Open 876 - Fire and Ice - Postgame

[Enchant]

I am also not in mood of coaching mafia.

[usesPython]

Ok, but scum don't want to make town lose, they want to win, don't they?

Just so we're on the same page when it comes to EV calcs, are we using the Charisma Model or are we assuming random lims?

-Nameless

[biancospino]

Random lims, except that the doc is never limmed.

Also I'm not really doing an exact calculation, I'm just simulating ~100k games and seeing the win proportions

[usesPython]

Random lims, except that the doc is never limmed.

Also I'm not really doing an exact calculation, I'm just simulating ~100k games and seeing the win proportions

Intuitively random lims and actions should decrease the chances of the same target = no nightkill rule applying (Since it means scum might be shooting people that are scummier than they are), I'll try to finish coding a simulation running on the charisma model and get back to you by the end of the day

-Nameless

[Alianna]

1.15

Votecount 1.15

usesPython (E-2): Oclaxian Empire, The Bulge, Aureal, Doctor Drew, Black
Doctor Drew (2): usesPython, Enchant
Black (2): Kokichi Oma, biancospino
The Bulge (1): furtiveglance

Not Voting (3): Malakittens, osuka, HighPrincessErinys

With 13 players alive, it takes 7 votes to secure an execution.

The day deadline is in (expired on 2023-04-20 18:48:28).

[biancospino]

Random lims, except that the doc is never limmed.

Also I'm not really doing an exact calculation, I'm just simulating ~100k games and seeing the win proportions

Intuitively random lims and actions should decrease the chances of the same target = no nightkill rule applying (Since it means scum might be shooting people that are scummier than they are), I'll try to finish coding a simulation running on the charisma model and get back to you by the end of the day

-Nameless

There is a specific reason why I don't want you to get back on me on that specific point for now. I'm sure you already see it yourself.

[usesPython]

Spoiler: Brick

Random lims, except that the doc is never limmed.

Also I'm not really doing an exact calculation, I'm just simulating ~100k games and seeing the win proportions

Intuitively random lims and actions should decrease the chances of the same target = no nightkill rule applying (Since it means scum might be shooting people that are scummier than they are), I'll try to finish coding a simulation running on the charisma model and get back to you by the end of the day

-Nameless

There is a specific reason why I don't want you to get back on me on that specific point for now. I'm sure you already see it yourself.

Yeah I haven't finished writing the script for this, obviously if it's -EV I'll just tell you I ran the numbers and shut up about it

-Nameless

[Oclaxian Empire]

Ok, but scum don't want to make town lose, they want to win, don't they?

In order for scum to win they need to make town lose though?

[usesPython]

That's what the nightkill is for isn't it? Why go for a "Flip Black and then flip me if Black flips town" gambit as scum here?

-A

Going back to this with python's recent posts in mind, I'm having a really hard time believing that they can't see the benefit of scum proposing a gambit here. I'll even go as far as to say they are deliberately acting obtuse about it

If what python thinks is true that town basically loses if we lim town here, it would make

complete

sense for scum to propose a gambit. That's one townie down and there isn't even a guarantee the scum gets limmed D2. It's so easy to get out of a gambit

I think this contradiction in python's logic suggests they are a wolf

VOTE: python

Actually I think this is coming from a mech misunderstanding: Just because groupscum can do something that is -EV for town doesn't mean that it's +EV for that groupscum. Your argument that we're forcing a 1f1 with a town does get that it'd be +scum winrate in that scenario but it misses the point that it'd be -groupscum winrate (e.g. If we were

Ice Mafia

then it'd cause

Town

and

Ice Mafia

win rate to go down while

Fire Mafia

win rate goes up)

-Nameless

Ok, but scum don't want to make town lose, they want to win, don't they?

In order for scum to win they need to make town lose though?

They also need to make the other groupscum lose though, so they won't take actions that put town slightly worse off if it means the enemy groupscum are even better off (e.g. a

25

/

25

/

50

winrate is worse for

Ice Mafia

than a

33

/

33

/

33

winrate)

-Nameless

[Oclaxian Empire]

I’m not gonna read any mech talk I hope you know that. I literally did not read any of the last three pages because of that + EV discussion.

[usesPython]

I’m not gonna read any mech talk I hope you know that. I literally did not read any of the last three pages because of that + EV discussion.

tl;dr scum also need to make other scum lose, not just town

-Nameless

[osuka]

skimmed the thread before work again and after discussing with my team of researchers, we have confidently come to the conclusion that we have never seen a more useless mech discussion before

[usesPython]

skimmed the thread before work again and after discussing with my team of researchers, we have confidently come to the conclusion that we have never seen a more useless mech discussion before

smh discussing the game outside the gamethread, Alianna plz deal with this rulebreaker

-A

[Black]

Yeah the mech isn't even some huge revelation. It's just circular and repetitive. I almost feel like python wants to drown the thread out and unmotivate people

[usesPython]

Yeah the mech isn't even some huge revelation. It's just circular and repetitive. I almost feel like python wants to drown the thread out and unmotivate people

Stop getting it wrong and I'll stop repeating myself

-Nameless

[Black]

Actually I think this is coming from a mech misunderstanding: Just because groupscum can do something that is -EV for town doesn't mean that it's +EV for that groupscum. Your argument that we're forcing a 1f1 with a town does get that it'd be +scum winrate in that scenario but it misses the point that it'd be -groupscum winrate (e.g. If we were Ice Mafia then it'd cause Town and Ice Mafia win rate to go down while Fire Mafia win rate goes up)

That's assuming the 1v1 actually results in both players getting limmed. Like I said it's too easy to get out of a gambit, so proposing the gambit on D1 should be a viable scum tactic to you

And I think it's something you as scum would propose here. We now know you think the town essentially loses if we lim green D1, so it makes perfect sense for you to propose a gambit based off of weak associatives to try and push a mislim here

And you'll just come up with some mech BS or find a way to wiggle your way out of your end of the gambit D2

[usesPython]

Our mechposts have been the basics of the basics and people are still getting it wrong, there's literally no point for me into deeper mech when the only person actually bothering to read and understand it is Bianco and it's frustrating as hell

-Nameless

[usesPython]

Actually I think this is coming from a mech misunderstanding: Just because groupscum can do something that is -EV for town doesn't mean that it's +EV for that groupscum. Your argument that we're forcing a 1f1 with a town does get that it'd be +scum winrate in that scenario but it misses the point that it'd be -groupscum winrate (e.g. If we were Ice Mafia then it'd cause Town and Ice Mafia win rate to go down while Fire Mafia win rate goes up)

That's assuming the 1v1 actually results in both players getting limmed. Like I said it's too easy to get out of a gambit, so proposing the gambit on D1 should be a viable scum tactic to you

And I think it's something you as scum would propose here. We now know you think the town essentially loses if we lim green D1, so it makes perfect sense for you to propose a gambit based off of weak associatives to try and push a mislim here

And you'll just come up with some mech BS or find a way to wiggle your way out of your end of the gambit D2

Please look at the wagon on us and tell us how we live past d2

-A

[Black]

Our mechposts have been the basics of the basics and people are still getting it wrong, there's literally no point for me into deeper mech when the only person actually bothering to read and understand it is Bianco and it's frustrating as hell

-Nameless

Not everyone likes math or is even good at it. Same with critical thinking

[Black]

Please look at the wagon on us and tell us how we live past d2

You're smart. I'm not putting it past you to figure it out

[furtiveglance]

Actually I think this is coming from a mech misunderstanding: Just because groupscum can do something that is -EV for town doesn't mean that it's +EV for that groupscum. Your argument that we're forcing a 1f1 with a town does get that it'd be +scum winrate in that scenario but it misses the point that it'd be -groupscum winrate (e.g. If we were Ice Mafia then it'd cause Town and Ice Mafia win rate to go down while Fire Mafia win rate goes up)

That's assuming the 1v1 actually results in both players getting limmed. Like I said it's too easy to get out of a gambit, so proposing the gambit on D1 should be a viable scum tactic to you

And I think it's something you as scum would propose here. We now know you think the town essentially loses if we lim green D1, so it makes perfect sense for you to propose a gambit based off of weak associatives to try and push a mislim here

And you'll just come up with some mech BS or find a way to wiggle your way out of your end of the gambit D2

This last line seems really uncharitable and disproportionate

VOTE: Black

[Black]

Besides, if you think the town essentially loses with a D1 mislim, how does you getting limmed D2 change those odds? Would your team still be favored to win?

[Black]

Actually I think this is coming from a mech misunderstanding: Just because groupscum can do something that is -EV for town doesn't mean that it's +EV for that groupscum. Your argument that we're forcing a 1f1 with a town does get that it'd be +scum winrate in that scenario but it misses the point that it'd be -groupscum winrate (e.g. If we were Ice Mafia then it'd cause Town and Ice Mafia win rate to go down while Fire Mafia win rate goes up)

That's assuming the 1v1 actually results in both players getting limmed. Like I said it's too easy to get out of a gambit, so proposing the gambit on D1 should be a viable scum tactic to you

And I think it's something you as scum would propose here. We now know you think the town essentially loses if we lim green D1, so it makes perfect sense for you to propose a gambit based off of weak associatives to try and push a mislim here

And you'll just come up with some mech BS or find a way to wiggle your way out of your end of the gambit D2

This last line seems really uncharitable and disproportionate

VOTE: Black

Is this just how you play Mafia now?

[usesPython]

Please look at the wagon on us and tell us how we live past d2

You're smart. I'm not putting it past you to figure it out

We're quite literally the designated miselim d2; Without even considering who in the list is scum here's the people that'd push us d2:

• Oclaxian
  
• Black
  
• Bulge/Drew
  
• Osuka
  
• Aureal

It's 6 to lim tomorrow (unless there's a no-kill in which case 7 to lim)

-A

[usesPython]

Besides, if you think the town essentially loses with a D1 mislim, how does you getting limmed D2 change those odds? Would your team still be favored to win?

Your entire argument here is based on the idea that it's somehow possible for us to not get limmed d2, please get out of your tunnel for a minute and really look at the gamestate

-A