TWG CXV: Just a Numbers Game

[E. Gadd Industries]

I'm not sure if it really matters too much if we organize our numbers. Leaning on random picking unless I'm convinced before then ig

I had that thought as well tbh. It seems like it doesn't really make a ton of difference picking random vs not unless we get a plan spun up.

Also, with the ~25-27% on the birthday problem application, I was thinking about it; IF we are going on the wolf picking the same number as two humans, wouldn't that be 26% * 1/100 (such that a wolf with no info whatsoever chooses the same number out of 100 as two humans)? If p=0.0026, that is certainly not the end of the world if we can't come up with a solid plan without letting the wolf know anything useful.

One other thought I had recently: how would it be best to go about day phases? The wolf has quite a bit to burn through to win, but we can extend that by not going for lynches straight out the gate (unless we're 95% certain we're going for a wolf). Unless you're the phantom number swapper, there's no fear that the wolf is going to be able to target any special roles (even as the number swapper you can't be sure of a targeted wolf attack), so we don't really have anyone to overtly protect aside from those who would bring good strategy to the human side. But even those can't be targeted.
Basically what I'm saying is, I'm not sure what the precedent is on this, but I think it could be worthwhile to not go for lynching during the day phases unless we're dead serious on the person we're putting on the chopping block. It gives us more time to make a strategy and we don't really have anyone specific to worry about dying due to wolfing, whereas anyone we lynch who isn't the wolf means the game still goes on and not in town's favor.
This does have the caveat though that we will need to find another way to flush out the wolf, although I'm not sure how that'll be functionally different from if we were to vote for lynch candidates before we had strong suspicions about a particular person.

[BlackDragonSlayer]

Omg @BDS what happens if the numbers are the exact same distance away? Say someone picks 6 and someone picks 8 and the wolf picks 7, who dies ?

Just for full clarification, as mentioned in the game post:

Whoever's number is closest to the number the Head Honcho picks dies. In case of a tie, all tied players die.

So both 6 and 8 would die.

[Toby]

I had that thought as well tbh. It seems like it doesn't really make a ton of difference picking random vs not unless we get a plan spun up.

Also, with the ~25-27% on the birthday problem application, I was thinking about it; IF we are going on the wolf picking the same number as two humans, wouldn't that be 26% * 1/100 (such that a wolf with no info whatsoever chooses the same number out of 100 as two humans)? If p=0.0026, that is certainly not the end of the world if we can't come up with a solid plan without letting the wolf know anything useful.

One other thought I had recently: how would it be best to go about day phases? The wolf has quite a bit to burn through to win, but we can extend that by not going for lynches straight out the gate (unless we're 95% certain we're going for a wolf). Unless you're the phantom number swapper, there's no fear that the wolf is going to be able to target any special roles (even as the number swapper you can't be sure of a targeted wolf attack), so we don't really have anyone to overtly protect aside from those who would bring good strategy to the human side. But even those can't be targeted.
Basically what I'm saying is, I'm not sure what the precedent is on this, but I think it could be worthwhile to not go for lynching during the day phases unless we're dead serious on the person we're putting on the chopping block. It gives us more time to make a strategy and we don't really have anyone specific to worry about dying due to wolfing, whereas anyone we lynch who isn't the wolf means the game still goes on and not in town's favor.
This does have the caveat though that we will need to find another way to flush out the wolf, although I'm not sure how that'll be functionally different from if we were to vote for lynch candidates before we had strong suspicions about a particular person.

Sorry I have to disagree with no lynching. Us lynching is the only controlled kill in this game.  The wolfings give us no information, but a lynching confirms if any suspicion, even slight is wolf or not.

We are best using our greatest power the most we can

[E. Gadd Industries]

Sure. It was a thought I had, wanted to see if the logic would float. But I suppose it was built on this idea that there's some deep level strategy that could be useful in outing the wolf by manner of play. I'm still not convinced that that doesn't exist in a game like this, but I'm also struggling greatly with the math side; it's an important part that I have little hope I'll be able to parse through in any meaningful way.

Also just gonna put it out there for any metagamers, I don't look to be as active this game as I have been. Activity is useful, but I'm also trying to practice healthy boundaries with online things in general. Really with everything in my life at this point. But anyway, yeah. I'll still contribute to the game and be as thoughtful with the posts I do make. Just... not as many posts or not as often online.

[SpecsFlyer17]

Also, with the ~25-27% on the birthday problem application, I was thinking about it; IF we are going on the wolf picking the same number as two humans, wouldn't that be 26% * 1/100 (such that a wolf with no info whatsoever chooses the same number out of 100 as two humans)? If p=0.0026, that is certainly not the end of the world if we can't come up with a solid plan without letting the wolf know anything useful.

Not quite, because the wolf doesn't need to pick the exact number of the duplicate pick; he
only needs to be closest to the pick. You're right that this makes the odds much lower than 26%, but its not quite a 1/100 multiplier.

On a broader scale, yes you are correct that I don't really think it makes much of a difference if we all choose randomly. We're talking about unlikely scenarios; and while there definitely is a way to avoid a duplicate pick, it requires 100% cooperation.

[BlackDragonSlayer]

A reminder to new and returning players to please not edit any game posts you make.

[SpecsFlyer17]

A reminder to new and returning players to please not edit any game posts you make.

Guilty as charged  It was just a typo for transparency to the other players. Won't happen again!

[BlackDragonSlayer]

Also a reminder that the phase will end in approximately 24 hours from now.

[magnet]

Oh, umm. The game's started, huh?  Sorry for not checking in earlier.
We're having quite a chilly winter here in NZ and I am so sick. I can't really think straight at the moment, but despite that, I've taken a look through your comments.

I checked your math on the birthday problem thing, and it looks sound to me, not that I'm an expert anyway. Regardless, I'm not sure how useful it is. As SpecsFlyer17 mentioned, the wolf doesn't need to pick exact numbers; whoever is closest will die. I don't have any grand plans yet so I'm just going to pick a random number. 

[A# Minor]

Oh my, I didn't expect to be so inactive today. Too bad I'll be inactive tomorrow as well because it's my sister's birthday. But I should be up until like 2 am so no worries


Anyways, random picking for the win, and I don't understand any of the math you guys are doing :')

[XiaoMigros]

i was joking

[E. Gadd Industries]

i was joking

What about? The Sun Tzu thing?

Anyway, yeah, today ended up being busy, but I did submit a number early on (although I did plan to change it if we got some sort of reasonable plan going).

[XiaoMigros]

yeah, i missed page 2 of the messages oops

[Toby]

What do you think about organising numbers, XiaoMigros?

[XiaoMigros]

I think it's the only way for us to bring any sort of strategy into the game and put pressure on the wolf, so I am in favor of it. However, it's largely pointless if not everyone participates (and for this reason THC is #1 on my sus list)