TWG CXXI: True Love Game Thread

[Oricorio]

</3 True Love </3

Wolves:
1. Wolf Lover
2. Wolf Lover
Town:
3. Town Lover
4. Town Lover
5. Town Lover
6. Town Lover
7. Town Lover
8. Town Lover

Mechanics:

• </3Basically the setup True Love</3
• </3Every player is placed in groups of two as lovers. Wolves are guaranteed to be paired with town</3
• </3If a player is lynched, their lover dies as well</3
• </3The game is played nightless</3
• </3Town wins by killing all of the wolves, wolves win when they're 50% of remaining players</3

Role PMs:

Spoiler

You are a Town Lover. Your lover is _____, and if they die you will die as well. You win by eliminating all wolves.

You are a Wolf Lover. Your lover is _____, and if they die you will die as well. You win if wolves make up half the remaining players.

[close]

Players:
1.</3 BDS </3
2.</3 ThatHiddenCharacter </3
3.</3 SpecsFlyer17 </3
4.</3 Nakah </3
5.</3 Raeko </3
6.</3 TheZeldaPianist275 </3
7.</3 mastersuperfan </3
8. </3 Toby </3

[Oricorio]

Okay, game begins now! Sorry for posting it in the wrong forum initially. I can extend the deadline by a couple minutes in case anyone cares, but for now phase ends in 48 hours.

[ThatHiddenCharacter]

Just to be clear, we are allowed to talk to our lovers in DMs, right? (With the host included, of course.)

[mastersuperfan]

huzzah!

I haven't thought about the setup too too much but my first inclination is that it's beneficial for everyone to reveal their pairings...? since the wolves know each other, by default the wolves know 2 pairings for sure and can narrow down the remaining 4 players to 3 possible pairings. meanwhile humans start off knowing only their own lover, with I think 15 possible pairings among the remaining 6 players. so it seems to me like it's just best to reveal it all right away to ensure everyone has the same information.

I have thought briefly about whether there's some sort of information-sharing scheme where we could equalize everyone's information more without giving away everything to the wolves, but since the wolves start off with such a strict informational advantage, I'm skeptical such a scheme exists.

thoughts?

[mastersuperfan]

on the other hand maybe it's possible that if we keep it secret wolves accidentally slip up and reveal they know too much about the pairings? hmm

[Oricorio]

Just to be clear, we are allowed to talk to our lovers in DMs, right? (With the host included, of course.)

Yes

[BlackDragonSlayer]

on the other hand maybe it's possible that if we keep it secret wolves accidentally slip up and reveal they know too much about the pairings? hmm

Personally, I think there's merit to the plan of everyone revealing their lover, but at the same time accidental slip-ups of wolves trying to defend the human lovers of *both* wolves is likely going to be a big part of the game. If we put everything on the table there's less chance of catching such slip-ups.

[BlackDragonSlayer]

In contrast, I feel like humans might actually be inclined to lynch their own lover if they genuinely believe that lover could be a wolf.

Since it's a no-nights game, we're going to be relying a lot more on reading into dynamics like this.

[SpecsFlyer17]

Personally, I think there's merit to the plan of everyone revealing their lover, but at the same time accidental slip-ups of wolves trying to defend the human lovers of *both* wolves is likely going to be a big part of the game. If we put everything on the table there's less chance of catching such slip-ups.

It's reasonable that a given player will defend their known partner in the interest of self-preservation, which as BDS mentioned, is one person per player. However, each wolf has 3 known players to guard: the other wolf, wolf 1's lover, and wolf 2's lover.

Keeping pairings hidden for the first day, noting who attempts to defend, then revealing pairings and seeing how peoples' defenses change could yield some information. Specifically if anyone defends multiple people across days.

Idk though.

[SpecsFlyer17]

Also worth noting that there's a 50% chance of eliminating a wolf D1; either directly or via their lover.

[ThatHiddenCharacter]

I think Spec's plan has a bit more merit to it than the others, but nothing that's been said so far seems particularly wolfy to me, either.

[mastersuperfan]

there are 6 humans. 4 are paired with other humans and 2 are paired with wolves. given uniformly random role distribution, from a human's point of view, there's a 2/6 = 1/3 chance you're paired with a wolf.

then:
- with 1/3 probability, your lover is a wolf. thus one other pair is human-wolf, and the remaining two pairs are human-human. then lynching one of the other 3 pairs at random has a 1/3 chance of killing a wolf.
- with 2/3 probability, your lover is a human. thus two other pairs are human-wolf, and the remaining pair is human-human. then lynching one of the other 3 pairs at random has a 2/3 chance of killing a wolf.

thus lynching one of the other 3 pairs at random, in total, has a 1/3 * 1/3 + 2/3 * 2/3 = 5/9 chance of killing a wolf.

tl;dr for humans:
- if you sacrifice your own pair, there's a 33.3% chance you kill a wolf
- if you sacrifice a different pair, there's a 55.6% chance you kill a wolf

so both humans and wolves have incentives to protect themselves day 1.

[mastersuperfan]

so both humans and wolves have incentives to protect themselves day 1.

(and their lover)

[mastersuperfan]

case 1: suppose a human-wolf pair gets lynched d1. then the remaining pairs are h-w, h-h, h-h. assuming all future lynches happen uniformly at random, there's a 1/3 chance that wolves win and a 2/3 chance that humans win.

case 2: suppose a human-human pair gets lynched d1. then the remaining pairs are h-w, h-w, h-h. then under the same assumptions there's a 2/3 chance that wolves win and a 1/3 chance that humans win.

then on d1, if you are a human and you get lynched, since your lover is a wolf with 1/3 probability, the probability you win is 1/3 * 2/3 (case 1) + 2/3 * 1/3 (case 2) = 4/9 = 44.4%. if you are a wolf and you get lynched, then the probability you win is 1/3 = 33.3% (only case 1).

in theory I think this means wolves have a stronger incentive to defend themselves than humans. I was initially toying with the idea of a human volunteering to sacrifice themselves, but since the win probabilities are 50/50 initially, a human volunteering to sacrifice themselves would still benefit the wolves, since the probability of humans winning would go down from 50% to 44.4%. so for a human's pov, the ideal play d1 is not to lynch themselves or their lover.

an exception being if you have sus on your lover and think they're a wolf.

[mastersuperfan]

It's reasonable that a given player will defend their known partner in the interest of self-preservation, which as BDS mentioned, is one person per player. However, each wolf has 3 known players to guard: the other wolf, wolf 1's lover, and wolf 2's lover.

Keeping pairings hidden for the first day, noting who attempts to defend, then revealing pairings and seeing how peoples' defenses change could yield some information. Specifically if anyone defends multiple people across days.

Idk though.

this might be interesting. a priori (before accounting for sus), each human should only have an incentive to protect one player (other than themselves), and if they don't, their win rate goes down from 50% to 44.4%. wolves have an incentive to protect three players other than themselves, and if they don't, their win rate goes down from 50% to 33.3%.

looking at this from another angle: if a human in a h-h pair is on the chopping block, both humans (2/8 players) will be opposed. if a wolf or a wolf's lover is on the chopping block, both wolves + the involved human (3/8 players) will be opposed. that's almost half the players, and so if the remaining humans aren't well coordinated, wolves could steer it away pretty easily. (of course, the process of doing so might look suspicious, depending on how conspicuous it is.)