[REVIEW] Open Setup Reviews

[popsofctown]

A much better way in general is to replace the Doc that would be following the Cop with a Jailkeeper. And an SK is an anti-town role, so an innocent result on one IS a false result, regardless of what the Cop might want.

Does anyone have a problem with this?

4 Mafia Goons

1 Serial Killer

(1-shot BP OR Inv-immune)

1 Cop
1 Jailkeeper
13 Townies

• Daystart
• Serial Killer chooses to be either 1-shot Bulletproof or Investigation-immune during the pregame
• The Serial Killer's chosen power is not revealed on death

The SK seems too weak. If he plays an antitown game, he is lynched and loses. If he plays his role very deceptively, a goodhearted jailkeeper could lock him up, preventing him from killing. Best play from there is to jail the same person again when you notice your action prevented a kill. Most rational scumteams wouldn't shoot the same failed kill twice in a row, the jailer might keep on jailing that individual, while a 1-shot BP can just be noted for later and handled if/when the jailer dies, or lynched. Most rational scumteams also wouldn't let the jailer roleblock the goon they selected twice. So a jailkeeper who jails the SK will probably jail him again, and then probably announce to the town that they've found the SK (unless the mafia is doing so well that it's not worthwhile, but the theme still is that this is rough on the SK).

Having both the jailer and either the kill or the investigation ringing a death knell for the SK seems too hard on the naturally difficult role.

I'd suggest an alternative, but it would turn into a tirade about why cop is a bad role. So.

[popsofctown]

That doesn't decrease the number of targetted night actions that are "investigating" the SK. The roleblock causes eyebrow raises the same way the jailkeep did. So you still have 3 night actions that are painful for the SK and only 1 shot of immunity, it's just now 2 of them belong to scum and one belongs to town whereas with the jailkeeper two belong to town and one belongs to scum.


The status quo is that SKs will either have a game-warping immunity or will be underpowered. So if the plan is to use only basic roles it will be hard to create something that is generous to the SK. Removing the cop is the easiest thing. If you want to keep the cop, you'll probably have to use something that's not completely vanilla.

Macho cop would cut down the number of searchlights the SK needs to dodge to 2. Preferably the variant where when the doctor protects the cop, it doesn't do anything.

Good points, how about this:

4 Mafia Goons

1 Serial Killer, 1-shot daykill

(immune to kills OR investigation)

1 Cop
1 Doc
13 Townies

• Day start
• Serial Killer chooses their immunity upon confirming

That's the best I can come up with that stays with very classical roles.

I kinda like goons being afraid to lynch the SK.

[popsofctown]

The effect on town behavior is nice.

The effect on scum behavior is interesting too. Scum are deterred from lynching or killing the SK, because they will be punished with follow-the-cop.

I dislike the difficulty for town to find scum when scum are SK hunting and it looks like scumhunting. So making scum unable to do so sometimes makes me happy.

Mafia might need an additional goon to compensate. But finding the SK and not shooting it is a skill. And there's always odds the doc is just randomly shot.

[popsofctown]

As I said earlier, Macho Cop works fine. I just thought that might run against the spirit of the setup.

[popsofctown]

I know.





Secretly I just like daykills.

[popsofctown]

A confirmed inno is more useful on a latter day than on day 1, better PoEing, gaining information that players shaded a slot that has been revealed to be IC, so to me it seems like an open question whether FV should claim day 1. I'd lean towards that you probably should I just don't think it's truly a forgone conclusion. After it claims it does nothing but be D1 IC because it can always target one player who said they were going to act anyway.

This setup is kinda similar to my Matsuda Mafia, and some of my drafts for that involved giving mafia members the same investigative role so that the mafia can disobey the procedures of the breaking strategy. That solution is an uglier fit here since using additional motion detectors creates implicative motion but there's maybe ways to do it, like saying the mafia who performed the kill can multiact a motion detect or saying one of the mafias is an Ascetic Motion Detector.

The leashing strategy doesn't investigate the entire town so you can probably tweak the numbers to get the desired 40ish percent town EV on a leashed playthrough of the setup but most players on site don't seem to want to play setups that should be leashed.

The "list of who acted" thing is pretty cool but gets excised by the leashing, the town should have an agreed upon list of who will act from the previous day.

[popsofctown]

Matsuda Mafia

[popsofctown]

Really douses the informed minority-uninformed majority feel when the mafia never appear as less than 33% of any particular discourse.

pedit: Seeing Double is probably more playable for that reason.

[popsofctown]

You have an unsavory situation when the first three lynches are mislynches and the Dark Angel was not found. That gamestate is 2 Mafia, 1 Dark Angel, 3 VT. Pro-town players don't control enough votes to actually push a lynch. Correct play for the antitown parties should be, both mafia claim mafia, Dark Angel claims, then the mafia and the dark angel vote either mafia member to be elected noble + a no lynch. The following night, the mafia get to decide whether to shoot a VT and immediately win alongside the dark angel, or be jerkwads and secret elect the Dark Angel so that the Dark Angel loses but the mafia still wins.
That's a pretty undesirable flavor of kingmaker.

[popsofctown]

If the town are also jerks they might refuse to vote the Dark Angel for the lynch. It's not like voting the dark angel for the lynch furthers the town wincon. Town and mafia could both say, "obviously scum won this game, but we are all annoyed by the dark angel, so let's no lynch then you scumzors make the Dark Angel lose before beating us".

3v2v1 should definitely be mod-resolved "Mafia and Dark Angel win, it is over", but I'm wondering if there's an elegant way to rewrite the mechanics so that outcome is inherent and natural.

[popsofctown]

Ah.

I think I misread "there is no Noble vote during LyLo" to mean, "the noble does not control an additional vote during LyLo". But what you meant was, "players do not vote to elect a Noble during LyLo".

[popsofctown]

That raises the question of whether there are additional vote in LyLo, are you going to let mafia force wins in gamestates that would otherwise be LyLo using the noble vote and hidden noble vote?

[popsofctown]

I programmed this thing up.
The first time I messed it up by not calculating how the mafia is twice as good at finding the Dark Angel as the town was, that version had to best 3rd part winrate ever. Alas it cannot be so.
The winrates are not that bad though. Assuming I got everything right this last time which might be a shaky assumption, the EVs I got are:

Mafia: 56.8% Town 43.2% Angel 41.8%.
Usual third party winrates are way more abysmal than that.

These assume compulsive lynching (once I account for the mafia's better Angel hunting tweaks seem to show compulsive lynching is optimal) and no preference for lynching players who are confirmed non-angel versus possible angels.
43% winrate is where sadistic people like me want to see the town winrate but TL might want to prop that up somehow. It's a "bad" 43% because the doublevoting should be antitown on balance and the EV can't account for it. Hidden nobling someone who has terrible reads is way of mislynching town without losing any towncred. I would run this anyway. Just roll vt onto 9 players with awesome perfect reads and be awesome and perfect. EZ.

[popsofctown]

If I increment the VTs, I get Mafia: 54.54% Town: 45.46% and Angel 42.5%

That's probably preferable, I expected that to tank the Angel winrate for the angel to need to survive longer, but it must provide a needle in a haystack benefit

or it's all bugged

[popsofctown]

I forgot to add a chance that the town lynches someone the mafia knows can't be an angel, but the town think is possibly an angel.
That should nudge the 3rd party EV up a little bit actually.

[popsofctown]

That made it M: 54.1%, T: 45.87%, A: 42.95%. Very marginal change (That's still with the 10th townie in).

[popsofctown]

It's impossible for the Lone Wolf since either Lone Wolf + Wolf claims a wolf victory and Lone Wolf loss immediately, or Lone Wolf + townie can't cause further player deaths, or Lone Wolf + Mafia can't cause further player deaths.
Oh, I guess the 1-shot kill could be used to crosskill exactly on the last night of the game, but that's not enough EV.

The mafia balance in original Jungle Republic is probably close to about right. The idea that you could gift them a single one-shot nightkill and a vanilla cop and punish them with removing one of their players AND adding two townies AND adding a cop and have something balanced after you're done is absurd, and could easily be sanity checked by the Nightless Expectation rule.

What is the goal with this? The contrast between the mafia having no nightkill and the werewolves having one is the centerpiece of the setup, so redesigning it with that element softened seems hard to square with: "I did like the original Jungle Republic".

[popsofctown]

That doesn't mean it's not a social deduction game it's just not a mafia game.

[popsofctown]

I think this setup's fun-ness could be improved by more closely matching the mathematically special part of Monty Hall and skipping the part where the metaphor of how Monty Hall is usually explained suggests doing a certain thing.

The metaphor is usually presented as "Alice picks door 3, hopeful it has a million dollars instead of a goat, but the gameshow host interrupts her to provide her some information." The metaphor is actually mathematically identical, though, if instead it is "Alice picks door 3 and promises never to pick that door for the rest of the game", or "Alice says door 3 is the ugliest door" or much more to the point, "Alice selects door 3 the be the only member of 'the set of doors that are not eligible to participate in the procedure where Monty identifies a door with a goat and opens it'".

The player with the most votes, pops ofc town who of course has zero votes, and the moderator Jingle are all three identical in that they will not be part of the procedure where a slot is selected to be revealed to be town. Requiring "the wagon with the most votes" to exist and be part of the precondition for the IC effect to trigger mimics Alice's exciting effort to lunge for a million dollars off the bat in the colloquial explanation of Monty Hall without actually mimicking the mathematical properties of Monty Hall. Those properties are already mimicked by having town select the group of two players eligible for ICing (or 3 to 4 if they're badz), everyone who's not eligible for ICing is equally ineligible.

Role Call I demonstrated getting people to agree on leashed stuff can be painful, slow, and unfun, so wherever you can achieve your setup goals with less leashed stuff surely you want that. This setup should be strictly better if it is just changed to : "a random player who is being voted is revealed to be town if there are two or more wagons".

[popsofctown]

Yeah I agree 2v7 is townsided and 3v6 is scumsided and I don't know where you go with that.

[popsofctown]

Is it flipless?

[popsofctown]

The balance is hard for me to brain.

It's maybe worth pointing out that this setup has a bizarre possible anomaly: it's possible for a Mafiosi to decide that it's the best play to lolhammer a partner to go off for help without any plan whatsoever of convincing the town it was a town lolhammer. And then you have an evil child hanging around the setup which is weird. Since it's a flag setup, completely eliminating the lolhammerer's chance of ever getting sent for help can still be defensible play. And since there's no lynching, the lolhammerer isn't removed from the game like they would be in standard mafia, they're just like, hanging around, and depending on moderator style they could possibly be required to post "game-related content" when all they want to do is confirm their slot scum and not give off associatives to the other partner. (is posting "My role PM says I'm scum" every 48 hours activity? In some sense isn't correctly and accurately stating a member of the scumteam the most meaningful activity you could possibly bring to the thread..?)

[popsofctown]

My brain can't wrap my head around combinatorics the way it used to, but if you -are- right, 140/332 (42%) town winrate is where you want to be. The community has some divide on whether target town EV should be 50% or 40% in setups with a night phase, but nightlessness definitely shifts favor towards town.

[popsofctown]

What about allowing suicide instead of lynches?

If someone prefers to live as guilty child and thinks they can get +EV from wifom I'm inclined to allow it

[popsofctown]

I guess the guilty child can do hammers

[popsofctown]

If the Mafia Reporters aren't multitasking I think you have all players use unprovable randomness to decide whether to fire their shot and fire 40, 50, or 60% of the time, not sure the value, and assign specific players that they need to report on certain other players. The Mafia Reporter that performs the kill becomes responsible for ascertaining whether a player acted but not having their hands free to do so, requiring them to kill that player which the town should consider heavily in NKA, or make up a result which has huge double digit chances of suiciding. Order shouldn't matter if you first go around the room with "I'm not saying I acted, but if I did, I received result X", then going around the room with "I acted/didn't act".

[popsofctown]

Does the "Bomb" in Thaumiel mean the same thing as Super Saint

[popsofctown]

9/10*6/8*5/8 = 42%

Is what I'm getting for cult EV, so 58% town EV. These numbers assume that town leashes the hammerer so it functions similar to but not quite as a lynch, so they always get a lynch no matter what, either stringing up someone who isn't lynchproof or bombkilling someone they demanded perform the hammer.
9/10 is the day 1 lynch, 6/8 is the day 2 lynch, and the 5/8 is a modified 1/2. At that point it seems like the ability to leash the supersaint mechanic becomes more partial, because people can refuse to perform the hammer, and usually the remedy to that is to lynch them instead, and if you lynch Thaumiel instead, you lose. Any of the other days, if you lynch Thaumiel instead, you still prove the original desired target was scum and still win, but the last day is LyLo so you can't clean up later.

It occurs to me the cult winrate should be just a little lower since it's possible Thaumiel's identity is known by that point. Not very likely though. Maybe 40% EV.

Cults historically have been able to win a whole lot, so I think 40% winrate for cults is fine. It's incredibly powerful in a social deduction game to be able to draft your team the way cults get to do. I would run this as-is, then if cults really seem to be losing over and over again I'd upgrade Thaumiel from Super Saint to Vengeful.

The idea of cults loverizing every member seems really clever and like a very appropriate counterbalance to how powerful they are. It'd be cool if in large closed themes loverizing cults were used, in that context potentially with the option of holstering the recruit action.

Have you thought about doing this as 10 VT, 1 Thaumiel,

Night start

? It doesn't seem to dangerous to give the cult that additional edge if their EV is 40% right now. And it seems really cute for 1 Thaumiel to be randed and every other winning member to be a player she chose.

[popsofctown]

Procrastination

1 Scum Mod-confirmed-role Beloved Princess
1 Scum 1-shot Watcher

1 Town Mod-confirmed-role Beloved Princess
2 Multi-acting Loyal Loud Friendly Neighbor Fruit Vendor
4 Vanilla Townies

What do you think?

[popsofctown]

You could make the traitors lovers to eachother (that don't know the identity of their lover) to increase the speed of the game and avoid adding players?
This might fill slow as a forum mafia game, many people aren't in love with the idea of rolling traitor.

[popsofctown]

Procrastination

1 Scum Mod-confirmed-role Beloved Princess
1 Scum 1-shot Watcher

1 Town Mod-confirmed-role Beloved Princess
2 Multi-acting Loyal Loud Friendly Neighbor Fruit Vendor
4 Vanilla Townies

What do you think?

Loud is redundant with FN. Night one, both FN's target both BP and one of them gets results, so scum is going to kill BPs early.

Lynch town D1, scum kills BP. Scum Kills not BP. Lynch Scum BP. Scum Kills 2x not BP.

Not taking into account the potential for shooting a FN, That's D3 MYLO. Watcher has a good chance of hitting one of if not both BPs. It's a 1/5 town win day one, functionally, and a MYLO with 0-2 conftown after that. Approximately a 47% town win if one of the FN's dies, otherwise a 60% town win. Doesn't seem untenable, but it's pretty punishing for town. It's an equilibrium on whether the optimal strats should be followed, and if they are it's functionally the same as a massclaim D1. Playable, but probably slightly scumsided.

I don't know why I was dumb enough to not realize loud is redundant on FN.

The rest of this analysis is largely this big pivot on my misunderstanding of the Beloved Princess Role: I thought Beloved Princess only skips a dayphase if

lynched

, in the same genre as vengeful and super saint, but after seeing your post and checking the wiki I now see that it dies upon a death of any kind (which is, a way worse role? Is she supposed to townspew to avoid getting mislynched or scumspew to avoid getting NKed??)

Does "skip the next day if this player is

lynched

" have a name? If it doesn't I guess I'd have to make up one. I believe the setup as written is a dumpster fire (no representation is made that any of my future designs are not also dumpster fires!)

[popsofctown]

We still do Marathons, last Marathon weekend was like last labor day.

[popsofctown]

I know Evenstar apparently siteflaked/siteleft after posting that setup, but it looks really dope and more interesting than the other cult setups people run, someone should run that the next time they run a cult game (if I foisted my preferences on everyone there wouldn't be any cult games at all but to the extent that there are going to be some whether I hate them or not it's a good design).

One recent cult game I watched had a watered down recruit action, one had a cap on the size of the cult. The Thaumiel setup has neither restriction and since converting player's alignments to cult and being a large faction generally are the two most unique things about cult I think it maximizes the cultiness of cult.

[popsofctown]

I guess it's probably balanced but it looks like a dethy-headache to play so that's why I didn't feel in a rush to comment.

[popsofctown]

No strings attached is super imbalanced, in a way that makes it possible to challenge if it qualifies as mafia. Manipulators differ from puppetmasters only in whether it's ok for some vanilla townies to be alive when they claim victory (Puppetmasters: that's ok. Manipulators: that's not ok.) and how they interact with the game's parity ruling (Mafia win at a parity of puppetmasters and manipulators, puppetmasters do not.) The parity rule is meant to feel unified with the rest of the game rather differentiate two factions only once the game reaches a distant endpoint.

The most popular classic definition of mafia/a mafia variant is "a social deduction game where the uninformed minority roots out an informed minority" and since the town puppetmasters have a symmetric amount of partnership info it's not quite mafia. In 2 Mafia, 2 Masons, 5 VT, the idea that the mafia will probably usually fracture the masonry with the factional NK is what makes the setup remain mafia, since this expectation is symmetric here and the vig shot can also work as a doc shot that can no longer be relied upon. (Some designers can/should?/could/would argue that the possibility the nightkills are super unlucky for the scum in such a setup is a problem even so. Not the social deduction game of 2 Mafia, 2 Mason, 1 VT is rather unmafia, both groups claim to be the Mason pairing and express willingness to vote the other mason pairing, and the game hinges 50/50 on the choice the VT makes between voting those two groups, but traditional scumhunting doesn't work, both players are refusing to vote a partner and not "trying to sort" the other group. Social deduction from the mafia leaking schaudenfreude that their victory would not be shared with the VT is the only social dynamic to deduce, which is a valid social deduction dynamic but not the one featured in the core concept of mafia.)

[popsofctown]

Trust No One is very much a mafia game. The town is informed, but the mafia are Even More Informed. Like Animal Farm where all the animals are equal but some animals are more equal than others, you know? The concern going into it is that is 6:3 Town:Scum and WHEW. You can make it work but man at a base level that tends to be strong for the scums WHEW.
Balance is like at least a little overrated though, that isn't actually even my number one concern with it! My number one concern with it is that it incentivizes bussing. General open setup theorycraft consensus is that mafia is fun when bussing is incentivized about as much as mountainous and if you make changes that shift that balance you prefer to go the direction of making bussing more costly rather than less. Since your 3-3 rule (that is clearly meant to address the overall out-the-starting-gate concern that 6:3 is WHEW) can't trigger if scum make a bus the setup definitely incentivizes bussing more than mountainous. No bueno, it is more interesting when scum get to ride the waves of their own bus/bloc meta etc.

I would use an equally brutal detriment to the one you have and let it trigger whether scum bussed or not. 3:6 is STRONG. Viggy doc masonry is very strong. But 3:6 is STARONG. You should be ok with something less conditional, like, "scum picks a player pregame, that player dies immediately following the third death of the game" or have that player picked on the fly, or have the same special wow the Puppetmasters dinged level 6 and are ulting thing special night immediately after the third death. There's other things you can do, but I think it is preferred if they impact all game branches instead of calling one possibility branch out specifically. That was in some ways the criticism that was leveled at Fusion Mafia that we just finished: that it's balanced for computers perhaps but maybe suffers in how some branches of game outcomes totally ignore the power-rebalance-mechanic. (I think this one is scumsided for computers so you have some room to work)

[popsofctown]

I think the ult needs to happen no matter what or almost no matter what, so either Puppetmasters need to be able to self target to protect themselves if they drop to solo, or the rules need to say some random towny gets nominated puppetmaster if parity is reached and both puppeteers are dead.
Having the ult trigger at parity rather than 3 kills is fine. I think it disincentivizes bussing a bit but that's the correct direction to push like I said earlier. Just as a personal preference I generally try to keep the bus incentive close to mountainous.
As an element of clarity it should be clear that the 3 targets the Puppetmasters can select can contain duplicate targets, that seems like it's definitely the point. Is it exactly three targets or up to three?
The outcome for "everybody dies" needs to be defined (I guess technically it's zero to zero parity). When parity is reached for the first time at 1 wolf 1 puppeteer the puppeteer is guessing how many times to hit themselves and how many times to hit the wolf and it can shake out that both the wolf and puppeteer die. I would want to rule that a town win. It's still possible for the wolf to win even if you rule that a town win, it would be a 50/50 on protecting themselves correctly with their action.
Unless it's not supposed to selftarget. Many many details.

[popsofctown]

I'm not very interested in a 9p where the evil faction doesn't know their partners but maybe some other players are.

[popsofctown]

On top of pops' comment, TSE's setup looks solvable with a massclaim. I think the SK/Guardian Angel/Witch trio would have a better chance of being viable in a larger game alongside more VTs and a regular mafia team.

Nah I don't think it has a massclaim issue once you read the weird way it was formatted. The third time I read it I figured out what the setups are supposed to be.
As an example one setup is 1 SK, 1 Witch, 2 Masons, 5 VT. Another one is 2 Serial Killer, 1 Botanist, 1 Roleblocker, 5 Vanilla townies.

SK + Guardian Angel does look cool to slip into a large theme, yeah.

[popsofctown]

Donnie Brasco

3 Mafia Goons

1 Town-Aligned Donnie Brasco

8 Informed Vanilla Townies

• Mafia Goons
  and
  Donnie Brasco
  share the scum private topic and vote for NK via plurality.
  
• Each
  Vanilla Townie
  is informed of the identity of
  Donnie Brasco
  .
  
• Each
  Mafia Goon
  has access to
  Day Kill
  that only works on
  Mafia Goons
  and
  Donnie Brasco
  . The
  Day Kill
  command is privately messaged to the mod during day time.
  
• Donnie Brasco
  maintains a
  Prevent
  command on a Mafia Player that is privately messaged to the mod and can be updated each night.
  
• Mafia
  win when they equal or outnumber the
  Town
  , or if the
  Day Kill
  hits
  Donnie Brasco
  , and
  Donnie Brasco
  did not
  Prevent
  the killing player.
  Town
  wins if all
  Mafia
  are dead.

Innovative. Lots of activity shenanigans here as written, once one goon is down, if one goon is verifiably afk for the day, Donnie Brasco can name the scumteam and prevent the scum player who is active, then start the following day with a prevent on the remaining solo scum.

It's implied but not stated in the setup that the prevent command will stop the daykill from happening (a reasonable interpretation is that Brasco will die anyway, but the win condition will be prevented.)

I don't exactly understand the design goal, there doesn't seem to be a social-deduction way to determine which goon is likely to submit the daykill, so it seems similar to "roll a d3 as a saving throw against the instant win".

[popsofctown]

Can mafia break it by figuring out which of those 4 the town are unwilling to lynch?

You can "break" 9p mountainous by figuring out which two players are unwilling to lynch eachother, because they are mafia . "Break" isn't the right word here. It is clearly intended that this Assassins in the Palace dynamic is going to overwrite some of traditional mafia and if you don't want that dynamic you wouldn't /in, but "break" isn't the right word.
Mafia can't perform the daykill in their first post. Donnie Brasco is in the scum PT, he knows who all the mafia are (presumably he is in the scum PT on a secret alt because otherwise the setup makes no sense and is absurd, scum read the PT access list and submit a daykill and win). I guess it's possible the intention is that all the mafia are also on secret alts but that seems like a much more scumsided setup. But assuming the mafia aren't on secret alts, if the yolo daykill is a miss, the daykill is expended and Donnie Brasco just names the scumteam in the game thread and the game is over.

Lynching Donnie Brasco to reduce the risk of losing by daykill is a play with value. In theory, though, the mafia can detect that is what's happening and submit a daykill on Donnie Brasco before he is fully lynched.
In the event they don't do that, in the mafia-aren't-on-secret-alts version, it's an instant town win: Donnie Brasco selfhammers and names the scumteam in the same post.

[popsofctown]

D1 put people randomly on L-1.
Donnie Brasco selfhammers and says who is scum.
The game turns into mafia random vigshooting each other.

Each time a player is put to L-1 randomly and does not selfhammer and say who the scum is, it becomes confirmed that they are not Donnie Brasco. So the theoretical EV should be close to even I think, scum have some sort of strategy where they take a lolshot after N players have been confirmed not-Donnie in this way.

[popsofctown]

Aren't all Vanilla Townies informed of who Donnie is? Theoretically the game turns into "can the Town pile on enough votes for Donnie to self-hammer before the Mafia vigshoots him".

That's definitely one of the activity issues I alluded to yes

[popsofctown]

omg Jingle's location changed

Compulsive lynch is probably o.k. for Lone Wolf.

[popsofctown]

It's easier to leave the state when you don't have any warrants, isn't it? Hope your attorney is working hard for you.

[popsofctown]

Oooooooh it's so spicy

[popsofctown]

The informed townies can feign interest in a Donnie Brasco lynch to counteract that. If not very many townies are informed, it becomes hard to keep the uninformed townies from hammering.

[popsofctown]

If town votes town in 5p LyLo, scum is supposed to win. That doesn't seem like a bug. I would avoid the awkward voting rule and add as many VTs as it takes for it to be a balanced game even if Donnie has to behave like a goon in LyLo sometimes.

[popsofctown]

I'm having trouble parsing what the setup is.

[popsofctown]

Yes it's easier to follow now. Clarifying that it's not a 26 player setup helps (you listed that there are 10 citizens, and 8+1+2 roles, and since that's 11 it makes it seem like the roles aren't citizens). "players" seems like a troublesome term for non-citizens because the citizens still have to queue up for the game.

It's unclear whether Citizens are voteless. If they are, their engagement might drop to the point they are difficult to read.

This seems incredibly townsided right on its face? 7:2 JK would be a newbie setup the is considered to have a weak town but not massively so, and you're stacking on top of that an additional townie, 3 additional "town hydra heads", and a public cop shot every time a player is lynched.
If the Citizens can't vote, then every single lynch is decided with 60% town control of the lynch, which isn't quantifiable town power but is an incredibly scumsiding dayplay dynamic that is likely able to overcome even these advantages. It's sort of untested territory, it is difficult to design a setup that stays in perpetual LyLo. My concern for voteless citizens is engagement, which isn't a balance concern. If I was designing the setup I would move the power from the JK into some kind of nominate a PR gimmick that only the citizens vote on just so the citizens have something to do.

[popsofctown]

Then how is a citizen a jailkeeper?? It has to aim?

[popsofctown]

Yeah, it's a treestumping setup. Which makes me less pumped.

[popsofctown]

You can ruin the conceit of the setup once you realize as a guard you can say "I'm a guard that knows among {Player 1 -" without finishing your sentence. Day one, a guard votes one of the two cantidates, all guards follow suit, the King and Assassin follow suit so they seem like guards. Even in the worst case scenario where you lynch the king first: a guard became a cultist, the Assassin gets one very difficult shot at hitting the Cult Leader (1/8, they can't rule out the follow on voters who want to seem like guards or are, but they can rule out the Guard who cast the first vote). Next, the Cult Leader is lynched day 2 and then it's not highly relevant whether you remove the Assassin for being a kingmaker or not, winning a nightless game with 8 guards against 1 cultist is pretty easy.

I'm not sure what the best fix is. You could delay flips until morning and give the Assassin an action to bring the last victim to life and kill them over again and kill the assassin when they guess this wrong, but that's much more all-in than I would

prefer

to make the Assassin play, I guess?

[popsofctown]

Yeah in standard parlance voting for a lynch is not considered "killing" someone, for example the way some people draft psychologist or detective role PM would ignore lynch participation.

[popsofctown]

Wait do you not even need to lynch the cult leader to win? I'm just now realizing that is maybe the intent here. You win by lynching all the cultists at any point after you wait for the first cultist to come to exist? That has to be what you meant if lynching the cult leader wins the game for the Assassin.

You have a sample size of 2 of people reading the setup and getting something different from it. It's not helpful that that win condition is -technically- fulfilled at the very start of the game when the first cultist doesn't exist yet.

[popsofctown]

I'm thinking up a fun different metaphor for this game

[popsofctown]

Deathsticks Daryl

1

Deathsticks Dealer Daryl

1

Coughsticks Kingpin

1

De facto Decision Making Drug Regulator

8

Paper Pushing Drug Regulators

Lynches are flavored arrests, players with honest flavor are arrested for unpaid speeding tickets.

Deathsticks Daryl You're Deathsticks Daryl, a chemistry hobbyist who invented a drug in his garage and continually networks to try to find a payoff for it. You start the game vengeful . Each night, you may target a player to tip reporters on an expose. An expose causes a Coughsticks Kingpin , De facto Decision Making Drug Regulator , or a Exgovernment Drug Dealer to subsequently get arrested and leave the game. An expose on a Paper Pushing Drug Regulator will have no effect and furthermore will make you lose vengeful as the newspaper is annoyed enough at you not to cover your inevitable arrest. You win the game and immediately make all other players lose the game when you are still alive and either Coughsticks Kingpin or De facto Decision Making Drug Regulator are out of the game, allowing you to flood the clean market or black market with your Deathsticks.

Coughsticks Kingpin Each night, you may target a player, and if they are a Paper Pushing Drug Regulator they will realize how much more they could be making per year and are recruited to become an Exgovernment Drug Dealer . They will join a PT with you where you can discuss how best to sell Coughsticks in alleyways. You win the game when there as many players in your PT as there are Regulators .

Paper Pushing Drug Regulator You are informed that player X and player Y are Coughsticks Kingpin and De facto Decision Making Drug Regulator , because you saw them at the high stakes tables at the casino, but you're not sure which is which. You win the game when you've arrested at least one Exgovernment Drug Dealer , proving your agency is effective, and you've also arrested every Exgovernment Drug Dealer , so that your agency is not corrupt.

Exgovernment Drug Dealer You've been recruited by player Q, Coughsticks Kingpin , and still remember the opulence of player P, De facto Decision Making Drug Regulator . You have access to a PT with Q and all other Exgovernment Drug Dealers . You win the game when there as many players in your PT as there are Regulators .

De facto Decision Making Drug Regulator You're the one with the real pull and agency power to decide whether Deathsticks belong on the clean market. You use that same power to embezzle your agency's funds a little bit. Each night, you can target another player for a tax audit, if it's either of the other two roles with a night action they become very busy concealing money laundering and are roleblocked . If it is a Paper Pushing Drug Regulator they will receive financial relief when the tax agency actually owes the impoverished bureaucrat money and that will render the Regulator unrecruitable for the night. You win the game when you've arrested at least one Exgovernment Drug Dealer , proving your agency is effective, and you've also arrested every Exgovernment Drug Dealer , so that your agency is not corrupt.

[popsofctown]

I realized during writing this up that denying the cult leader PT access but still letting recruits talk to eachother can keep the cult leader in the dark about who the king is, making it easier for the assassin to continue searching for the cult leader. I think that seems preferable?

[popsofctown]

AitP hasn't exactly taken the world by storm. Maybe it'd help for the assassin's win not to be exclusive

[popsofctown]

You didn't state in your version fact that Daryls win will end this game, which is not obvious for 3rd fraction. His win has to be exclusive.

You are correct, I edited the post. It was intended to be an exact reflavor of your first draft and not reflect any suggestions for changes. I feel like Jingle might have smarter ideas about maximizing balance and playability compared to me on this one.

[popsofctown]

This seems like a good overall design but needs a slight tweak of the knobs to not be townsided. 50:50 EV for nightless happens at 9:3, this setup gives town an additional townie and an additional clear over that.
Most of the powers can be counted as not being a balance concern, but the dead PT is purely town motivated because the dead PT can make sure not to vote in dead scum's interests and they get the equivalent of dayvigging to decide the game in LyLo where LyLo is usually a third or two fifths controlled by the mafia.
I think this would be good with an innocuous neighborhood added to bump it to 4:11 since 4 player scumteams historically do crazy things and that dead PT ability is similarly crazy.

[popsofctown]

Yeah, if you want to keep it at 10:3 I think you would want to empower the mafia somehow, either by allowing some nightkills (not every single night, I think that would scumside it), or some sort of black flag condition (which would be an attractive option due to powerful dead thread ability becoming more and more powerful towards lategame) or some other way.
I would not say "Open games are usually 13 peeps". Open micros are very very common, and open games go beyond thirteen players when that's the best fit for the setup concept, like with FakeGod's dance mafia. But frequently small opens see some very high demand.
"Open games usually have fewer than fourteen players" is definitely true and is a strong argument against making this an 11:4, the Venn diagram for players interested in open setups and players interested in larger games doesn't look great, the dances were more of an exception to the rule.

[popsofctown]

I am trying to make a new setup which I am putting in the undertale opens setup thread but I want some feedback on it if I can please get any: viewtopic.php?p=11661807#p11661807

[popsofctown]

The flu is a type of virus, you could improve the flavor there.
Also this game is ten times more fascinating with the same EV balance if the Virus's target becomes the new "Virus" and it's the original one that dies.

[popsofctown]

That setup looks really favorable to town if you're telling Commuters what kind of Commuter they are. The day one massclaim doesn't degenerate the game to a single point with 5 duplicates of each, but it degenerates it a lot, and people don't usually want to play games that are day1 massclaims.

[popsofctown]

in the context of an open there's not really a downside to making a commuter compulsive and even or odd and refusing to tell the commuter what kind of commuter it is. In a closed "i'm gonna stop commuting so you can check me" is the fanciness but that's not possible here so you're just removing an element people usually don't like that much.

[popsofctown]

you would get this cute dynamic where the virus might guess a player's evenness or oddness based on where they expected the flu to shoot after a shot fails even though the victim doesn't know their own evenness or oddness

[popsofctown]

viewtopic.php?p=11702993#p11702993

[popsofctown]

viewtopic.php?p=11742110#p11742110