The New Set-Up Review Thread

[GuyInFreezer]

Looking for few reviewers to look over my 13p mini theme setup for balance. The setup contains a special mechanic. The flavor knowledge is not required since I'm not planning to out the theme now.
Please PM me if you're interested!

Bump :<

[GuyInFreezer]

Looking for few reviewers to look over my 13p mini theme setup for balance. The setup contains a special mechanic. The flavor knowledge is not required since I'm not planning to out the theme now.
Please PM me if you're interested!

Bump :<

All filled. Thanks!

[Ryansee]

Hey all , I have created my first set-up for a game and I am looking for some reviewers to review the setup and tell me what they think. I am a new mod.

[callforjudgement]

Hey all , I have created my first set-up for a game and I am looking for some reviewers to review the setup and tell me what they think. I am a new mod.

If you want to run it here on mafiascum.net, it'll need to be a Mini Normal game, and so you'll be allocated a review panel when you join the queue, so there's not much point in having it reviewed separately.

If this is just for offsite and you want someone from here to take a look at it, this is the right place, but you should give some details (number of players, approximate complexity of the setup, etc.) in order to get appropriate reviewers. (See GuyInFreezer's post for a good example.)

[Ryansee]

Thank you for your answer. Its for another web but truth be told. I found this website randomly while I was working on mafia setup that I had planned to just take a week of work and post it right away to them.. I thought I was done and just needed to look at some stuff but then, I ran into the wiki and the stuff i read made me feel that I just didnt it the proper way it should have been done. I basically told them I wont deliever it until I am sure its good enough. Its not a complicated setup. It is based on this system which will be given to all players. The roles are mostly all unique. I would really love for people to talk to be about the setup. I am willing to redo it from A to Z as long as i will have a good finished product that I can work as much as needed on. It has a story aswell which I will give the reviewers with the roles list.

The amount of players are around 15(+-2)

Setup Info (to be posted in the setup thread) (based on 15p example with 4 scum and 1 serial killer)
There will be 4 villagers and 11 power roles in the game: 8 are open and 3 are secret
6 open roles are given to town, 2 are given to scum
Scum gets 2 secret roles, the serial killer gets the last one.

Open roles are divided this way:
2 killing roles: Roles that let you kill
2 lynching roles: Roles that let you mess with votes or lynching
2 investigative roles: Roles that give you information about other players
2 protective roles: Roles that let you protect other players (or yourself) from kills

No roles will overlap with other categories, a role will fit strictly into one and only one category.
The actual roles will not be known ahead of time.

The secret roles can be anything and everything our twisted minds are able to imagine. So if someone claims the ability to blow up the game and end it in a tie, he’s probably scum.

[Porochaz]

Some of us still prefer to do it the old-fashioned way.

Mhm.

[callforjudgement]

The amount of players are around 15(+-2)

The actual roles will not be known ahead of time.

Right. Because some of information is secret, this is probably best taken to PM (in case one of your players stumbles across this thread in a search engine; it isn't totally unheard of). I would help, but I'm not particularly good at balancing setups larger than 13 players (I nearly always do Mini or Micro reviews). Perhaps someone else will help out. If nobody else does within a couple of days, PM me and I'll take a look at it then, although I can't guarantee I'll get a mid-sized setup like that balanced.

[Elscouta]

Setup Info (to be posted in the setup thread) (based on 15p example with 4 scum and 1 serial killer)
There will be 4 villagers and 11 power roles in the game: 8 are open and 3 are secret
6 open roles are given to town, 2 are given to scum
Scum gets 2 secret roles, the serial killer gets the last one.

Open roles are divided this way:
2 killing roles: Roles that let you kill
2 lynching roles: Roles that let you mess with votes or lynching
2 investigative roles: Roles that give you information about other players
2 protective roles: Roles that let you protect other players (or yourself) from kills

Problem with this setup is that giving killing roles to scum can be extremely unbalancing. A Town Vig is a balanced PR (albeit swingy) but a Scum Vig would be totally insane. Any game that would be balanced if the scum gets the two protective / lynching role would be totally scum sided if the scum gets the two killing roles instead.

[Ryansee]

The roles arent given out randomly. I will have a proper list of all the unique roles in the game and which side has them. Although I didnt know a scum vigi in unfair? even if i just give him an extra Night kill once the entire game? , I want to talk to you more about this

[Elscouta]

1-shot Mafia Vig is not totally unbalanced, but it's still very very strong.

It might depend on your local meta. The element of surprise of such a role makes for a significant part of its power. If it is not uncommon in your play circles to have scum being able to double kill, it's a lot more reasonable.

[Ryansee]

Can you explain more? I would love to talk to you more about the setup if you have the time ?

[Ser Arthur Dayne]

Something like a 1-shot Mafia Day Vig that has to announce the day vig in the thread would be a bit more balanced.

[Ser Arthur Dayne]

Also if you're interested in roles that aren't Vig but also "killing" you should have a look at http://wiki.mafiascum.net/index.php?tit ... ling_roles

[Ryansee]

The roles arent standard , none of them are at least. I will give you the roles in a pm if your up for reviewing and the entire setup details.

[Ryansee]

I guess no one is interested

[Ryansee]

I guess no one is interested

[inHimshallibe]

I need a clean slate of reviewers for a Large Theme (no, not WoW, sorry reck). Theme is the Renaissance.

[Guyett]

I'm putting together a game with football manager flavour. Its 9 player role madness game. Anyone want to have a look to see if its ok?
Knowledge of the flavour would be appreciated

[neil1113]

Anyone have a quick minute to look over the balance of a 15 man game? Just some quick input is really all I need... pm me real quick. Thanks!

[LeMidget]

Setting up a large (17p) game off-site. No vanilla roles. No super complex roles either, though. Would love a review. Thanks in advance.

[Who]

I've got an idea for a game and would like advice on it.

The setup as I think it will be right now:
1 Polygamist Cult Leader with Daytalk (If they or anyone they recruit dies, the town wins, they are free to recruit one person at night but can always not recruit if they want. I'm trying to make it so that not recruiting is an option they should seriously consider)
8 VTs
Open
Daystart

I need advice on:
The size (Went with 9 because Micro, not sure if it would be better with less/more)
Whether or not I should add a few PRs (And, if I do, should I make the setup closed or leave it open?)

[callforjudgement]

Let's see. The only balance implications here are how often and when scum recruit.

Lylo comes earlier than normal; if there's

n

scum,

n

+3 townies, then if one townie is lynched and one is recruited overnight, we have

n

+1 town,

n

+1 scum, = scum win. So when comparing this to other setups, we have to take the early lylo into account. This is a pretty major balance consideration, on the order of giving scum a daykill (which has a similar effect on lylo timing). There's also the usual cult benefits (you get to recruit the good players). The Polygamist drawback is pretty noticeable, though.

Luckily, there are only a few possible scum strategies, all of which can be reviewed separately (the strategy descriptions assume that scum will always recruit if they win the game as a result, regardless of what the strategy would dictate):

• Never recruit:
  This is basically 8:1 Nightless with town limited to 4 mislynches. Slightly worrying as a strategy, in that that has a EV of 44% for town, which is much lower than it should be for a scum strategy we want to be suboptimal. I fear that best scum play here is to simply not recruit until it wins the game, which would lead to a very boring game.
  
• Recruit once, on N1:
  This is reminiscent of Black Flag Nightless after one correct lynch (which is even a 7:2 setup, so the numbers match exactly). That setup is townsided (because Black Flag Nightless is theoretically balanced, and in practice seems reasonably balanced, maybe a little townsided because scum can't eliminate strong townies). However, in that setup, town can go down to 4 members and still remain with a chance of winning; in this setup, they have one mislynch less (two mislynches total), which is pretty much the only balance-affecting difference (there's a marginal difference if the D1 lynch is the player that scum were planning to recruit N1, forcing them to change their mind, but how often does that happen?). I was initially inclined to say that with the missing mislynch, this becomes quite strongly scumsided. However, I think I was wrong; the EV is 52% to town, meaning that either town can win over half the time via scumhunting, or if they feel that they can't, just randomize and still win that way.
  
  It's worth trying to quantify the ability for scum to control the lynch, too. There's no real data for D1 of a 2:6 (a cult setup is like having all days as D1s because you don't know who had their alignment changed), but we have plenty of data for 2:5 (Original Newbie) and 2:7 (more recent Newbie setups) For day 1 of a 2:7, town's chance of lynching finding one of 2 scum ranged from 20% to 27% depending on the setup (the theoretical value is 22%); for a 2:5 (Original Newbie), it's around 23% (disregarding day 1 no-lynch) compared to the theoretical 28% (and noting that scum are very likely to claim a PR if run up for lynch that day in that setup, which skews the statistics).
  
  So I think there's no reason to worry about this strategy much. It's somewhere between "balanced" and "somewhat townsided", depending on just how much of a impact the cult mechanic makes, and the scum should probably not choose it.
  
• Recruit twice, on N1 and N2
  : This is similar to the above, but scum expose themselves quite strongly on day 3 in exchange for more control over the lynch. The EV is around 62% to town, so this is incredibly risky from the scum point of view, but town end up needing every townie to vote for the same scum at 7p, which is a tall order. I don't have much experience in balancing this setup, although what evidence we do have (mostly from Themes) anecdotally suggests that town doesn't have as much of a problem as it might seem. I have no issues with this tactic existing in the setup; I wouldn't use it as scum, but I can imagine scumteams who might. And I expect them to get horrifically screwed over by it :-)
  
• Recruit once, on N3
  : This has the start from the first strategy, and the endgame from the previous one. Basically, you get another vote to help you out in 5p lylo, in exchange for giving the town another target they can hit to win the game. I doubt it's worth the tradeoff. EV is 55% to town.
  
• Recruit once, on N2
  : This is probably the best strategy for scum, unless they feel (based on D1 play) that they're seriously likely to get lynched D2 instead, in which case they could recruit a player to help defuse the lynch. EV is 44% to town, the same as with the never-recruit strategy; it's better than the previous strategy in pretty much every way, except that scum have less lylo control (at 5:2 rather than 3:2). I think a town might be hard-pressed to win this; when lylo comes on day 3, one of the players has
  only just been recruited
  , thus no associative tells apart from those happening in lylo itself, and PoE can't be used either for the same reason; not to mention that scum can recruit strong scumhunters in order to deprive town of their benefit. This strategy seems rather scumsided, probably unbalancing the setup as a result; 44% to town is viable as a reasonable EV for a "standard" sort of nightless game (where town normally play above their EV), but not in cult setups, which are even worse than mountainous setups in terms of the difference between the EV and the predicted town win rate.

Based on these numbers, 9p does seem to be around the right size, but with optimal strategy, it spends most of its time not being Mafia. If scum play correctly, there probably won't be two scum until lylo, and possibly not even then. Adding VTs doesn't really help; I think scum's optimal play with any number of VTs is to lie low until the setup reaches 6:1 overnight, then move into a 5:2 lylo (because they can recruit to win at 3:3 the next night after a mislynch).

I'm not sure if this issue is at all easy to fix; I can't see a way to do it using town power, for instance, and if you limit the ability to recruit close to lylo, or start with multiple scum, there are few reasonable choices for the scum at all, defeating the point of the setup. One crazy idea I considered is to increase the number of VTs, but to give the cult a kill they can use instead of their recruit (making kill-or-recruit compulsive, so that the town always know how many scum there are), but I can't come up with a reasonable argument that that helps either. At least this setup is very easy to calculate EVs for (that's usually the case when all scum are lovers with each other), which makes potential ideas easy to test, so if anyone has ideas for fixing the setup, let this thread know and I'll take a look.

[JasonWazza]

Make a anti-cult doctor would be my suggestion (can block the recruit essentially) this gets rid of most of the problems with the setup in my opinion (can't bring LYLO a day ahead as easily, can't just recruit the towniest of the town.)

The only draw back is that the town have no clue how many cult there is EVER.

[Bulbazak]

Run this as a marathon and see what happens.

[callforjudgement]

Make a anti-cult doctor would be my suggestion (can block the recruit essentially) this gets rid of most of the problems with the setup in my opinion (can't bring LYLO a day ahead as easily, can't just recruit the towniest of the town.)

That role's called Alarmist, and I guess it would help (scum would be more inclined to recruit earlier than later if they knew their recruitment wouldn't be guaranteed, meaning choices would become more meaningful). However, the problem still remains that scum's best strategy is to recruit on N2 or not at all (with both having the same EV). If scum attempts to recruit on N2, then if it works, fine, if it doesn't work (say you hit an alarmed player), still fine, you just don't recruit and go for the solo win. And you still don't have a groupscum team until D3. The only real change adding an Alarmist does, then, is to increase town win rate slightly due to the chance that scum miss with their recruit when it's 1:3 or 2:4 overnight, a 33% or 25% chance respectively (recruiting successfully wins scum the game in those settings, so they have no reason to do anything but randomize their target, or aim for a claimed Alarmist if there is one).