Mafia theory discussion

[Zag]

I have a couple of comments I'd like to make about our mafia games, and I'm interested in others' opinions. Some of these discussions can only be had while we're not in a game. Since only a side-game is going on now, and I think it is near the end, this seems like the best time.

1. 'No lynch' vote is always considered to be a pro-scum action.

I don't agree with this, especially in a no-vanilla game like mine or the one Sentran just ran. In any game where every night brings the town new information (either from some type of investigative role, or because of "little birds" or some such where random people learn things), then dragging out the game with fewer deaths is very helpful to the town. In my last game, I did a fair bit of analysis concerning how much the townies were likely to learn by day N, and I purposefully had to build in some doubt in what they knew, or just the collective investigative knowledge would drown even the best play by the scum.

2. Theme games should have safe claims.

This comes from a point that Deception made early on in Sentran's Game of Thrones game. If you create a theme game where knowing a player name will pretty-much tell everyone his alignment, then you really need safe claims for the scum. Fortunately, GoT has enough grey area that we (the mafia) were able to forestall pressure for a mass claim, but we would have been sunk if all the townies had pressed for it.

3. If a player is terminally inactive, should the mod:
- A. Just let the character float
- B. Mod-kill and end the day
- C. Mod-kill and continue the day
- D. Mod-kill and conditionally end or continue, depending on the alignment of the person killed
- E. Find a replacement (and what's your choice if one can't be found?)

4. Have we been too lenient on time limits? What do you think about strict (and fairly short) time limits imposed on the days, so we don't get the first day that lasts 30+ real days as we did in both Sentran's and my games? Do you think that this is a big benefit for the Mafia if you do impose such a limit?

If anyone else has discussion points they would like to talk about, I'll summarize here.

[Jedo the Jedi]

1. I didn't know this was the collected opinion. I usually see Townies doing this. I usually hate no lynches simply because they don't give the Town information from the flip. I can say in the early game they definitely shouldn't be used. In the end game, no lynches can be extremely useful, but they have to be used properly and with great caution. *shrug*

2. Oh my gosh, yes! The scum don't need a role, just a safe name to use. Believe me, most groups are smart enough to come up with some sort of role to go with a fake name. Still, you often run into a problem that the mod has used all of the main characters of a given theme for actual roles and the scum get some passable minor character. Towns will see through that every time. I encourage mods to sprinkle in a decent sampling of minor characters and perhaps leave out some major ones. That will balance the meta-gaming which seems perpetually to be against the mafia.

3. Always find a replacement first. ALWAYS! The mod designed the game to be balanced with the roles it has, so it will stay balanced most easily if all the roles stay in until they are killed in the game. If there can be no replacement found, I say mod-kill and end/continue based on balance.

A big thing in conjunction with this is that mods need to be careful about mod-killing one role and leaving another alive. The players will meta-game this knowledge and assume the role left alive is either an important power role or scum. That's a big problem.

Edit: Stop adding numbers as soon as I correct mine.

4. Strict deadlines are pretty important, in my opinion. This is something I learned from MS. Not having strict deadlines adds too much swing into the game. Generally speaking, in a large game like we just had, 3 weeks for the first day is reasonable followed by 2 weeks for all remaining days. Nights should last around 72 hours, and go full the time. Unless everybody is coming onto the forum hidden, people will look at logon times and meta-game.

As far as this approach helping the Mafia, well it really helps Town also. Discussion is bad for the Mafia; but aimless, circular discussion hurts Town. Ultimately, I think it is more beneficial for both because it removes possibility. Are you more likely to finish a project quicker with a deadline or without? I think the former because it says a decision has to be made by this point. Now, that decision could be poor with too short of a deadline, but that's why I think 2 weeks is plenty.
_________________
Paragon Tally: 18 mafia, 3 SKs (1 twice), 1 cultist, numerous chat scum...and counting.

[Zag]

(Now 4 points. )

I forgot to mention that in my game, the Secret Theme game, the Mafia would have had some very tricky choices if the town had pressed for universal name claim early. They would have to decide whether or not to claim their real names, without real knowledge of whether or not it does say anything about their alignment. However, if they had chosen to make something up, they would have been discovered as soon as the town knew what the theme was.

[Jedo the Jedi]

I know one point we often argue about on Day 1 is the place of claims and the response to them.

If you don't know my opinion by now, you must be new. Somebody has to be lynched on Day 1 (preferably, unless you also disagree with me on the No Lynch thing ), and as Monk has well-said, the Mafia are not going to claim something that sounds bad. The best thing is just to test the waters and get information from that flip so you can gauge interactions the next day. You can't depend on the mafia for this because they usually go for the most uninformative kill.

One counter I usually hear to this argument is we don't want to lynch the cop or some other powerful role on Day 1. Being gun-shy isn't going to help anything. Additionally, it might help if these people take solace in the fact that the mod (hopefully) didn't design the game around one super-role. If the game is balanced properly, losing the cop on Day 1 isn't going to cripple the town. (Never mind that I think games should be won through analysis and rhetoric than through powers. )
_________________
Paragon Tally: 18 mafia, 3 SKs (1 twice), 1 cultist, numerous chat scum...and counting.

[Sentran]

[Amb]

Safeclaims didn't use to exist. In strongly themed games, that presented the mafia too big a problem. The first game I played with a safeclaim seemed more balanced to me, as it forced players to not rely on the role claims, but the behaviour. I personally think that Mods can use them at their discretion, but they shouldn't tell anyone.

Regarding No Lynch - I believe that No Lynch can be a great strategy. When I was mafia once and the town No Lynched, it slowed down my strategy to kill town. It meant at dawn, there was an extra person alive - and that person could vote me. But in every game I ever suggest No Lynch, I get lynched. So guess what happened: As town I pushed for No Lynch and died. As scum, I shut up and never mentioned it and survived. These days I dont generally bother bringing it up at all. It gets people too mad...

[MNOWAX]

I'm anti -no lynch in the early art of games later on I'm for it.

I'm pro-safeclaims in themed games. other than that its up to mod discretion if they are actually given out.

as for your inactivity, I always feel like unless there was something major that happened, a modkill ends the day as if it were a lynch. as for MKing inactives, it depends on the situation and the game.

lastly, this is up to mod discretion again. Just make it clear what your intentions are at the beginning of the game, and it is all good.
_________________
The Man The Myth The Legend
MNOWAX

[Undercover Monk]

1. No Lynch: I am never in favor of no lynch on Day 1. Period. after that Im still generally against it but some situations make it a necesity. I believe as Jedo does that rhetoric and logic eventually win the day over investigations and night actions so I prefer the cold hard fact of a dead body when I am scum hunting.

2. I believe theme games should have safe claim. not claims. too many safe names are a huge advantage for the mafia who already have a major advantage of information on the town. jedo would argue that the mafia in sentrans game needed a safe claim but I would point out just how close they came to winning without a second safe claim.This game isnt supposed to be easy to win.

3.inactiveness is a virus that needs to be ended. I think 20= games are just too much right now. we have a decent group that we could afford one or two newbs in a 16-18 game and still have a few vets left to replace in if need be. problem with this game everyone who plays mafia on the forum was already playing.

4.I think our deadlines are fine its the day 1 fiascos we have to stop after day 1 the game flowed naturally. deadline extensions are necessarry as they act as a pressure valve if the game is getting stagnant or is flowing fine. the mod should be an invisible hand pushing the game along unseen. metaphorically speaking of course.

5. as has been said before on day 1 claims (other than cop) should be ignored. If you are the day 1 sacrifice deal with it. your death helps the town whether you are town or mafia. dont make me and jedo go forum to forum busting heads.
_________________
The Classic Blunders:
1. never get involved in a land war in Asia
2.Never go in against a Sicilian when DEATH is on the line
3. Never release Peyton Manning

[Jedo the Jedi]

[Zag]

New question:

5. Is a Serial Killer role too hard to win, especially in a big game? Not only are you all alone, but, unlike most roles, you have to be alive at the end to win. In a really big game, any win condition that requires you to survive is extremely unlikely.

I was thinking that a reasonable compensating advantage would be to tell the player who everyone else is. That is, he still has to be the sole survivor, but he knows everyone's alignment and possibly even everyone's role. He has the ability to seem prescient, but that will get him targeted by the mafia, so he'll have to temper how much of that info he gives out.

I know it would be fun to play, at least. It doesn't seem overpowered, to me, but I'm on the fence. Certainly just knowing all the alignments isn't too much.

[Jedo the Jedi]

The SK used to be a reasonable role when it was just doc and cop as the power roles. With all of your convoluted roles now, it does need a small perk.
_________________
Paragon Tally: 18 mafia, 3 SKs (1 twice), 1 cultist, numerous chat scum...and counting.

[Zag]

Hmm. Not much discussion on that one.

6. What's the right ratio for town to mafia?
In a 20-person game, I think I had 5 mafia. But if there is a higher ratio of townies, I think that it just makes the mafia more able to hide out and let the town self-destruct. It makes it more likely that they don't really have to commit to any of the early bandwagons, because they will likely be competitions between townies, anyway. Therefore, they will likely give up no information at all, and the town will be going into the later game blind This, of course, is only true if there are few or no power roles, since you have to have enough mafia to have compensating power roles in those cases.

But consider, for a moment, and all-vanilla game with 18 town and 2 mafia. The scum would have to be unlucky or foolish to get targeted at all in the first few days, so they'd be down to 12 town and 2 mafia and I doubt they'd have given up much in the way of information. Since they can somewhat manipulate things with their night kills, they can keep any opposing factions that have formed equally balanced and stay out of the spotlight for a long time.

So my point is that while it is obviously unbalancing to have too many mafia, I don't think that it hurts much to have too few.

[Thok]

[Deception]

5: SK is hard to win, which is why they get anti-investigation, anti-night kill, and a killing power. Often they get even more abilities. I think though that giving them the whole setup would be overpowered in the hands of a smart SK though. In a 15 player game with 3 mafia and 1 SK, I'd say they have a god chance to win as anything. Remember that 3 people always make it to LYLO - for the SK it's all about using what abilities you're given to make sure you make it. Good SKs will have a win rate over 15%, bad ones might never win. 15% comes from some research done on mafiascum a long time ago; remembering of course that every setup is different.

in a bigger game the SK is only winnable if they can't be killed at night I think. In a game with more than 20 players, you need significant power; any more specific would get problematic. It is the hardest role to win with though.

6. 20%-33%

For ordinary setups - 25% scum
For Town-powered setups or multiple scumteams - up to 33% scum
For low-powered setups - down to 20% scum

10% is just too low. 2v18 is 8 mislynches. Good luck with that

[itisally]

6. I agree that with fewer Mafia the hiding becomes easy, but there is no room for error either. If the Mafia made and early mistake the game would be over.
Imagine how disheveled the town would be after several mis lynches. Town gets frazzeled and makes silly mistakes at that point because they second guess all thier assumptions,
Is it possible that fewer mafia is actually a benifit to the town once you look past the statistics to the human factor?

It seems like 1 to 4 is about right and then power the roles accordingly. Adding nuterals completely throws me though.
_________________
I could agree with you, but then we would both be wrong.

[3iff]

I'm in the process of preparing for a game and I've asked a few of the regulars...now I'll ask everyone.

(Numbers added after the suggestion from Sentran)

7. Balance
The main concern I have is for balance. I've read the mafia-wiki site a few times, most of the important bits and some of the less important areas, but how do you game designers make sure one side doesn't overwhelm the other side (assuming one side doesn't get ridiculously lucky).

8. Leeway/a Mod's freedom of action
Do you leave yourself leeway to adjust balance during the game? If so, at what point is it that you're interfering in the outcome?

Someone suggested that a role could have 'fluid powers' where information might/might not be disclosed depending on the state of the game. Is this a good idea? I actually like this option.


I've also read threads here (and maybe stolen an idea or two).

All advice and suggestions appreciated (probably).

[Sentran]

I consider 3iff's questions on leeway and balance to be question 7 and 8.

9. Is there a "standard" order of resolution for night actions?
_________________
Sentran
"Speaking of double negatives, I haven't read greylab yet today." - Lifeinmomland

[Undercover Monk]

7. When you've mastered how to balance a game you can stop playing LOL. seriously no matter how many times you mod this is still the toughest part as we all like non vanilla and themed games it gets tricky if you screw up you take what you learned and adjust. Its why Ilike mafia here its about making a good game and logic and philosiphy not just playing a fun game.

8. lee-way some time you have to make decisions on the fly as a mod if some unexpected situation but I dont think fluidity in a role is a good idea. IMHO the less direct effect the mod has on game play the better. The mod has to be impartial.

9. Not that I know of. Its really up to the mod to determine that. I usually give role blockers first dibs but thats just me.
_________________
The Classic Blunders:
1. never get involved in a land war in Asia
2.Never go in against a Sicilian when DEATH is on the line
3. Never release Peyton Manning

[Jedo the Jedi]

[Deception]

7. You just gotta have the experience to know if something is balanced or not. Always check endgame situations that are broken and repair them. Check for bad role interactions as well as how powered each teams are and compare it to other setups you know that were balanced.

I can review any setup you want me to, but I offer the caveat that I'm not the best person to ask because I don't know what is fun - Only what I think is balanced; and of course if I look at your game I won't be able to play in it.

8. I appreciate zero changing the setup or balance after the game has started. There is the narrow exception which is so scenario-specific and rare that it isn't even worth discussing, and if you did a good job pregame then it should be no issue. Like Undercover Monk said: Impartial Mod

9. Jedos' link is my preference as well.

[Zag]

[Deception]

I don't because, for instance, immunity to investigation is thought of as an intrinsic property of the role. The role itself is the ability to, say, kill.

Immunity to investigation isn't an active movement, it's just a property of the role, not a function.

BUT, each moderator is different, so it's not like one interpretation is wrong (ALL interpretations that aren't mine are wrong! ); I would personally call your type of roleblocker a temporary vanillaizer, so it's not like your policy would ruin me in a game or anything when faced with this ability.

[3iff]

Interesting ideas. Thanks for the different views on dealing with things.

I did used to run a D&D dungeon. That allowed me to adjust things on the fly to the situation at hand...I wasn't actively trying to kill the dungeoneers but I've had occasions where they were outclassed and I allowed them an escape route for the sake of the longer term game.

Mafia is different where there are factions battling away at each other and it should be that I don't care who wins. It's important that I don't favour one side for whatever reason. If one side gets blasted away in a few days then perhaps they weren't good enough...assuming my setup is fair to begin with.

That's why I'm going to attempt a small game so I can learn first hand what (not) to do. I shall keep the game mechanics to myself but thanks for the offers of direct help.

[Elethiomel]

Small games can be harder to balance though... or at least more swingy.

[3iff]

Possibly...but in a smaller game there are fewer variables to deal with and less people to annoy by getting it badly wrong.

I'd prefer to start small.

[Jedo the Jedi]

10. How do you differentiate scum-tells and opposing theory?
_________________
Paragon Tally: 18 mafia, 3 SKs (1 twice), 1 cultist, numerous chat scum...and counting.

[itisally]

10. I don't, but mostly that is beczuse I don't believe in tells because of diffrences in play style. It is more about something being off in thier play or interactiond. But I am learning to trust Intuition more as I play.
_________________
I could agree with you, but then we would both be wrong.

[Deception]

10: It works like this: Can you see X saying what X said coming from town? Scum? What is the belief-factor of each option? Can't decide? It's a null tell, move on.

Theory is usually not indicative of alignment because it is influenced not from ingame actions, but from out of game philosophy which was determined before their role.

[MNOWAX]

[Zag]

Heh heh. Of course, any time the two of us interact in any way, from now on, everyone else will be screaming, "They're both scum! Just lynch them now!"

[itisally]

Meta gaming. It is one of the things that seem to get under so many people’s skin one way or the other.
Here is my thing on it. The whole point of Mafia it to look for patterns in behavior. Any one person’s patterns don’t change that much between games. As we get to know one another we see personality patterns. If someone acts outside of those patterns we ask why.
In that way no game can truly stay in the past. What is difficult to keep in mind is if they were scum when I saw that or not.
Time stamps are another one of those things. It is a piece of evidence. It is available to everyone who plays and we know that all of our posts in and out of the game thread will be read. Why do people get so bent out of shape when players reference these things as reasons?
_________________
I could agree with you, but then we would both be wrong.

[Zag]

Now you've stumbled on to part of MNO's genius. He can be completely different from one game to another, but then it turns out he had the same alignment. If he could avoid the foot-in-mouth disease that he occasionally gets, he would be the best player here, precisely because meta-gaming is useless against him.

[itisally]

but because of his randomness he lacks the credibility for an effective endgame.
_________________
I could agree with you, but then we would both be wrong.

[Undercover Monk]

Which is why he gets lynched so often. We know he's nuts but at some point during the game he always becomes a distraction.
_________________
The Classic Blunders:
1. never get involved in a land war in Asia
2.Never go in against a Sicilian when DEATH is on the line
3. Never release Peyton Manning

[itisally]

So what is the advantage to that style of play (not to rag on you MNOWAX, I just wanted to analyze a little outside of a game)

He has fun there is no doubt.
He gets people to react so that there is observable content.

Howeveer, it ends up being detrimentat to what everside he is on.

Is is just a resignation that winning is unlikely so stur things up while you can? It always had that kind of feel to me.
_________________
I could agree with you, but then we would both be wrong.

[Jedo the Jedi]

I'm putting this here because I don't want to start an entirely new thread. If it seems like it needs its own thread, I'll split it off.

Could we maybe actually set deadlines? It occurs to me these games last so long when they shouldn't. That may not be a problem for anybody, but I think it's a little unfair to those people not in the current game--they have to sit around and wait until the current game ends for the next one to start. If we don't want fixed deadlines, could we maybe bring back the question of having a mini and a regular game going concurrently?

I offer you these questions. Think carefully.
_________________
Paragon Tally: 18 mafia, 3 SKs (1 twice), 1 cultist, numerous chat scum...and counting.

[MNOWAX]

I like running mini games concurrently with regular games. Enforcing set hard deadlines seems like a bad way of shortening games, and honestly I think it tends to work in the mafia's favor. However, allowing an unlimited amount of extensions on a deadline ( always starting off with a soft deadline) may work as well.
_________________
The Man The Myth The Legend
MNOWAX

[Jedo the Jedi]

Well, all of the last three games (H&V, Survivor, and GoT) have gone almost exactly three weeks for Day 1. If we put hard deadlines of between two-three weeks, I think that would encourage people not to fall off the earth. Conversation always picks up when a "soft deadline" is brought up, so why not just have that constant pressure to keep the conversation going?
_________________
Paragon Tally: 18 mafia, 3 SKs (1 twice), 1 cultist, numerous chat scum...and counting.

[esme]

[The Great Crep'er]

@Jedo: Because there are people here who genuinely want to play and discuss in a leisure format. Putting deadlines feels somewhat like a punishment for those who need more time to discuss and think, enforced by those who have to be almost "coaxed" into caring and playing. It's a good idea, but making it a "hard" deadline almost seems to hurt those who wish to take a pro-longed interest in it (which is exactly what those types of things are intended to produce).

It feels like the one thing that would effectively reduce players who fall off the earth is replacing them with players who want to get into a game of mafia (and getting them in there fast). Not to take a page out of the bot's book, but it almost seems like a good idea to advertise. (Like a failing company would do)
_________________
Potato. Belgium. Eight. Random Lynch # Mafia awaits.