Page 3 of 3

Posted: Mon Apr 29, 2019 6:15 pm
by Lycanfire
There's going to be an optimal strategy. The point is to make people fight over what it is in the thread so people can form reads off that, call people scumfucks, threaten to replace out, and tell people not to join games with them again. That's the real reason why mountainous sucks. It's a rehash of baggage contained to that one thread and nothing else to argue over, so you're just having the same argument from page 3 to page XXX.

Posted: Wed May 01, 2019 1:14 am
by FakeGod
Hmm... I need a way to incentivize roses from picking day 10000.

There has to be a gamble option for picking early day rose. Thoughts?

Posted: Thu May 02, 2019 2:53 am
by callforjudgement
It's not obvious to me in the current setup why roses have specific days at all.

Posted: Thu May 02, 2019 3:25 am
by FakeGod
Right. Roses would just pick day infinity to avoid being outed as a rose.

Can you think of anything that would incentivize roses to pick an earlier day?

Posted: Thu May 02, 2019 3:26 am
by FakeGod
Or maybe just accept it as part of the game?

You pick a night + Gun or rose.

Posted: Thu May 02, 2019 4:23 am
by callforjudgement
I think either the roses need an additional side-effect, or else to not be night-specific.

The current setup is effectively "town can choose at the start of the game to be a night-n vigilante or a VT; scum win by killing all the VTs". The core mechanic of "power roles don't count towards the town's wincon" is interesting. Getting a working setup built around that can be hard, though.

Posted: Thu May 02, 2019 4:25 am
by callforjudgement
Oh, and I just realised a trivial way to stop town all picking N1 or whatever: don't reveal gun holders and rose holders, and let scum choose a night on which to be bulletproof. Town probably won't all pick N1 in case scum do.

Posted: Thu May 02, 2019 2:13 pm
by FakeGod
Hmm, so if rose/gun are not revealed, then mafia needs to solve the game old-fashioned way, which can be ok. Scum being bulletproof will only fix the N1 cheese I guess, since the vig will just tell people that "hey this guy was bulletproof last night - he's mafia".

I think I want to explore some other options though.

Posted: Thu May 02, 2019 2:21 pm
by implosion
I think there are many, many different directions you can take this idea in that are all very viable.

Posted: Fri May 03, 2019 1:09 am
by FakeGod
I slept on it, and I think a more elegant solution may be:

NO reveal on rose/gun - only when the player flips, they will be revealed their role
Guns pick a night and can shoot during that night
Rose pick a night and is bulletproof during that night

Posted: Sat May 04, 2019 5:11 am
by FakeGod
I think the setup is still missing something.

What if I made the number of town roses alive public?

Posted: Sat May 04, 2019 5:23 am
by callforjudgement
That's probably a good idea for the same reason that people dislike no-reveal games.

Posted: Mon May 06, 2019 6:30 pm
by FakeGod
Setup


  • 9
    Town players
    vs 2
    Mafia players

  • All players must secretly pick either a
    gun
    or a
    rose
    .
  • All players must also secretly pick a night.
  • A player who picked a
    gun
    is a
    1-shot vigilante
    during his/her
    chosen night
    .
  • A player who picked a
    rose
    is
    immune to night kills
    during his/her
    chosen night
    .
  • Upon
    death
    , it will be revealed whether the
    dead
    player picked a
    rose
    or a
    gun
    .
  • Number of remaining
    Town
    -aligned
    rose
    players will be announced at the
    beginning of each day
    .
  • If there are no
    Town
    -aligned player who picked a rose,
    Mafia
    immediately
    wins
    the game, regardless of whether there are any living
    Mafia
    members in the game.

Posted: Tue May 07, 2019 3:40 am
by callforjudgement
I like it. It'll be hard to know where the balance lies, because there's so much WIFOM involved, but the core mechanic seems correct now.

Posted: Tue May 07, 2019 3:51 am
by FakeGod
I can run some EV calculations as soon as I know what kind of assumptions I can make.

I'm testing with low numbers to see the setup in action.