Limbo Setup Design [Large Theme]

[Jingle]

Ego. I'll have notes tomorrow.

[Jingle]

Currently, it looks too scumsided. The easiest fixes would be to either throw a couple more VT's into the game or to switch one of the goons to town and add one or two 0/negative utility roles to town as well (To drop the power of the Neopolitan).

Lawyer could use clarification. I would switch the wording from "most favorable" to "least incriminating" to head off the argument that most favorable to the lawyer would be a false guilty on scum.

Ninja and tracker are both not in your list of included roles, which you should fix .

Because it's a modular setup, you can just use a list of Possible Roles instead of sample PR's. Format the Role PM's for every role you list that might be in the game and make template addition sections for your modifiers. It's extra work, but it'll raise the quality of your game immensely, imo. It also means you can specify the type of results Lawyer leaves behind (tracker would recieve No Result, for example).

Weak Ninja by MS standards wouldn't function as described. I don't think there's a role that does explicitly what you want (although I DO like the role), but you could have a modifier in your list that would be "This players action fails if [PREDETERMINED EVENT] happens." I'd probably name it Conditional, but that's a personal choice, as long as you make it clear what it does somehow.

[Jingle]

For ease, divorcing from flavor:

Weak Ninja
Vanilla Cop
Goon x2

Tracker
Neapolitan
Elite Bodyguard
Lawyer
VT x6

[Jingle]

X-Shot

[Power]: [Description]. You may only use this power X times over the course of the game.

Conditional

[Power]: [Description]. If [Condition] happens, your action will fail.

etc.

That's exactly the why of the Neapolitan power drop, btw. The power of the Neapolitan is in it's ability to generate innocents. If there are fewer people it can accurately get that result on, it becomes less and less useful. Especially if there are nonsense roles like Weak Townie (If it targets scum it dies, but it can't target anyone), nonconfirmable roles like Informed Town (if their information is something you would've announced anyway or something inane) or outright negative utility roles like hated. I would shy from anything that targets though to prevent unintentionally beefing up the tracker. Maybe a visitor, but nothing confirm-able like Fruit Vendor.

Balance wise, 10v3 with some fairly significant town power is the norm. Something like Cop + JK. Adding a scum player takes the number of required mislynches down from 4 to 3. Adding a townie on the other hand doesn't change the number of mislynches, just whether you're on the MYLO or LYLO track. It ups the power of the EBG and lowers the power of the Ninja, but not enough to throw too much into question.

Power role wise, Tracker is going to get limited guilties, because most scumteams are going to use the Ninja while it's alive. Neopolitan can't get guilties, only innocents. Tracker can't get real innocents either, unless scum is already losing. And the Lawyer is going to function as a free mislynch most of the time despite the fact that it DOES have town utility in this setup. EBG is by far the most powerful role, but it can only give one inno/guilty per game.

The limited pool of fakeclaims hurts scum, but on the other hand the limited pool of roles they can be fighting lets them try to figure out exactly what they might be up against and need to play around and those two things are roughly equal depending on the strengths and weaknesses of the scumteam themselves.

Flavor wise, if you want to use more roles, you could do Acedia or one of the other prior incarnations of the sins/virtues, but honestly I think the modern ones make for a nice theme without feeling too forced.

[Jingle]

AFAICT, the PM's themselves look good to go. There were a couple of grammatical errors that I can hunt down and point out if you care, but nothing that should influence readability. (the one I can remember is an motivator instead of a motivator.) You should probably leave in a sample VT PM (The despair one was good if you still have access to that) just so scum can see the formatting, or at least put it in the scum PT. Randomizing flavor in the role PM's works, but you could also leave it vague ([Flavor Goes Here]) if you want and achieve the same goal. It kind of just depends on what you think looks better.

One thing I just saw is that you have a dead player list outside of spoilers in your OP. You might consider spoiling it just so that people reading the game after it's conclusion can choose whether to be spoiled or read along without spoilers. Not a big deal either way, but I think convention leans towards spoiling it nowadays.

11v3 is worse for mafia by sheer numbers, especially if you have any roles like doc or vig to take the town off of evens. (EBG is basically guaranteed to do so.) What that means is that you should give town less power or scum more power to compensate. 10v4 is a much larger swing though. It's not that mislynches are more likely, although they probably are since there are more scum influencing things, it's that it changes the number of mislynches scum needs to win.

In a standard 10v3 game, after the first mislynch, it's 8v3, after the second it's 6v3 and after the third it's 4v3 LYLO. In a 10v4, that changes to 8v4 then 6v4 meaning they need to lynch one less town player to win. Power helps a lot here, but your game is pretty close to balanced if you drop a VT. Maybe a little townsided, but not horribly broken.

For balance a 10v4 needs some insanely high levels of town power. Something like a guaranteed scum lynch or multiple frontloaded innos. I could see a setup with 2 pairs of masons, a ridiculous number of docs (3 would be my bet), true follow the cop with a little bit extra, a D3 guilty child, a mafia traitor who can't be recruited... etc. There's lots of ways to balance it. It just shifts the focus of the game away from dayplay towards nightplay, and I tend to argue against a lot of the roles that can do that. For example, a guilty child is probably the easiest role to fix the balance with, but for most players they're probably going to see that role PM and just not want to play the game (there are exceptions of course, but hoping you get one of the few people who would want to play guilty child is fairly naive). You could also go the route of limiting the scumkill, but games that mess with core mechanics (scum can only kill every other night, for example) tend to be very difficult to balance by their very nature because they alter the underlying concepts of mafia.

An 11v3 on the other hand just needs to have more power for the scum or less power in the hands of town. In your setup, the easiest ways to do that are A. limit the power of the EBG, B. Lower the power of the investigators, C. Give scum additional power, or D. add negative power to the town. A combination would of course work as well.

A. is pretty straightforward. You could instead of making it an Elite BG make it a 1-shot BP BG. You could give it an X-shot modifier. You could make it indecisive (can't target the same player on two consecutive nights). Anything to lower the chances that he actually hits one of the scum before they find and hit him.

B. is what I suggested. Currently, the tracker is largely a WIFOM investigator with a high chance of utility if he lives to the late game. Your big source of power here is the Neapolitan. Cutting down the number of players he can safely investigate cuts the power by making it more likely to get useless results. Additionally, more weird and useless claims gives scum more room to fakeclaim. If you put a Weak VT into the setup, who's to say you didn't also put in an informed townie who is told that flavor x is role y?

C. is usually the traditional go to, but honestly I like games with goons. You could add a scum rolecop to help find the investigatives. You could add a roleblocker to mess with Neopolitan/tracker results and allow scum to kill past the EBG. But the power light status of the scumteam seems like a feature, not a flaw, of your design.

D. is interesting, but again, I wouldn't rely only on it because most negative utility roles just aren't fun. Miller? Cool, I have to die some time! Hated? Cool, I'm a liability to town! Most of them are just not fun role PMs to get. If it works with the setup, then go for it, but I advise against negative utility unless you have a flavor or setup reason to want a specific one in the game.

Hopefully that helps.

[Jingle]

Currently, the tracker is largely a WIFOM investigator with a high chance of utility if he lives to the late game.

Can you elaborate more on this, specifically the utility aspect? I have a vague idea of the Tracker's usefulness in late game, but clarification's always good!

Tracker basically becomes a full cop if there's only one scum left. Not

quite

that level of power, but close enough for the world of balancing a setup.

As the setup was I would add 2 0 utilities or 1 negative utility. Also, Green Goo could be useless against goons depending on the mod interpretation. It depends on whether you define PT access and nightkill ability as an ability of the player or the faction. Basically it's more about how you word the scum PM than the Goo PM. Between the two, I prefer the Toxic Goo. It has less of a swing factor, although it is still negative utility and very swingy.

Ooh! I like your suggest Indecisive suggestion for the EBG! In theory, it sounds good. In practice, what would be the chances - do you think - that the EBG would target the same person two nights in a row?

If they haven't hit before claims? High. Particularly if the Neapolitan and the Tracker claim at the same time. It's a toss up which is the better kill, and if they guess opposite the scum, they probably then can't stop the scum the next night either. Otherwise, it's still a decent chance, because if the EBG is wrong about a NK, they're still targeting for likely nightkill targets, meaning that whoever they targeted is likely to be a high priority target the next night. On the other hand, it also means they can't accidentally sit their protect on a single particularly towny scum player for the first three nights, so it's not all bad news for them.

I'll run through balance and swing in it's own post, because it's likely to turn into a wall before I'm done.

[Jingle]

Okay, so balance versus swing.

First off, Swing is not a good v bad gradient. There are fun games that are high swing and fun games that are low swing. It's all about preference and game design. As you play, you'll likely find that you like a certain level of swing in your games. As you moderate, you'll probably find that you like a certain level of swing in your games. Those will not necessarily align. Until you have the experience to gauge how much swing there actually is (which is really just reading and playing games), I always advise shooting for a middle of the road approach. On the other hand, there is most certainly good and bad balance.

Balance refers to the chances of various factions to win the game. Generally, the rule of thumb is that every player should have between a 40 and 60% chance of winning the game, with the closer to 50% the better. Obviously, this is more of a science than an art because there are so many subjective elements to mafia. Examples of bad balance would be, in the extremes, setups like 6 VTs versus 5 goons or 2 goons vs 4 cops and 3 VTs. Obviously, no one is going to want to play those games. The exception to this rule is in the cases of games involving Jesters/Cults. Both of those roles are ridiculously hard to balance, but properly advertised Jester and Cult games can be fun for the right people even if they're awful from a game design sense.

In the more complicated cases of multiball/third parties, you have further consideration. There is a bit of a preference issue involved in balance with multiple factions, and there isn't universal agreement on the correct determination. Basically, some people think that town should have a 50% chance and the scum should split the remainder, and some people think that all factions should have an even chance. I personally lean towards the former, but just be aware that when balancing multiball you will have to answer that question in your own mind.

Swing on the other hand, is a measure of how much impact the roles and mechanics will have on a player's chance to win. In general, the more complicated a setup, the swingier it is. A mountainous game would thus be the game with the lowest amount of swing while role madness tends to be high swing.

Certain roles have higher swing than others. Take for instance Cop vs Tracker. Both function in a very similar manner, giving their alignment additional information. Cop is definitely the stronger of the two (at least in the hands of town). On the other hand, Cop is a very low swing role, because it's hard to play cop in a way that hurts town. Barring getting lynched or killed without investigating, a cop is going to give useful information to the town. Tracker, on the other hand, could be very good for the town if it gets a guilty, bad for the town if it outs a PR on accident, or very bad for the town if it tracks something like a cop to the nightkill. Therefore, tracker is a high swing role. Other roles that are notoriously high swing are things like PGO (Could kill scum, could kill Doctors, and both because you were playing well as town) and Redirectors (notoriously hard to balance around). Basically, the more likely something is to generate faulty results or bite your faction in the ass, the higher it's swing.

As to where a game should fall on the swing scale, you want to have at least some, but honestly for inexperienced mods the best bet is low to medium. You could theoretically perfectly balance a 500 player mountainous nightless game using math. And no one would play it because that sounds like the most tedious garbage of all time. Similarly, Great/er/est Idea games are all more or less balanced based on the random chance element of the setup, but the level of swing involved often makes them unfun for many more 'serious' players. Queues also tend to have a certain expected level of swing to them. Normal games are usual somewhere in the mid to low range, although I have seen good Role Madness Normals with high levels of swing. Opens are almost universally low swing because of the player knowledge. If there's a way to break the setup, people will probably find it pregame, and any truly zany roles and mechanics are easily anticipated and recognized when they impact the game. Themes tend to be higher swing because that's the queue that lets you play around with weird and crazy mechanics. Regardless of your queue though, it's just a good idea to give your players as accurate an idea of the level of swing in your game as possible.

As an aside, modding philosophy:

As a mod, your number one goal should be to provide a good experience for the players. (Note, this is my opinion and is not universally shared.) My golden rule when it comes to modding is to always ask "Would the average player want to play this game?" Note, this does not translate to "Would I want to play this game?" because you'll probably find that if you make weird and zany games, you'll answer yes to the second and no to the first. That's not to say that weird and zany games are bad, but that you should always let people know what to expect from your style of modding.

Now, as for how balance and swing play into each other, that's a little more complicated. The short answer is that eventually you'll get a feel for what is and isn't balanced. How long that process takes is entirely based on the individual, and even the best sometimes drop the ball. The long answer is that there are a series of things you can look at to determine what your rough level of balance is, but it's still largely subjective.

First off, the common milestones. Most games are based around an established set of numbers of players. The closer you get to one of these sets, the easier it will be to balance the game. These are, for singleball,

7v2
10v3
13v4
18v5

Multiball is harder, because you have a larger range based on how you want to balance the factions against each other. Additionally, adding factions adds swing, which makes balance harder.

Once you've got your starting point, (or you've tweaked your starting point) you can start playing with roles and mechanics to bring things more in line. The things I always consider/tell people to consider when balancing a game are pretty straightforward.

How many of your roles are unlynchable? By that I mean things like Innocent Child, Doctor or Mason, where barring incredibly bad play, a claim should get you out of most jams. Obviously, scum fakeclaims can override that, but you can pretty much assume that the Doc is never going to be a mislynch.

How many mislynches does scum need to win? This, in most cases, is pretty straightforward. Cases like vigs, hiders and suicidal modifiers can change the math, but you can usually tell exactly how many town players scum are going to have to get to the gallows to win at the outset of the game. (This is the number of mislynches bit that came up earlier).

How many potential innocents do you have per night? (2, but most commonly one)

How many potential deaths do you have per night? (2)

How many potential guilties do you have per night? (1)

In the best possible play for town, how early is the game unwinnable for scum based off of innocents? (D3) (EBG and Neapolitan are probably unlynchable roles, Neapolitan can inno one per night. Neapolitan can get an extra inno off of Motivator. Motivator targeting town PR makes it unlynchable. Therefore, possibly 6 unlynchable slots with 10 alive on D3, with optimal results (none of the unlynchables/innos killed, Neapolitan targeting only VTs, Motivator targeting Neapolitan D1.))

In the best possible play for scum, how early is the game unwinnable for town? (D4) (EBG misses every night and claims/dies, town mislynches four times.)

Generally speaking, anything more than three for the first three questions is too swingy and anything earlier than Night 2 for the last two is too swingy. If a game can be decided D1 and it isn't in the micro queue, you should probably put a lot of effort into making sure people know what they're signing up for.

[Jingle]

I would give this setup the stamp of approval, provided that the following is true. Green Goo does not affect Goons (which as far as your Role PM's go is a perfectly reasonable interpretation because the factional kill and communication are not listed as possible powers, meaning that they are likely considered a factional feature and thus available to anyone of that faction, something the goo doesn't change.) and Motivator works as expected in that a player with a power is notified that they may use that power twice at the beginning of the following night, but a player with no power is not told anything. I'll check for cosmetics, additional mod information you might need, and anything else tomorrow or the day after, but it looks awesome to me. I always suggest a second reviewer if you can find one to weigh in. I'm not infallible, after all.

I would also have been fine with simply the addition of either the Goo or the Ascetic to the previous iteration, FWIW.

(Also, as a note, Goos tend to have a bad reputation, and you might have a better reception naming them Reflexive Poisoner and Reflexive Vanilla-izer. I honestly wouldn't here, because I think making reflexive, poisoner, and vanilla-izer into possible roles muddles the waters a bit more than you'd want with your list of roles, but it's food for thought.)

[Jingle]

Oh, and one last thing before I stop for the night. I've been toying with the idea of writing a modding guide for a while now, so if there's anything you think would be good to address/you feel isn't well defined I'd love to try my hand at helping/answering. Even if it's something you know now but you don't think was addressed anywhere easy to find or something that has nothing to do with this setup, I'd like to hear about anything you think would be helpful.

[Jingle]

Hi Ari!

Balance wise you're a little scumsided at this point, because honestly the chances that Neapolitan hits anything vanilla just aren't very good. Interestingly, VC is pretty much negative utility at this point because it's more likely for scum to aim for Goos and unlikely to get valuable information. Swing wise, double goo with 7 targeting roles is pretty high. I think its unlikely the Goos activate early (In fact, they should probably holster outside of weird circumstances) but you definitely have a chance of a night with 7 deaths. Which... not good for town. If you're fine with the swing being that high, you could probably combine the goos into one role to make them a little less impactful on the vanilla cop/ Neapolitan, at which point I don't think you need the hated modifier.

As far as a flavor name for the game, something along the lines of Vitium et Virtus (Vice and Virtue) would be my leaning, but you could also go with something a play on the balance or the seven and seven nature of the game as well.

[Jingle]

Exactly.

[Jingle]

I think the vanilla cop is fine as is. If you want to up it’s power, rolecop is probably the way to go. Poisoned players should absolutely be allowed to vote/be lynched. Scum choosing to fake being poisoned to live a day or fake not being poisoned after killing the goo is a viable strategy in some cases.

[Jingle]

Full role cop ups scum utility drastically. I'd also argue against it because I think it kills a lot of the beauty in this setup. The coolest thing you've got going, design wise, at least imo is that both scum and town have roles that give them an idea of what someone is, but not perfect information. Neapolitan has hard innocents but not hard guilties. Vanilla cop can tell scum if someone has a role but not if they're a good kill.

Role cop instead would let scum know whether that role was good or bad for them to shoot at. Imagine for a minute the scum role cop hits the Lawyer. They out the result, and town lynches the obvious scum claim (Ninja Lawyer, come on! How could that be anything but scum?). Not only did scum just get a free mislynch, they got a confirmed role that may well be town in this setup that played their role in an incredibly town manner. That's the kind of interaction that can decide a game by itself, if scum is halfway decent.

[Jingle]

I wouldn't announce publically, because my standard policy is to never confirm roles unless absolutely necessary, and a public confirmation does exactly that. Definitely would tell the players themselves when they get poisoned though, particularly because just targeting the goo doesn't mean the player got poisoned.

[Jingle]

You can, if you'd like, and it isn't really a concern balance wise. I wouldn't, to be honest, but would answer if the goo asked me if the poisoned players were notified of their status. Mostly because I intentionally try to push players to ask the mod questions if they feel their role is ambiguous/confusing and actively try not to give them ideas on how the roles should be implemented. That's my own little crusade to shift meta towards bombarding the mod with questions, though, so it's entirely up to you

[Jingle]

Scum:
Vanilla Cop
Goon
Conditional Ninja

Town:
Indecisive Elite Bodyguard
Ascetic Townie
Ninja Lawyer
Neapolitan
Conditional Loved Tracker
Indecisive 2-Shot Motivator
Activated 1-Shot Hated 50% Toxic Goo
4 Vanilla Townies

Assuming Ari doesn't see any gaps in my logic (which to be fair, is entirely possible), I'd sign off on the balance with the disclaimer that it's a little swingy. I'd definitely put the modular and the list of possible roles in the advert for the game to ensure you attract the right kind of people, as well as the deadline rules because they are nonstandard (standard being 2 weeks/2 days with 48 hour prods).

[Jingle]

Rules look good to me. The game will end bit is a bit redundant when you tell the players it's 11v3, but I assume that's just because you're going for a reusable rules template and it's perfectly fine to include.

You hit the big ones, like DBAD (Don't Be a Dick), Deadline/Activity, and boilerplate MS rules, which takes care of most things. The right to append and pretending to break a rule bits should cover everything else, so

[Jingle]

What do you think of the having prod timers be the usual length (48 hours), but players can only be prodded twice before being replaced on the third time? Do you think it's reasonable?

That was standard when I was new, FWIW.

[Jingle]

Also, as long as you make it public, you can use whatever activity requirements you personally think will make the game most fun.

Be aware that no one signs up for BAM games anymore.

[Jingle]

You are go for takeoff.