Open Setup Autopsies (HELP)

[mhsmith0]

I'm going to base this on the argument that C9++ is, in practice, balanced. Its history seems to suggest so. However, town has only won 1/8 JK9++ games before I won the last one. The most recent iteration had an extremely imbalanced playerlist (D1 non-bus scum lynch) and only 2 mafia.

c9++ is vaguely sort of balanced in terms of overall factional win equity between town and groupscum, but (imo) that's mainly because half of the rands have serial killers, which tank town win equity, and half don't, which tanks wolf win equity.

[mhsmith0]

I think that, in fact, there's a substantial problem for modular setups in the c9++/jk9++ vein is that win equities for everyone will vary drastically based on the 50-50 odds of a Sk showing up or not.

[mhsmith0]

Yes, that would def. be the logical thing.

I mainly think that there's an inherent problem where you're essentially balancing wincon odds based on omething entirely outside of players' control, i.e. whether or not there's an SK.

But that's a personal preference thing mainly *shrugs*

[mhsmith0]

I suppose the next question would be; can the serial killer be swapped out for another 3rd party role? While still maintaining equal, or better, balance than the SK had in games?

Edit: I'm referring to JK9++.

Essentially the nature of the balancing means that if the game is overall reasonably balanced for town/scum winrates, then you're probably going to get town wins solidly >50% when there's no SK and solidly below 50% when there is

Ideally, you'd instead hvae a setup where no SK is a given and then it's balanced given that assumption
And SK is a given and it's balanced given that assumption

but that's not any of the _9++ setups.

I mean, it's not really conceptually impossible to balance around a modular or other form of semi-open setup without using the possibility of an SK rand as a crutch to make the win rates more even. That the setup uses it as a crutch is a weakness inherent in its design.

[mhsmith0]

It's probably possible to salvage a switch-like format by designing a game vaguely like it but without any n0 actions, but I feel like it's better to just *switch* your attention instead to a format that isn't in large part a giant RNG outcome (since games with heavy n0 action usage are blatantly RNG given lack of control of who shoots/peeks/protects who)

[mhsmith0]

The essential problem with masons and monks in terms of balance is that it's super reliant on cross-kills, since 9/2/2 with no killing or protective roles makes life challenging for town given site meta. Town could lynch correctly the first three days and get:

9/2/2->9/2/1->7/2/1
7/2/1->7/1/1->5/1/1
5/1/1->5/1/0->4/1/0
1 scum left, but day before LYLO

or...
9/2/2->9/2/1->7/2/1
7/2/1->7/2/0->6/2/0
1 scum team fully operational, and the day before MYLO

So even three straight correct lynches could, int eh absence of cross-kills or overlapping kills, net out to being one day before LYLO/MYLO

I'd almost say just check a town vig or doctor into the mix and be done with it *shrugs*