

[zoraster]

in the past on another site I have done similar things to richard, an important role goes to someone who is a VI or something you want to re-roll it.

Right or wrong it is his right to set his games up however he wants so like he said if you don't like the way he does it just don't join his games. I wouldn't be surprised if more people actually did this but most wouldn't have the courage to speak up.

This is my point. I think Richard is just an example of someone who is willing to admit to it, whether he's misguided or not. I suspect he is just one of many. Which is why I don't think it's wrong for me to say I don't trust mods to secretly randomly generate without manipulating.

That being the case, there are gains by being able to verify (or at least partially verify) random generation. As a mod, I like that. As a player, I like that.

---

All that said, this is just one very tiny benefit of private topics on the whole. There are certainly bigger ones, and I don't want that to get away from us.

[zoraster]

ANNOUNCEMENT

The following will be added to the Private Topic policy:

Moderators of Mish Mash games who would benefit from the use of Private Topics should PM Xalxe for permission to create them. Permission will be granted on a case-by-case basis, depending on their intended function. Generally, they should only be used for topics which require multiple users to have access, not one-on-one discussions.

[Mr. Flay]

This is my point. I think Richard is just an example of someone who is willing to admit to it, whether he's misguided or not. I suspect he is just one of many. Which is why I don't think it's wrong for me to say I don't trust mods to secretly randomly generate without manipulating.

I'm legitimately curious what you think you gain by a permanent paranoia about the randomness of role generation.

Even if a mod fudges 15% of the time, it's not anything you can rely on or make decisions based on. Newbie Games showed a general and systematic bias against some setups, and we fixed that by having the LM roll for them. But for an individual mod, in an individual game? The fact that Richard MIGHT reroll tells you absolutely nothing about the setup. The fact that I might reroll tells you nothing. The setup is already so complex that the small chance of a non-random factor literally vanishes unless they are blatantly assigning PRs to their friends. In which case, would you really play that mod's game to begin with? It's not like you can 'verify' that they have randomly assigned the roles BEFORE that particular game is over, so what are you really gaining by all the hoops?

[zoraster]

This is my point. I think Richard is just an example of someone who is willing to admit to it, whether he's misguided or not. I suspect he is just one of many. Which is why I don't think it's wrong for me to say I don't trust mods to secretly randomly generate without manipulating.

I'm legitimately curious what you think you gain by a permanent paranoia about the randomness of role generation.

Even if a mod fudges 15% of the time, it's not anything you can rely on or make decisions based on. Newbie Games showed a general and systematic bias against some setups, and we fixed that by having the LM roll for them. But for an individual mod, in an individual game? The fact that Richard MIGHT reroll tells you absolutely nothing about the setup. The fact that I might reroll tells you nothing. The setup is already so complex that the small chance of a non-random factor literally vanishes unless they are blatantly assigning PRs to their friends. In which case, would you really play that mod's game to begin with? It's not like you can 'verify' that they have randomly assigned the roles BEFORE that particular game is over, so what are you really gaining by all the hoops?

And I think that's a legitimate question that has a couple different answers, both of which are true.

1. Does there have to be anything "gained" by permanent paranoia? Particularly so if it is justified paranoia? I don't base my beliefs on what I'd like to believe is true.

2. That said, the reason this came up was when several people balked at some of us saying that using dice rolls in a private topic might be a good way to have verifiable randomness. If all mods everywhere always perfectly randomize their roles and never manipulate and every player believes this to be the case, this would be unnnecessary. But because I believe mods absolutely do manipulate, and I think other players share my belief, it works in my favor to provide new and better ways of proving to players that I am doing everything I can to give them a random setup.

You've made a lot of assumptions about what players might and might not take into account. I do not make the same assumptions.

As for what's gained even though they don't verify until after: I tell them beforehand that things are randomized using this method. Unless they believe I'm lying about that in a way that will get me in trouble when the game is over, they now have assurance I did not manipulate the roles. Regardless, the vast majority of players in my games, particularly my large games, are repeat customers. I tend to run multiple games a year.

That built trust is important to me, even if it is not to others.

You are free to randomize (or not randomize) your roles as you see fit. I have not suggested nor do I plan to suggest we mandate anything here. But if you believe there is an implicit criticism of those who do not strive to prove their randomness when such a thing is readily available, your belief is correct.

[Mr. Flay]

You've made a lot of assumptions about what players might and might not take into account. I do not make the same assumptions.

I believe the burden of proof is on you to show that imperfect randomization breaks games. That's why I asked the way I did - not to prove that it happened or didn't happen, but that it can have an impact on the game,

during the game

. There's too much noise and incomplete information to derive anything about the role X has in Y's game.

[zoraster]

Why in the world would that be? That is a patently absurd statement to make, and there is absolutely no reason for me to need to do that. It's incumbent upon you to show that imperfect randomization is not a problem if you want to justify it despite the fact that randomization is a pretty crucially held concept in mafia games.

[Zachrulez]

Zoraster, my problem isn't the aspect of verification per say. I'm not hugely for it or against it. It's that I find dice rolls an inferior way of actually randomizing alignment and find it an overall worse method than simply using my current method. If I was able to use my current method in a verifiable way, I would as it's not really a big deal to me one way or another.

That is to say I don't really understand why it's important to verify a mod's honesty in distributing the roles, but I don't really have a problem with it outside of the fact that a push toward it might force or pressure me to use what I feel is an inferior randomization method.

[zoraster]

Well it's a little bit more work for the mod to put together, but is it inferior outside of that?

[Zachrulez]

That it's more efficient to put together is why I prefer my current method. How much value we put in the ability to verify that it was put together honestly is where we part company.

In fact I've as much as said that I would be happy to use my current method in a verifiable way if it was possible, but I don't find the advantage of verification alone a big enough one to use dice tags.

[Majiffy]

This looks like fun, I'm glad it's finally ready.

[Majiffy]

in the past on another site I have done similar things to richard, an important role goes to someone who is a VI or something you want to re-roll it.

Right or wrong it is his right to set his games up however he wants so like he said if you don't like the way he does it just don't join his games. I wouldn't be surprised if more people actually did this but most wouldn't have the courage to speak up.

I am aware of this occurring with certain mods off-site, but I'm not aware of it with anyone here.