ILLEGAL Assasin's Gambit Sit-N-Go

[Kaiveran]

shhh don't tell the mods
i'm actually running this here because I'm not sure it counts as mafia by MU standards

Setup

x1 Red Assassin
x1 Blue Assassin
x8 Townies

Both Assassins kill every Night.

2 Townies are selected to be Wards, one for each Assassin.
An Assassin knows who their Ward is, and must protect them. If a Ward dies, the associated Assassin loses and leaves the game.
Wards are not aware of their status.

4 Townies are selected to be Targets, two for each Assassin.
Two are Cold Targets – known to the Assassin that must kill them, but not knowing they are a Target.
The other two are Hot Targets – they know they are a Target, but the Assassin that must kill them does not know them.

All Townies except Hot Targets receive an identical VT-like PM.
----------------

An Assassin wins when both their Targets and the other Assassin are dead.

The Town wins when both Assassins are dead.

48/24
Will have midnight deadlines -5:00UTC and start ~24 hrs after filling.

[Kaiveran]

Signups

Empty / N/A

Extra Queue

Empty / N/A

Subs

Empty / N/A

Spectators

Empty / N/A

Use:

• /in to join as a player.
• /out to drop out and remove yourself from any list.
• /sub to join as a substitute.
• /spec to join as a spectator.

Game start: August 31st, 2023 at 9:43 PM EDT

[Giga Punch]

/in

[Kaiveran]

@matt

[Zugzwang]

Do the assassins have any kills? Town has a 100% win breaking strategy if they don't.

[Kaiveran]

Do the assassins have any kills? Town has a 100% win breaking strategy if they don't.

Yes, both of them can kill every Night. Probably should have included that in the setup box (Thought it was obvious)

[Zugzwang]

Do the hot targets know which assassin must kill them?

[Kaiveran]

Do the hot targets know which assassin must kill them?

Yes.

[Zugzwang]

Do the hot targets know which assassin must kill them?

Yes.

I quickly made something (hopefully no bugs!) to simulate this setup many times. I assumed no claims (which is probably suboptimal but way easier to simulate), and that assassins always kill their cold target if they're alive; otherwise they kill randomly.

When using the wincons exactly as stated in the OP, the results are:
Red winrate is 0.3937
Blue winrate is 0.385
Town winrate is 0.3563
The chance 0 players win is 0.0
The chance 1 players win is 0.8738
The chance 2 players win is 0.1174
The chance 3 players win is 0.0088

When using the wincons in the OP but also requiring assassins to be alive to win, the results are:
Red winrate is 0.3283
Blue winrate is 0.3153
Town winrate is 0.3564
The chance 0 players win is 0.0
The chance 1 players win is 1.0
The chance 2 players win is 0.0
The chance 3 players win is 0.0


So assuming claims don't change too much, this looks pretty well-balanced.

[Kaiveran]

Do the hot targets know which assassin must kill them?

Yes.

I quickly made something (hopefully no bugs!) to simulate this setup many times. I assumed no claims (which is probably suboptimal but way easier to simulate), and that assassins always kill their cold target if they're alive; otherwise they kill randomly. When using the wincons exactly as stated in the OP, the results are: Red winrate is 0.3937 Blue winrate is 0.385 Town winrate is 0.3563 The chance 0 players win is 0.0 The chance 1 players win is 0.8738 The chance 2 players win is 0.1174 The chance 3 players win is 0.0088 When using the wincons in the OP but also requiring assassins to be alive to win, the results are: Red winrate is 0.3283 Blue winrate is 0.3153 Town winrate is 0.3564 The chance 0 players win is 0.0 The chance 1 players win is 1.0 The chance 2 players win is 0.0 The chance 3 players win is 0.0 So assuming claims don't change too much, this looks pretty well-balanced.

@Zugzwang Hmm, why is the Red Assassin getting an edge here? Both Assassins kill every Night and their targets are symmetrical so I don't understand.

[Zugzwang]

Do the hot targets know which assassin must kill them?

Yes.

I quickly made something (hopefully no bugs!) to simulate this setup many times. I assumed no claims (which is probably suboptimal but way easier to simulate), and that assassins always kill their cold target if they're alive; otherwise they kill randomly. When using the wincons exactly as stated in the OP, the results are: Red winrate is 0.3937 Blue winrate is 0.385 Town winrate is 0.3563 The chance 0 players win is 0.0 The chance 1 players win is 0.8738 The chance 2 players win is 0.1174 The chance 3 players win is 0.0088 When using the wincons in the OP but also requiring assassins to be alive to win, the results are: Red winrate is 0.3283 Blue winrate is 0.3153 Town winrate is 0.3564 The chance 0 players win is 0.0 The chance 1 players win is 1.0 The chance 2 players win is 0.0 The chance 3 players win is 0.0 So assuming claims don't change too much, this looks pretty well-balanced.

@Zugzwang Hmm, why is the Red Assassin getting an edge here? Both Assassins kill every Night and their targets are symmetrical so I don't understand.

It's a simulation with random choices, not an exact calculation, so a little variation is normal (only 10000 runs, which isn't that much here).

[Kaiveran]

It's a simulation with random choices, not an exact calculation, so a little variation is normal (only 10000 runs, which isn't that much here).

Hmm, methinks we should have a chat about game simulation sometime...I have many ideas!

Anyways, thanks for verifying that this setup is theoretically sound @Zugzwang, even if no one here will ever play it XD

I do think I have an alternative venue to try though