[CHALLENGE] February + March Challenge

[Jingle]

8p Choose Your Rival

Phase One: Playerlist is randomized. Starting with player 1, each unpaired player picks another unpaired player that is guaranteed not to be their scum buddy. Two scum are chosen with double the chances for the players who chose. Pairs are published.

Phase Two: The first lynch is on a rival pair. Both players flip.

Phase Three: Normal nightphase.

The game progresses normally from there.

Spoiler: Example

1
2
3
4
5
6
7
8

1 recieves a list of 2-8 to choose from and chooses 2. 2, being paired, doesn't get to choose. 3, being next in line, gets to choose between 4-8.

After pairings are determined:

1 and 2
3 and 7
4 and 6
5 and 8

are the rivals. The first scum is determined by a d12 roll on the list of

1,1,2,3,3,7,4,4,6,5,5,8. 1 is rolled. The second scum is determined by a d9 roll on the list of 3,3,7,4,4,6,5,5,8 since neither 1 or 2 can be scum #2.

Day one is a vote to lynch one of:

1 and 2
3 and 7
4 and 6
5 and 8

Day 2 is either 4v1 or 3v2.

[Jingle]

As much as I like the idea of forcing pops to do calculus to determine setup, you should probably limit number submissions to integers (or at least reasonable numbers that don't require comparison of limit sizes).

Additionally, how do you resolve ties in priority?

Unfortunately, I think Jester/Unjester/Survivor are all precluded by the no 3p rule.

Do the switch kills end the day? If so, this setup is crazy broken for scum.

This autowins for town.

Setup one, Bjerg massclaim:

Maf claims: No cc, you lynch Bjerg and they flip maf. Quicklynch into 2v1 LYLO and Bar the doors.

No maf claims: Known setup 1, Quicklynch into LYLO and Bar the Doors.

Setup two, Bjerg massclaim:

Maf claims, Bjerg ccs. Known Setup 2 quicklynch into LYLO and Bar the Doors.

Maf doesn't claim. Lynch Bjerg and Bar the doors.

[Jingle]

Interesting.

Falls into the design pit of forcing wagon placement management, but it might be interesting enough to play it anyway. I'd guess slightly scumsided, but definitely marathonable (and probably blitzable with a group that are all on the same side of the hyper-optimize or meme-away fence).

[Jingle]

Oh, hmm. I missed that. So it's a D1 coinflip based on claim trustworthiness.

[Jingle]

Why would you ever not no lynch with no claims until mafia takes the 1/5 shot for autolose?

Otherwise it's an exactly 50/50 choice you have to make on Day 1. Town can't lose Setup 1 except by wrongbarring.

[Jingle]

Interesting gambit, but that still falls into the it's a 50/50 solved on D1 criteria. I was wrong that it autosolves, but the game still isn't about finding mafia. It's about guessing whether you're in one or two. Whether that's a flaw or a benefit of the design is subjective.

I do think by the logic of #45 it'd be better as a 7 person or without the option of no lynch.

[Jingle]

I agree. I usually ignore setups that i think are inherently bad.

[Jingle]

Is information reciprocal? Like, say I'm informed I'm not Starcrossed with Pops and Oka, are Oka and Pops informed they're not Starcrossed with me? Is it confirmed that tragic fates will be in the information? Is the relationship between the mafias static? Are they always Starcrossed/Tragic?

[Jingle]

If switches are day ending LYLO happens 2 days sooner. (Scum can just mash buttons at daystart.)

If switches aren't day ending, leashing activation and using buttons exclusively for as long as possible is optimal, although yolo shooting is still likely from town, tbh.

[Jingle]

If they pair S-T they literally cannot win unless scum self sacrifices before a town sacrifices which I think is unhealthy outside of marathon. (Since town sacrificing scum generates an IC and now you just force every pair to sacrifice)

?

S-T
S-T

Remaining would be a scumwin.

It's probably not optimal, but it is theoretically winnable.

[Jingle]

Ah yes, scum do have a log-in draw. Snazzy. That is not the One Thing making or breaking whether I view that as a playable setup

I'd guess it's intended as a parity win (which is fairly standard), although I agree that S/T pairings being possible (if ill advised) doesn't mean its a great setup.

[Jingle]

theoretically optimal play is to mass claim

since power roles are conf town

It's not really potentially 3 clears. It is 3 clears. Scum have to kill one of the two PR's, but even if they do the public cop goes through and clears/guilties a player. If scum cc's public cop they die. If scum cc's named townie they trade a scum lynch for potentially a clear lynch.

[Jingle]

It's a simple enough clarification for Hectic to make, and he probably should.

I'm more interested in the switches answers anyway, though.

[Jingle]

4 players get informed.
4 players uninformed.

Massclaim information D1.

Pair all pair claims outside of the impossible to OTP information until the lategame.

If there are 6 players in the mafia pool, there are 15 possible pairings among those 6. You can't quite brute force the autowin (a single player must be paired 5 times for autowin) but you can get it close enough. Scum therefore shouldn't fakeclaim to make a 6v2 split.

If it's a 5:3 split, you should be able to logic at least one pairing for the three based on information claims early. You have a good chance of a first day OTP.

4:4 split, 32% chance of no OTP in the first phase if S/S starcrossed, 42% if S/T starcrossed.

tl;dr, I still think X=3 is better for balance.

Also, as written there's no mafia loss for both mafia dying. is that intentional?

[Jingle]

The way I parse it each player has 1 of each type.

So Player A would be starcrossed with Player B, Tragic with Player C, and Informed they're not starcrossed with players C and D.

[Jingle]

A Sample setup would be:

Starcrossed:

A:B
C:D
E:F
G:H

Tragic

C:E
D:H
B:F
G:A

Informed:

C: H,B
D: A,G
B: G,F
G: H,A

Mafia: F, D

I'm still not certain if the tragic fates are guaranteed to be included in the 2 informed for the informed players, but I would assume not because it's already a ton of information.

[Jingle]

Miller is functionally the same as Named town there, since they're claiming D1 anyway.

Question, is the dayvig limited shot or can they continue shooting on future days if they survive?

[Jingle]

I'm probably going to have to make setup 1 use a public tracker or other unreliable public investigative. I'm thinking public tracker and named townie might work, if not maybe D2 public cop and named townie.

Could give scum a rolestopper in Setup 1.

If the dayvig is a full dayvig, that setup is viable as a choice too.

70% to hit scum D1 and go into 4p with a conftown and a vigshot (63%). Overall that's a
55% town EV when you take into consideration the 1/10 where you catch both scum D1.

[Jingle]

I'd play it, but it's probably better as a marathon or a blitz than a prolonged game.

[Jingle]

Once per day, he confirmed that for me.

[Jingle]

There's probably a better flavor for it.

You could flavor it with the Avalon board game.

Merlin/Percival/Morgana/Mordred 2 Knights 2 Councilors.

As worded, town wins if they lynch Xayah D1 and Xayah and Rakan aren't paired. I assume that's not intended and Mafia wins if Kogmaw is never lynched.

I'm torn as to whether vote for town or massclaim is a better option as written, and would have to do the EV calcs to figure it out. Massclaim is exactly 50%, so assuming that playing it as a vote for town is viable it's almost certainly more fun.

Also, combining vote for town and flagbearer mechanics is pretty interesting.

[Jingle]

I feel like that's probably balanced based on cop 9er and vig 9er being roughly balanced. I'm not sure how likely mafia being hit D1 is in practice.

Playability wise, it's probably fun if the Ascended player works well with the town, and less so if they, y'know, don't. Doubt it's chaotic though, especially compared to half the other suggestions. It's like... A reasonable game of mafia or something.

[Jingle]

so Ascended do nothing on first night bc he is immortal anyway
lets say on day 2 mislynch happen
night 2 is 4 town vs 2 scum
one out of 3 players is dead and one out of 5 players is checked with cop, no result probably means maf covering maf
With green result becomes 2 IC vs 1 townie and 2 scum.
The swing is real here.
After that checked player can be protected with doctor, which leaves us with
in case of fractional kill: 2 IC vs 1 scum
in case of scum not killing: 2 IC vs 1 townie and 1 scum
Remember that ascended still has vig shot, so in case of mislynch he can nightkill last scum.

This discounts the possibility of the cop target hitting the NK, which is very nontrivial. I'm also unsure why the IC wouldn't act N1.

I'd actually wager that Vig -> ScumFlip -> Cop
-> 2TownFlip -> Doc

would be slightly more optimal, though not enough to make it an always choose sort of thing. Note: I haven't done any of the math to support or refute this position, just my thoughts at a glance.

[Jingle]

The best patch is my final submission, best I got

You should make it clear that town aren't allowed to post e or b, otherwise they can conftown by doing so and not being lynched. (I assume this is intended, but making it explicit is probably a good idea.)

[Jingle]

viewtopic.php?f=115&t=82646

[Jingle]

I disagree with assuming a creator who didn't submit a list would vote for only their own, FWIW.

I didn't vote but wouldn't have ranked mine as top setup either.