View Full Version : Calling All Mafia Players
This is probably more apt in the frontroom, but I figured I'd get a better response in here.
I'm slowly trawling through the creation of the roles of my mafia game (See sig), but I need a third party with experience in hosting mafia games to take a look over it. I need to see if it's both balanced, and if the feasibility and fairness of some of the roles would not impede the fun of the game, and the same with certain game mechanics. Not a huge in depth analysis of the game or anything, but just an experienced eye who would be able to give me some advice on my first mafia game (That happens to be a huge one, although I'll probably run a small one prior to that) However, I obviously want someone who will not be playing it in the summer, as that would spoil it a bit.
Thanks in advance. Should any of you be kind/stupid enough to take up this offer, I will show you some concept art for the banner made by our very own splitpersonality.
pevergreen
01-27-2010, 00:21
Happy to do it, if you don't mind the potential loss of me as a player.
I'm sure Pevergreen has wayyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy more experience than me, however I would love to get an insight of the game too (and will abstain from signing up) Besides, I have at least hosted a few successful game.... and a few (many) less successful... :sweatdrop:
Double A
01-27-2010, 00:34
Oh come on, you humble yourself, Medieval 2 mafia was brilliant!
Oh come on, you humble yourself, Medieval 2 mafia was brilliant!
ATPG finished it though (plus it was horribly unbalanced), I haven't the staying power :laugh4:
Anyway, shush ,this is subotan's thread, (but thanks all the same) :whip: :egypt:
This might be as good a thread as any to bring up something I've been thinking about for a while. Would anyone be interested in having an in-depth discussion about creating and balancing mafia games? There's a lot of resources on the web with respect to small-sized games, where balance seems to have been finely honed. However, there's almost no guidance about balancing games of 30+ players which we do constantly. I'm interested in talking about stuff like:
1) What's the ideal mafioso to town ratio?
2) How do multiple mafia families and/or independent roles impact game balance?
3) What roles are overpowered/underpowered and overused/underused?
4) Does the entertainment benefit of giving all players something to do outweight the balancing difficulties that result?
5) What's the best way to minimize lack of participation/WoGs?
There's lots of other stuff as well, but that's just off the top of my head. Would anyone be interested in and participate in such a thread?
This might be as good a thread as any to bring up something I've been thinking about for a while. Would anyone be interested in having an in-depth discussion about creating and balancing mafia games? There's a lot of resources on the web with respect to small-sized games, where balance seems to have been finely honed. However, there's almost no guidance about balancing games of 30+ players which we do constantly. I'm interested in talking about stuff like:
1) What's the ideal mafioso to town ratio?
2) How do multiple mafia families and/or independent roles impact game balance?
3) What roles are overpowered/underpowered and overused/underused?
4) Does the entertainment benefit of giving all players something to do outweight the balancing difficulties that result?
5) What's the best way to minimize lack of participation/WoGs?
There's lots of other stuff as well, but that's just off the top of my head. Would anyone be interested in and participate in such a thread?
Yes, would be a good sticky to guide people hosting a game too.
Double A
01-27-2010, 00:50
That would be a great resource for us to have... I still don't have half a clue how to balance games.
pevergreen
01-27-2010, 01:03
It gets more complicated when you deviate from the normal roles.
Work through your game. Simulate it on paper from your hopeful number of players.
Each of mine, I expected to come down to only a few people left. Inishmore did, as did The Riftwar.
When you introduce things like Lucifer in Netherworld 1 or Loki in Midgard 1, it becomes harder to predict.
I go through and when I think my game is balanced, I ask one other player I respect for their opinion.
This might be as good a thread as any to bring up something I've been thinking about for a while. Would anyone be interested in having an in-depth discussion about creating and balancing mafia games? There's a lot of resources on the web with respect to small-sized games, where balance seems to have been finely honed. However, there's almost no guidance about balancing games of 30+ players which we do constantly. I'm interested in talking about stuff like:
1) What's the ideal mafioso to town ratio?
2) How do multiple mafia families and/or independent roles impact game balance?
3) What roles are overpowered/underpowered and overused/underused?
4) Does the entertainment benefit of giving all players something to do outweight the balancing difficulties that result?
5) What's the best way to minimize lack of participation/WoGs?
There's lots of other stuff as well, but that's just off the top of my head. Would anyone be interested in and participate in such a thread?
Quick answers as I am tired:
1) The main principle is informed minority (Mafia) versus Uninformed Majority (Town). Thus, balance has to be the ratio which doesn't unbalance either side.
2) Depends on the victory conditions. One of the recent large games possessed a really underpowered mafia. Mafia should be the most powerful "faction", at least at the beginning (when there is zero information). Multiple familes and independents affect balance a lot. Too many famiilies can make it basically impossible for the town to win, and true-independents are often underpowered and basically a Mafia/Town random counterbalance.
3) YLC's role in Netherworld 2. It is a combination of a roleblock + investigation + immune to night-kill especially with the original town-aligned objective. Totally impossible for a Mafia to get rid of. Only way he was disposed of was due to being recruited by a Mafia-aligned. There are many underused roles, however, the most overused in a bad way is probably the "serial killer" or the joker. Only in games such as Lift-Escape or Bakerboy's last game, is where a joker role is quite balanced.
4) Answering directly from personal experience from the .ORG series, amazing large number of people do not do any actions at all, even if the game expects them to do it. This also unbalances the game as if a counterbalance isn't being used, it can overpower other abilities or roles. It does come down to the argument that a game host should select certain people who know how to play a role, however, this results in elitism and other various community problems.
5) You want the really harsh answer to this? Refuse people participantion in your game who are known to act this way. By being refused, it means they are then expected to act in a certain way to actually join a game, however, then it gets into a circular argument where some one so bad cannot seek to show they changed their ways as no one will trust them, as it will ruin their game. Again, this results in elitism. There are however, things which can result in low participantion.
a) Bad hosting quality, as in lack of write-ups or imagination by the host. Sometimes, a player would like more than what looks like a half-hearted sentence as a end of phase write-up.
b) Lack of Ability: While this doesn't effect everyone, some people are simply sick of being bog-standard townies. They want to be the mafia! They want to be the doctor! When some players are on a role-dry spell, they simply cannot be bothered playing much due to lack of reason why.
c) All the active players get lynched. Trust me, I know this from experience, so will many other regulars. This leaves the lurkers and as they don't bother to contribute, the game can dry up.
There have been some successful counters to this. My favourite was infact how I handled it in socially unacceptable mafia. I enforced a rule where some one had to post a minimum of three times per round. This forced anyone purposely trying to lurk, make their lurking obvious, which then generated discussion. Infact, I might re-implement this system again for my large game...
pevergreen
01-27-2010, 01:57
5) You want the really harsh answer to this? Refuse people participantion in your game who are known to act this way. By being refused, it means they are then expected to act in a certain way to actually join a game, however, then it gets into a circular argument where some one so bad cannot seek to show they changed their ways as no one will trust them, as it will ruin their game. Again, this results in elitism. There are however, things which can result in low participantion.
a) Bad hosting quality, as in lack of write-ups or imagination by the host. Sometimes, a player would like more than what looks like a half-hearted sentence as a end of phase write-up.
b) Lack of Ability: While this doesn't effect everyone, some people are simply sick of being bog-standard townies. They want to be the mafia! They want to be the doctor! When some players are on a role-dry spell, they simply cannot be bothered playing much due to lack of reason why.
c) All the active players get lynched. Trust me, I know this from experience, so will many other regulars. This leaves the lurkers and as they don't bother to contribute, the game can dry up.
There have been some successful counters to this. My favourite was infact how I handled it in socially unacceptable mafia. I enforced a rule where some one had to post a minimum of three times per round. This forced anyone purposely trying to lurk, make their lurking obvious, which then generated discussion. Infact, I might re-implement this system again for my large game...
I fully support almost all of this.
I refused entry, and would have refused entry to quite a few players (they didnt attempt to sign up) to a player who I knew would end up not even logging in after a few days. He pleaded and promised he would, but two days after that, he hasn't been back since.
Write ups - Need to be done on time. I have no skill in creative writing, but I still attempt them. They always look cooler in my head though :shame:
Bog standard townies are boring to play for many people. I like to think that everyone enjoyed themselves, as everyone had a role. Of course, I had the full backstory of 9 books, but details. :rolleyes:
Bog standard townies are boring to play for many people. I like to think that everyone enjoyed themselves, as everyone had a role. Of course, I had the full backstory of 9 books, but details. :rolleyes:
I know from experience, giving everyone a role doesn't mean they will contribute more either. Even had some one whine saying their ability was rubbish, when it wasn't (and the actions of others with the same/similar ability showed this)
pevergreen
01-27-2010, 02:07
You can't make everyone happy.
If you could, I'd be the same role I was in Fimbulwinter, cannalbalistc rune carver. Serial killer and leave messages for the town making them think i'm a detective. :laugh4:
Beefy187
01-27-2010, 02:22
1) What's the ideal mafioso to town ratio?
2) How do multiple mafia families and/or independent roles impact game balance?
3) What roles are overpowered/underpowered and overused/underused?
4) Does the entertainment benefit of giving all players something to do outweight the balancing difficulties that result?
5) What's the best way to minimize lack of participation/WoGs?
I'll give it a shot.
1) Found few topics on Mafiascums. But they were starting to discuss the meaning of balance it self so I gave up.
When I started playing here, I often heard that ideal ratio between mafia and town was 1:6 or 1:7. That seems to be the case with 12 player games (2 mafias, 1 Pro town, 1 kill each) where both mafia and town has a fair chance of winning. However in large games mafia seem to have larger ratio like 1:9, 1:10 or bigger and still pull off a good game. I guess it depends on number of pro towns and inactives.
2) A lot. Very rarely does each mafia families attempts to work together. They normally blow each other up early on, which favours the town. Though, if the town are to follow someone just because they contributed in lynching a mafia, they will suffer the consequences.
3) Role blockers can go either way. But in the games where each mafias can kill, Role Blockers can be both medic and a detective at the same time. So in that sense, they are quite overpowered. Jokers are overpowered too.
4) I enjoy the classical games every now and then. Simple games like Mafia series and God Father should be fun even for the towns.
5) I try not to give the important roles to player I don't know that well. Getting roles and getting wogged just ruins the game. As for preventing inactives, I guess smaller number of players and a interesting storyline/write up should provoke activity.
pevergreen
01-27-2010, 02:33
Jokers are overpowered too.
Jokers don't deserve to exist. Its simply too easy to get yourself lynched. Maybe if you say that they have to be lynched with at least x number of votes, on y day, including at least z of *small list of players*
Askthepizzaguy
01-27-2010, 02:35
This might be as good a thread as any to bring up something I've been thinking about for a while. Would anyone be interested in having an in-depth discussion about creating and balancing mafia games? There's a lot of resources on the web with respect to small-sized games, where balance seems to have been finely honed. However, there's almost no guidance about balancing games of 30+ players which we do constantly. I'm interested in talking about stuff like:
1) What's the ideal mafioso to town ratio?
2) How do multiple mafia families and/or independent roles impact game balance?
3) What roles are overpowered/underpowered and overused/underused?
4) Does the entertainment benefit of giving all players something to do outweight the balancing difficulties that result?
5) What's the best way to minimize lack of participation/WoGs?
There's lots of other stuff as well, but that's just off the top of my head. Would anyone be interested in and participate in such a thread?
I am VERY interested in such a discussion.
Dozens of game hosts have come to me asking for my assistance in balancing their games, because I have some experience with creating and designing games that seem to be balanced enough.
1) What's the ideal mafioso to town ratio?
1) This is largely dependent on what abilities the mafia have, how large the game is, what abilities the townies have, and how many different factions there are. However, 4 mafiosi seems to be the maximum number of mafia in a vanilla game with up to 60 players, because that is a rather large advantage when it comes to voting. You'd need to win any vote by at least 5, probably 6-7 votes just to make sure that you even have a shot at lynching a scum, at least until a mafioso dies. 5 or more scums on one team versus a vanilla town of any size is still greatly advantageous to the mafia IMO, due to the teamwork versus uninformed majority. 3 scums can successfully sneak through most of a large game, 4 is plenty.
The best answer here is that you'd have to define what the intended game is, before there is a guideline response. 1 family, 2 separate mafia families, large, small, or huge game... etc. Too many potential combos for there to be an easy guideline.
2) How do multiple mafia families and/or independent roles impact game balance?
2. If the mafia is overpowered when their rivals are defeated early, the game isn't balanced. Otherwise, I've had a game (Treehouse of Horror) where I successfully balanced 4 mafia families of 2 players, the boss of all 4, which had its own agenda, and the Capo (his boss) with HIS own agenda, and of course, the serial killer. The basic rule is, opposing mafia of almost any typical size are balanced, because the mafia will hit each other. I'd still recommend having at least 3/4 players in the game being townie, so that the town always has a decent shot of winning.
3 scums versus 3 scums versus 30 townies is probably balanced, but the town has a good chance.
Special note must be taken of how many murders each family has, multiple murders per family almost requires pro-town roles.
3) What roles are overpowered/underpowered and overused/underused?
3. 100% effective Detective roles are incredibly cheap and overpowered to me. They are prone to meta-gaming. I just don't like the idea that you can nail a scumbag on round (two) any other way than by a random vote. Detective roles should be balanced by only scanning for night actions (at best) or having a risk of failure, something. IMO. These are purely for balancing an overwhelming mafia advantage, such as having 4 members (too many, IMO) in a small game. And it isn't always effective, so if the detective dies early in a game like that, game over.
Joker roles are decent tricks for mini games, but the general consensus is to leave them out of larger games, FOREVER. :laugh4:
Any pro-town role becomes 100-200 percent more powerful when private reveals are allowed. Then a detective, just one detective, could form a pro-town cabal of 4-5 proven townies before he dies even if he catches no mafia.
4) Does the entertainment benefit of giving all players something to do outweight the balancing difficulties that result?
4. Depends. I personally find a game where everyone has an action to be insane to balance, especially if private reveals are allowed. Remember that every roleblocker can be a PROVEN pro-town role, every vigilante can be a PROVEN pro-town role, and every defender can be a PROVEN pro-town role: They can clear themselves merely with their actions alone... whereas a detective needs to correctly identify a scumbag or whatever to be a proven role (and even then they might have just been lying and guessing.)
In my opinion, roleplaying can help cover up the "boringness" of basic townie roles. But honestly if people need pro-town roles before they give a darn about mafia games, then they will be lame townies anyway. Some players don't want to play unless they have a cool role. And besides, the value of being a basic townie is that you can pretend to not be a basic townie and try to get yourself murdered or lynched instead of a pro-town role, the "askthepizzaguy method" of basic townie play.
Entertainment should be derived from good host writeups, good story, game balance, and the fun of talking and accusing each other, and being surprised by the results.
5) What's the best way to minimize lack of participation/WoGs?
5. I personally don't invite players who habitually drop from games due to inactivity. If they sign up and play, that's fine, but I have a mini-blacklist for repeat offenders, especially when no reason is given (such as RL unforeseen difficulties, which is a perfectly fine excuse). Players can also choose to lynch the inactive players themselves to take care of the problem (also known as day one, day two, and day three lynches... wink wink wink).
There should be an interesting story and an interesting discussion. If you have those things, the only reason why people aren't voting is because they either are too busy to play the game (and will be killed off eventually, and rightly, by the WOG, and probably shouldn't have signed up) or because they are trying to hide, or have no opinion. Voting (at least an abstain) should be mandatory.
Ultimately it is the player's decision to sign up to a game they can't play, so theres nothing the host can do. However, you can encourage participation by making participation mandatory.
I'll limit my inquires to the basic balance issue to start with. I'll start with the random assumption, based off of Beefy's post, that 1:9 is a good mafia:town ratio in a 30+ player game. So, in a 30 player game there will be 3 mafioso and 27 townies, with the mafia getting one kill per person. Assuming that is balanced, what then happens if we add on a single detective role? Does that skew the balance enough towards the town to require compensation on the mafia side? How much compensation? How about a doctor?
Essentially, I'd like to try and figure out a basic blueprint for a standard, balanced large game. Working from a balanced mafia and townie-only game with no roles, we could then 'weight' all other common roles (i.e. detective, doctor, serial killer, survivor, etc.) according to how much they shift the balance towards the town or the mafia. For an example, I will randomly assign the following values to roles, with + numbers benefiting the town and - numbers benefiting the mafia.
Detective +2
Doctor +1
Mafioso -4
Serial Killer -1
If we assume that the 'balanced' ideal has a starting value of 0, then adding in a single doctor would unbalance the game by 1 point in the town's favor. That could then be corrected by adding in a SK to restore the score to 0. If 2 detectives were added in, a bonus mafioso could be added to balance out the detectives.
Again, the above are just randomly assigned numbers with no basis in reality... I just made them to illustrate the system I'm thinking of. If we could arrive at some kind of consensus about the proper base mafia:town ratio, and the proper weighting of basic roles, that could provide a solid foundation for people to build their games off of. Of course it is pointless to try and account for every possible role and rule foible, but at least with a system like this people would know that if they added something on one side, they would need to then add something else back to the other side.
Thoughts?
Sasaki Kojiro
01-27-2010, 04:02
The majority of successful mafia games on the org have tons of roles, multiple factions, possibility for the townies to get promoted, or great writeups.
When doing GH style games, 2 mafia 28 townies 2 kills a night was pretty standard. But honestly, I don't know how well a game like that would do. They seem to be the exception rather than the rule. It's just hard when 2/3 of the people playing don't post or make a 1 post essentially random vote.
Which is ok by me, because crazy private intrigue and plotting make for the best games anyway.
pevergreen
01-27-2010, 04:05
One must also take into account if mafia do their own write ups, if hosts gives clues at death, if signatures are used for kills. I might sit down tonight and work on it. Beefy/ATPG/anyone else that uses MSN up for it?
It's wild that mafia:town rations here tend to be so low -- at another place I play, 1 in 4 is typical for games of 16-24 players, which is the norm. Occasionally 1 in 5 or lower if the games are at the upper end of that range or higher. Somehow it all works out just fine.
How is that possible? Balance is a weird thing.
(One thing is, these games have no free communication except among the mafia and sometimes masons. But if anything that seems to give the Mafia an edge, not the town.)
pevergreen
01-27-2010, 04:13
That would push the balance towards the mafia, but where you play elsewhere, is it the same group of people?
I've been playing mafia constantly for nearly 3 years, with a core of the same players.
Introduce such things as Holmes...
Sasaki Kojiro
01-27-2010, 04:23
It's wild that mafia:town rations here tend to be so low -- at another place I play, 1 in 4 is typical for games of 16-24 players, which is the norm. Occasionally 1 in 5 or lower if the games are at the upper end of that range or higher. Somehow it all works out just fine.
How is that possible? Balance is a weird thing.
(One thing is, these games have no free communication except among the mafia and sometimes masons. But if anything that seems to give the Mafia an edge, not the town.)
1:15 is the original standard. 24-48 hour days, mass lurkers, WoG.
Different people. Much longer days (about 120 hours on average). No WOGs, that's a point. Vigs are extremely common, as are serial killers, but then, they are here, too.
Beefy187
01-27-2010, 04:33
One must also take into account if mafia do their own write ups, if hosts gives clues at death, if signatures are used for kills. I might sit down tonight and work on it. Beefy/ATPG/anyone else that uses MSN up for it?
Sure. Tell me when to get on, otherwise I'm going to take a walk outside :smash:
pevergreen
01-27-2010, 04:43
3 hours 20 minutes from now? That may put it at like 6-7pm your time though.
Beefy187
01-27-2010, 04:52
That would be 3: 15 pm for me. Sure thing
pevergreen
01-27-2010, 04:54
What?
Aren't you ahead of me in the time zones?
Turns out no. Hour behind me.
:laugh4:
GeneralHankerchief
01-27-2010, 06:56
In terms of balancing, I don't really have a magic formula. The basis for Mafia I was one RL game I had played several years before, in which there were 60-odd players and 4 mafiosi, four kills per night regardless of the number of scum left alive. After Mafia I got 26 players, I realized that keeping the 4 scum/4 kills number intact would be suicide so I cut it down to two. That seemed to work well so I followed it for future installments. Even as the player count went up to 39, it always seemed to come down to the last round, so I never bothered to mess with a good thing.
Silver Rusher's Godfather series, which I borrowed to host Godfather 3, was based off my system but added an extra mafioso in the form of a Godfather. However, he balanced this by kills being limited depending on the number of grunts alive and the fact that a lynch of the Godfather automatically ended the game.
Pirate Ship Mafia I spent a long time balancing, and while the results may say I still need more work, I maintain that was mostly thanks to the Horsemen's freakish luck, so we'll operate on the assumption that it was fairly balanced. :tongue: Generally I just went by intuition there, basing most decisions from my observations with the Capo series (remove some mafia paranoia by making recruitment forced/automatic, nerf the protection groups by only allowing them one successful block per night, etc).
Generally, the advice I give to players asking for balancing help is if you're not sure which way to swing, usually err on the side of less potential kills. A tense 3-round endgame is better and more dramatic than there being five kills a night.
pevergreen
01-27-2010, 06:59
Generally, the advice I give to players asking for balancing help is if you're not sure which way to swing, usually err on the side of less potential kills. A tense 3-round endgame is better and more dramatic than there being five kills a night.
True.
In my build for Riftwar, I had 45 players, and if somehow everyone with a kill ability used it, there would have been i think 10 kills.
But most were delayed. Delayed serial killers, or sleeping mafia families are a great thing. It adds tension and scares the town as well :yes:
GeneralHankerchief
01-27-2010, 07:19
In response to two of TinCow's points, which I don't think have been fully addressed yet:
3) What roles are overpowered/underpowered and overused/underused?
Detectives, especially when you multiply them. Luckily, there seems to be a trend of removing them from games or making their investigations take more time to complete (see deep investigations in Capo, my tiered system from Omerta and Pirate Ship). I know khaan hates the role. Mafia IX will not have a Detective.
Finally, if there is an overpowered role, such as YLC's in Netherworld II or the Frenchman in Pirate Ship, compensate by making his objective more difficult. This usually works better in bigger games (Huge or a large game with 35-40 players and up), where said overpowered role also has to slog through more townie lynches. Attrition and general fatigue definitely becomes a factor in the longest games.
4) Does the entertainment benefit of giving all players something to do outweight the balancing difficulties that result?
From my experience with playing and hosting games over the years, this boils down to a simple formula, ignoring the balancing aspect for a minute:
- Multi-role, elaborate games: More fun for players while alive, much less so when dead
- Classic, vanilla-like games: Less fun for players while alive (unless getting a role), only a minor dropoff in fun once dead
A big part of what makes the Mafia series and Godfather 3 (and GF2, for that matter) memorable is the high participation rate of dead players. They can still pull for their team and heavily influence things in the thread with their arguments. I don't really need to give examples of this as you've all seen it. Whereas in the more elaborate games, sure you have a cool role, but once you're dead, you're pretty much out of the game. Gone is your ability to successfully argue in the thread since so much of the lynches there are based on the results of private discussions which you are no longer privy to.
So in the end, the more elaborate games are higher risk/higher reward when it comes to player satisfaction whereas the vanilla games are a safer option. This is assuming all other things are equal when it comes to balancing and host effort, etc, although the vanilla games are more likely to be better-balanced.
seireikhaan
01-27-2010, 07:52
A few bits.
1) YLC's role- :shifty: Y'all will find out.
2) Personally, I dislike the very concept of detectives. 100% detectives are too strong and can break the game in a whole other direction just by themselves. Detectives with even a 1% fail rate are worthless, since there's always the possibility of being wrong. Hence, its a broken class, imo.
3) Balancing. Not really a magic formula once you start adding more and more roles. GH has it down pretty well for vanilla games. However, large games with many roles and factions start to come down to intuition and scenario planning. Think- what if team A does this? If I was player B, what would I do? If there's one thing that I've quite evidently failed at, and something I would warn against if you make a more complex game, is built in mechanisms that prevent meta-gaming. While one needs to make roles that are balanced and fun, one also needs to make sure that good safety mechanisms are in place.
Lastly, some roles are just hard to figure out what the consequences are for. Best examples are serial killers and vigilantes. They could hit the mafia night one. They also could keep whacking away townies and let the mafia railroad the game.
1) What's the ideal mafioso to town ratio?
2) How do multiple mafia families and/or independent roles impact game balance?
3) What roles are overpowered/underpowered and overused/underused?
4) Does the entertainment benefit of giving all players something to do outweight the balancing difficulties that result?
5) What's the best way to minimize lack of participation/WoGs?
1) It depends on the abilities and if kills drop when a mafioso goes down. 1:9 for a large game with two kills each night, regardless of how many mafioso alive seems good to me.
2) It depends on what happens. Night 1 cross kills can ruin your game balance. You can make a scenario and twitch and tweak, but you can't prepare for all scenario's.
3) A detective able to reveal in private is a real game breaker imo. Detectives should only be allowed to reveal in public, imo. A detective can be tossed in for balancing purposes. For instance, if you have 45 players, you can pick 5 mafiosi with 3 kills a night and add a detective and a doctor for balancing. Alternatively, you can pick 3 mafiosi with 4 kills each night. Balancing is not exact science; you can't prepare for an unexpected turn of events. My first large game (The Secret Formula) ended prematurely because the serial killer and the last mafioso targeted each other. In later games, I developped a behind the scene-not known by the players system of rolling dices to prevent this from happening.
4) I don't think it's necessary to give all players a role. In all honesty, the vanilla games and the Godfather series are always enjoyable games. In fact, my next game will probably be either a vanilla game or a game using the Godfather mechanic, with some minor tweaks, mainly for story telling purposes. Townie is a great role if you dedicate yourself to it.
5) Tips for hosts:
- make the write-ups interesting or fun. You can do this by telling an ongoing story, funny write-ups, putting clues in your write-ups etc. If the write-ups are interesting, people will check each phase, if only to read your write-ups.
- make sure your phases don't drag on forever. Of course, we all have a life outside the .Org and extensions can happen, but phases should never last longer than 48 hours and end preferably within 24 or 36 hours. A phase longer than 48 hours could kill your game.
- dedication. Things have improved now, but there has been a period when hosts frequently abandonned their own games: not good.
If a host does all of that, then he shouldn't be blamed if the game suffers from a lack of activity. All a host can do is present his game and run it as good as possible. The rest depends on the players.
@YM & Pevergreen
Thanks, but I'd rather have someone who wouldn't be playing the game anyway. I'd hate to force someone out of the game just because I'm not experienced enough to know if it's balanced.
Bog standard townies are boring to play for many people. I like to think that everyone enjoyed themselves, as everyone had a role. Of course, I had the full backstory of 9 books, but details. :rolleyes:
That's one of the things I'm afraid of in my mafia game. One, I'm worried that people will find the setting confusing (Good write ups/backstory of Modern Chinese history should help though), and I'm widening GH's officer idea to include four people, plus a "captain". However, I don't want that to turn into a game where the officers run the whole show, and ordinary townies are left in the dark, and get bored.
3. 100% effective Detective roles are incredibly cheap and overpowered to me. They are prone to meta-gaming. I just don't like the idea that you can nail a scumbag on round (two) any other way than by a random vote. Detective roles should be balanced by only scanning for night actions (at best) or having a risk of failure, something. IMO. These are purely for balancing an overwhelming mafia advantage, such as having 4 members (too many, IMO) in a small game. And it isn't always effective, so if the detective dies early in a game like that, game over.
Joker roles are decent tricks for mini games, but the general consensus is to leave them out of larger games, FOREVER. :laugh4:
Any pro-town role becomes 100-200 percent more powerful when private reveals are allowed. Then a detective, just one detective, could form a pro-town cabal of 4-5 proven townies before he dies even if he catches no mafia.
Interesting. I originally thought that the detectives in my game were underpowered :laugh4:
Pirate Ship Mafia I spent a long time balancing, and while the results may say I still need more work, I maintain that was mostly thanks to the Horsemen's freakish luck, so we'll operate on the assumption that it was fairly balanced. :tongue: .
Although you'd think that I'd be the last person to want this, I want to reduce the power of vig groups in my game, in order to avoid a repeat of the Four Horsemen. That said, I still want townies to be able to take matters into their own hands, to keep it interesting.
Generally, the advice I give to players asking for balancing help is if you're not sure which way to swing, usually err on the side of less potential kills. A tense 3-round endgame is better and more dramatic than there being five kills a night.
That's good advice actually.
pevergreen
01-27-2010, 12:10
Maybe I just won't sign up at all then, :smug:
pevergreen
01-27-2010, 13:11
My interest in Chinese history goes as far as the silly stuff Dynasty Warriors makes up, and no further.
Some more points.
Any pro-town role becomes 100-200 percent more powerful when private reveals are allowed. Then a detective, just one detective, could form a pro-town cabal of 4-5 proven townies before he dies even if he catches no mafia.
Private reveals aren't necessarily bad. A good mafioso will manage to deceive his townie foes. But role pm reveals give an inherent advantage to the town though.
4. Depends. I personally find a game where everyone has an action to be insane to balance, especially if private reveals are allowed. Remember that every oleblocker can be a PROVEN pro-town role, every vigilante can be a PROVEN pro-town role, and every defender can be a PROVEN pro-town role: They can clear themselves merely with their actions alone... whereas a detective needs to correctly identify a scumbag or whatever to be a proven role (and even then they might have just been lying and guessing.)
I'm reminded of the words of Buddy from The Incredibles: "When everybody's special, no-one will be."
Entertainment should be derived from good host writeups, good story, game balance, and the fun of talking and accusing each other, and being surprised by the results.
:yes:
5. I personally don't invite players who habitually drop from games due to inactivity. If they sign up and play, that's fine, but I have a mini-blacklist for repeat offenders, especially when no reason is given (such as RL unforeseen difficulties, which is a perfectly fine excuse). Players can also choose to lynch the inactive players themselves to take care of the problem (also known as day one, day two, and day three lynches... wink wink wink).
Are any of those players on the sticky?
Detective +2
Doctor +1
Mafioso -4
Serial Killer -1
Again, the above are just randomly assigned numbers with no basis in reality... I just made them to illustrate the system I'm thinking of. If we could arrive at some kind of consensus about the proper base mafia:town ratio, and the proper weighting of basic roles, that could provide a solid foundation for people to build their games off of. Of course it is pointless to try and account for every possible role and rule foible, but at least with a system like this people would know that if they added something on one side, they would need to then add something else back to the other side.
Thoughts?
I haven't played enough games to know how powerful each role is. But it seems like a good idea though, but we'd want to avoid a situation where every game has a balance value of zero, so townies can go "Well, if there are two doctors, three detectives, then for it to balance there must be a serial killer i.e. Subotan". Balance isn't the be all and end all, and a slight imbalance in favour of one side makes it a challange. I just want to avoid totally one-sided games.
In response to two of TinCow's points, which I don't think have been fully addressed yet:
Detectives, especially when you multiply them. Luckily, there seems to be a trend of removing them from games or making their investigations take more time to complete (see deep investigations in Capo, my tiered system from Omerta and Pirate Ship). I know khaan hates the role. Mafia IX will not have a Detective.
I shall be applying your tiered system to A Bridge Zhou Far, if you'd allow me. Come to think of it, a lot of aspects of my game are influenced mainly by PSM (E.g. There's a captain role plus officers, two mafias etc.), but I have lots of other influences as well :yes:
My interest in Chinese history goes as far as the silly stuff Dynasty Warriors makes up, and no further.
You'd be surprised. I used to think like that about Chinese history, and I realised I was really, really wrong. It's totally engrossing, at least for me, and I'll take care to make sure that everybody understands the situation of the war(s) at the start of the game (Unlike the Riftwar :whip:)
Askthepizzaguy
01-27-2010, 13:56
@Subotan- I allow anyone to play if they sign-up. Maybe they will change their habits.
It's just that I don't invite certain repeat offenders, who are IMO irresponsible players.
However, I decline to name names because many of them might have been good players at one point and simply lack time for the .org now. It isn't about revenge, or humiliation, it's about inviting active people or people who don't often get WOG'ed. That's all.
Feel free to maintain your own list. My advice: one WOG isn't a big deal. Several is a big deal.
I thought so, but the sticky is a list of people who want to be invited, rather than just play.
On the mafia:town ration, there's got to be a mathematical way to figure this out. Assuming the random lynching of a person each night, there simply must be a ratio (probably not linear) at which both the town and mafia have equal shots at victory when the game starts. If we could figure that out, I think a lot of people would find it a very useful starting point. I think we need to find out the ratios for two specific game types:
1) Pure mafia vs. town, with 1 kill per mafioso.
2) Pure mafia vs. town, with a static number of kills for the entire mafia team
I've read through this paper (http://arxiv.org/PS_cache/math/pdf/0609/0609534v4.pdf) on mafia mathematics, but it mainly deals in optimal strategy and doesn't discuss optimal game balance as far as I can tell (though much of the functions in there are greek to me). If there are PhD-level papers being written on the mathematics of mafia strategy, someone must have figured out the proper statistical balance for a vanilla mafia/town setup.
On the detective issue, I'm fond of handicapped detectives. I think they're more balanced when their information is less certain, without being blatantly false. For instance, I like detectives that can see if a person is active, but not what they were doing or what their alignement was. That is useful information, but it has to be handled very carefully as it can risk exposing a pro-town role just as much as a mafioso. It also rarely fully 'clears' someone, as it would not be useful against an inactive mafioso. Similarly, I like target-oriented detectives, for instance a detective that only gets a positive result on a single specific mafioso, not the rest. Having only 1 possible target makes it much harder for the detective to find that person and it doesn't risk the other mafioso, but at the same time it still has the potential to aid the town decently under the right circumstances.
On participation issues, in my Rubicon game, I tried to mimic Seamus' success at involving everyone by allowing every role to have night actions with a group. While that seemed to work relatively well, it was a stupid amount of work and I don't relish having to do it again. I very much like the idea of each person having something interesting or unique about them, just to give the game flavor and make people more interested, but I am now working on ways of doing this that don't involve night actions. I think it should be possible to give every player a minor 'quirk' without unbalancing the game too much. Some of those can be the ability to pull off a night action, like a single-shot detective/vig/doctor or whatever, but others can be passive stuff like the ability to survive a night kill (also a potential balance to a higher mafia kill rate). Slightly tweaked game objectives are also worth exploring IMHO.
One of the innovations I really like about Reenk's game was the goal he gave to some people to "outlast" another person. That, IMO, is a superb word to use as it allows a person to have their target lynched or killed at night, but doesn't require survival until the end of the game. It's a very minor alteration which has the potential to add some confusion and intrigue to voting phases beyond the normal mafia confusion. I think the "outlast" goal is definitely worth looking into some more.
We first find a randomized strategy that is optimal in the absence of detectives. This leads to a stochastic asymptotic analysis where it is shown that the two groups have comparable probabilities of winningexactly when the total population size is R and the mafia size is of
order R^-1
...
Interestingly, as soon as there is one detective present, the game becomes fair only when the mafia consists of a linear fraction of the total population.
Wow! This seems to be exactly what we're looking for.
On the subject of giving everyone something special: a duel system where everybody can challenge another player is always a nice addition.
Wow! This seems to be exactly what we're looking for.
Yeah, but a history major like me can't translate that into actual numbers. What does that mean for a game of 20 players, 30 players, 40 players?
On the subject of giving everyone something special: a duel system where everybody can challenge another player is always a nice addition.
I've yet to see a duel system that actually accomplished anything more than entertainment value. It seems like something for bored townies to do to kill each other, and thus should be considered a mafia bonus which needs balancing on the town side. Am I wrong in that?
Askthepizzaguy
01-27-2010, 15:15
Duels can be a nice way to break tied votes. :2thumbsup:
I've yet to see a duel system that actually accomplished anything more than entertainment value. It seems like something for bored townies to do to kill each other, and thus should be considered a mafia bonus which needs balancing on the town side. Am I wrong in that?
The thing is, the duel system can be used as a tool to take out suspects, but often townies can't help themselves and challenge away. I consider it more as an entertainment and fun factor, but if town would focus, it can be used for finding/killing mafia.
Mafia doesn't necessarily need to have a high dueling score, in fact, giving them (or some of them) low scores makes it more interesting.
Also, never underestimate the power of random luck.
Yeah, but a history major like me can't translate that into actual numbers. What does that mean for a game of 20 players, 30 players, 40 players?
Means if there's no detective, then for 20 players, 4/5 mafia have an equal chance of winning, 30 players, 5/6, and 40 players, 6 mafia. That's without a detective of course, and probably just a vanilla game. I dunno what linear fractions are though.
Means if there's no detective, then for 20 players, 4/5 mafia have an equal chance of winning, 30 players, 5/6, and 40 players, 6 mafia. That's without a detective of course, and probably just a vanilla game. I dunno what linear fractions are though.
Based on the way I read that paper, that is with a 1 kill per night system for the mafia. So, that's a partial answer for #2 in my previous post. However, 1 kill per night for an entire mafia team is too low for Org standards, so we should try and find stats for 2 kills per night and 3 kills per night as well, as those would be more commonly used numbers in the Gameroom. We also need to find an analysis of the ratio when every single mafioso can kill individually, which I think it the most common setup on the Org.
I'd like to start writing down actual concrete numbers on this, so could you list exactly where the break points are for the different mafioso numbers? For instance, 4 mafioso with 18-24 players, 5 mafioso with 25-34 players, etc.
Based on the way I read that paper, that is with a 1 kill per night system for the mafia. So, that's a partial answer for #2 in my previous post. However, 1 kill per night for an entire mafia team is too low for Org standards, so we should try and find stats for 2 kills per night and 3 kills per night as well, as those would be more commonly used numbers in the Gameroom. We also need to find an analysis of the ratio when every single mafioso can kill individually, which I think it the most common setup on the Org.
We could write to them, asking for help/data. I'm sure they'd interested to know that their paper was being used by people who actually play the game.
I'd like to start writing down actual concrete numbers on this, so could you list exactly where the break points are for the different mafioso numbers? For instance, 4 mafioso with 18-24 players, 5 mafioso with 25-34 players, etc.
Obviously, it's the square numbers, so 9, 16, 25, 36, 49 and 64 etc.
So no-one is interested in taking a look at my game? Well, I guess that means no concept art for you :snobby:
However, as I am half blind at the minute, for reasons too complicated to explain, I won't be in position to discuss mafia for some time.
Beefy187
01-28-2010, 01:28
So no-one is interested in taking a look at my game? Well, I guess that means no concept art for you :snobby:
However, as I am half blind at the minute, for reasons too complicated to explain, I won't be in position to discuss mafia for some time.
I would like to, but I'm more interested in playing :yes:
Beefy187
02-01-2010, 15:13
Here is couple theories based on patterns
Normally our mini games is on 7 player max with 1 mafia, our standard small game is on 16 with 2 mafias. For 3 player, 36. (Previous game times number times 2+2n)
Assuming each mafia gets 1 kill each, 7 player game will have 3 lynch opportunity, 16 will have 5 and 36 will have 8. (N+ the previous N)
Number of lurkers/bandwagoners for each numbers 1, 4, 10. ( (previous n+1)x2) )
If that makes sense... I just thought those patterns is beautiful and it would be surprising if those work as a balancing standard.
I am actually finding this useful for planning my upcoming game, beefy. :grin:
Though I'm happy with how I balanced it anyway.... :shifty:
I haven't read all this read in detail so forgive me if this in unhelpful, but just from an overall mathematical view (not accounting for serial killers or masons etc):
Ok well this is no help at all, far more complicated than I first thought... :embarassed:
A large game with 30 players
1 mafia (1 kill)
29 town
chance of lynching mafia round 1,2,3,4,5 = 3-4% per round approx
6,7,8,9,10 = 5% per round approx
etc etc, round 20 = 10% approx per round
well one mafia was out the window anyway but still
2 mafia (2 kill)
28 town
chance of lynching 1 mafia round 1,2,3,4,5 = 6-7% per round approx
6,7,8,9,10 = 9-10% per round approx
etc etc, round 20 = 10% approx per round
well lets just think logically, on say a 30 player game, 5-6 mafia is too much (too many kills) 1-2 isn't enough (almost guaranteed to be lynched at some point) I think the 3 mafia margin is sensible, some things need common sense more than technicalities...but because of how individual each game is it can be hard to balance all ideas equally, besides if all games with a certain number had the same amount of good to bad it would be predictable in some ways, its for the host to make the decisions, if they get it right, bravo, if wrong they know how do improve in the future, but no amount of working out can account for luck, if mafia are all inactive or are all voted really quickly, then its out of the hosts power.
You wouldn't necessarily need 1-mafia 1 kill
You wouldn't necessarily need 1-mafia 1 kill
I know, which is why I scrapped that system, I was talking about a standard game though, you show perfectly that mafia games vary so much that they simply can't all be balanced.
Sasaki Kojiro
02-03-2010, 00:29
I saw a setup on another forum that I think we should run here some time.
This game is an open setup. You KNOW that the following and only the following are in the game:
17 Townies
5 Mafia
However, what roles you have is a bit trickier. Roles will be assigned via draft.
How does this draft work you say?
Everyone will, when they /in, send me a PM with two numbers between 1 and 15 (X,Y). The first number closest to 1 that no one else picks will get first pick. The next closest that no one picks will get second, etc. If you pick a X number that someone else has the Y number closest to 1 no one has picked will get to pick next. If two people have the exact same numbers then I will random.org.
Once all players that have the first number are done, all 2 numbers, then three:
An Example Draft:
Player 1: 4,9
Player 2: 1,4
Player 3: 3,5
Player 4: 1,7
Player 5: 1,4
Player 6: 15,15
Player 7: 15,14
The draft order would be:
Player 3
Player 1
Player 7
Player 6
Player 4
Players 2/5 based on random.
The reason that order is the way it is (and this is how this one works as well) is that the numbers that are picked singulary (X) go first, then the ones that are doubled (X), THEN the ones that are tripled (X).
Within X sets, the (Y) number is used for order just like the X is in the first parse. So that final set is:
Player 3: 3,5 <-- Single X
Player 1: 4,9 <-- Single X
Player 7: 15,14 <--- Double X, Y closer
Player 6: 15,15 <--- Double X, Y farther
Player 4: 1,7 <--- Triple X, Y single
Player 2: 1,4 <--- Triple X, Y same (random)
Player 5: 1,4 <--- Triple X, Y same (random)
The Draft:
Once the order is done, you will then straight down the list attempt to pick a power role. If someone above you on the list has picked that role you will be vanilla.
Note: The mafia will be able to talk before the draft begins for a period of approximately 24 hours.
The roles:
Each night you may target a player. If successful, you will get a result of town or mafia.
Each night you may target a player. If successful, you will kill this player.
Each night you may target a player. If successful, you will prevent any and all kill attempts on that player. You may NOT target yourself.
You are bulletproof. You are immune to night kills.
Each night you may target a player. If successful, that targets night action(s) will be successful.
You are the bomb. If you are successfully killed at night you will explode killing all players that targeted you.
Each night you may target a player. If successful, that target will not be able to perform any night actions.
Each night you may target a player. If successful, you will see all players who targeted that player that night.
Each night you may target a player. If successful, you will see all players that player targeted that night.
Each night you may target a player. If successful, you will be given the role that player has.
You will be given a copy of the first power role eliminated (lynch or night action) in this game. If multiple are killed simultaneously random.org will be used to decide your role.
Once during the game, you may PM me to stop a lynch.
Each night you may send me two names. These two players may then communicate outside the thread. You may send your own name as one of the names.
If you are lynched you may post "Kill: X" (where X was on the wagon that lynched you). Both of you will then die at the end of that day.
Each night you may send me the name of one player. If that player is targeted for a kill action you will be killed instead. You may NOT target yourself.
Of course, different balance would be required for the org. More townies and more roles I think. Reflexive cop could be added, neighborizer removed. Could think up some weakish new roles, or make some of them twice choosable. The mafia get their extra ability in addition to killing power. Not clear how many kills they get, would have to re-balance for the org anyway.
Well, I probably won't run it, but it'd be cool if someone did.
-edit-
Lot's of extra roles could be added...vote switcher, double voter, bus driver...
I saw a setup on another forum that I think we should run here some time.
This game is an open setup. You KNOW that the following and only the following are in the game:
17 Townies
5 Mafia
However, what roles you have is a bit trickier. Roles will be assigned via draft.
How does this draft work you say?
Everyone will, when they /in, send me a PM with two numbers between 1 and 15 (X,Y). The first number closest to 1 that no one else picks will get first pick. The next closest that no one picks will get second, etc. If you pick a X number that someone else has the Y number closest to 1 no one has picked will get to pick next. If two people have the exact same numbers then I will random.org.
Once all players that have the first number are done, all 2 numbers, then three:
An Example Draft:
Player 1: 4,9
Player 2: 1,4
Player 3: 3,5
Player 4: 1,7
Player 5: 1,4
Player 6: 15,15
Player 7: 15,14
The draft order would be:
Player 3
Player 1
Player 7
Player 6
Player 4
Players 2/5 based on random.
The reason that order is the way it is (and this is how this one works as well) is that the numbers that are picked singulary (X) go first, then the ones that are doubled (X), THEN the ones that are tripled (X).
Within X sets, the (Y) number is used for order just like the X is in the first parse. So that final set is:
Player 3: 3,5 <-- Single X
Player 1: 4,9 <-- Single X
Player 7: 15,14 <--- Double X, Y closer
Player 6: 15,15 <--- Double X, Y farther
Player 4: 1,7 <--- Triple X, Y single
Player 2: 1,4 <--- Triple X, Y same (random)
Player 5: 1,4 <--- Triple X, Y same (random)
The Draft:
Once the order is done, you will then straight down the list attempt to pick a power role. If someone above you on the list has picked that role you will be vanilla.
Note: The mafia will be able to talk before the draft begins for a period of approximately 24 hours.
The roles:
Of course, different balance would be required for the org. More townies and more roles I think. Reflexive cop could be added, neighborizer removed. Could think up some weakish new roles, or make some of them twice choosable. The mafia get their extra ability in addition to killing power. Not clear how many kills they get, would have to re-balance for the org anyway.
Well, I probably won't run it, but it'd be cool if someone did.
Great roles, That list should really add variation to games ran.
I'll add a few (I've used a few of them myself actually)
Clumsy Assassin (for lack of a better name)
Each Night you may Target two players, if successful, there is a 50% chance you will kill each target, giving a 0-2 kill ratio.
Distorter
Each Night you may Target a player, if successful, there night action can be directed at someone else of your choice (send who to redirect to) it cannot be redirected to yourself or your original target.
Taunter
Each Night you may Target a player, if successful, they cannot vote in the next day phrase
Rigger
Each Night you may target a player, if successful, there vote in the next day phrase will be controlled by you.
Gambler
Each night you must have at least 1 kill attempt on your life, if you don't receive one, you die. If you receive one or more, you survive.
Some of these roles look a tad unbalanced but can work depending on what other roles in the game consist of, anyway, just my two cent.
Beefy187
02-03-2010, 07:48
I've read the papers that TinCow linked. And I couldn't finish reading it with lots of numbers and codes. But it did tell me about few things. This is what I gathered. You can ignore it, as I can't think with a straight head most of the times.
Firstly, in a game with no detectives.
Let total number of the townies be R
Let number of mafia members be √R
So If its a 16 player game, there should be 4 mafias. For 36, there should be 6.
However Org mafia players will know that the amount of mafia in the said example is ridiculously high.
This is because there is couple of rules which they used.
1. Members can communicate each other anonymously and simultaneously. (Similar to Letter system I sometimes adopt)
2. All members are active each round (Never.)
And some other silly things we don't have.
We normally have half or a third of mafia members. That could be because of the amount of lurkers. And the fact that some players are clearly better then others in terms of mafia finding. If they get taken down first by meta gaming, mafia will have much better chance of winning the game.
Also they seemed to have constructed the strategy based on how much information there are.
Mafia starts with all the information while the townie starts with none.
Faster they can work up all the information, the better chance of townie winning.
So lets stay active.
That theory seems to be quite accurate.
But few things hinder this information theory.
- Many of the players have known each other for quite a while. This could be the advantage or disadvantage for the mafia.
This already ruins the patterns. We are not equal in the terms of mafia playing, so existence of one player can influence the outcome quite a lot.
- Some player even has a following. Namely Pizza guy, who normally gets many followers in the shape of bandwagoning.
If star players like Pizza guy for example is on the other side, it could severely influence the outcome.
So there fore the article of probability can't be the case in Org mafia.
Anyways, in terms of balancing goes it is hard to balance the game without considering the following.
- Activity level. How many lurkers are you going to expect
- Whos in the mafia. Who is the protown?
- How simple is the game? Is there going to be any twists, which could trouble the town and mafia?
If you keep those in mind and make the game, I think its going to be fairly balanced.
And I think the more appropriate mafia to town should be in the range of 1:7 to 1:10.
Also, nice roles Sasaki, TM :2thumbsup:
I might use one or two from that lists
Since Pevergreen is helping me balance the game, as promised, here is Splitpersonality's concept art for the banner (Which is as of yet unfinished)
Pretty cool huh?
Beefy187
02-04-2010, 08:26
Whats the flag on the bottom?
Is that the national party flag?
An interesting essay on balance:
http://www.mafiascum.net/forum/viewtopic.php?t=208
I thought people might like to hear some thoughts on the subject of game balancing, since there always is a bit of discussion about it here and there, and I had some thoughts recently that I wanted to share.
A common question people will ask of a game to determine it's quality is "Is it balanced?" It's a good question. But not a perfect one. The reason that balance is important is that one of the major points of a game is the tension that comes from the fact that you don't know who will win initially. It's a flawed question though, because balance isn't the important part of it. The important part is winnability. A player should be able to win, that way the tension of who will win is maintained.
I just lied. Winnability isn't important either. The only reason that winnability is important is because the player sees that he has a chance and continues struggling for victory. But if he didn't understand the game's situation completely, or something was hidden from him, the player might think he could win when really he had no chance.
It stands then that what is important is not winnability itself, but the illusion of winnability that allows a player to believe in it.
This is how mafia is balanced. At the beginning of the game, the players do not know the entire setup of the game. They probably all have some information, but not enough to make any guess as to the likely outcome of the game. Bear in mind that a player could have no chance to win and still believe that he does. This is really based on a sort of balance anyway. We could call it metabalance.
Metabalance is basically this: While perhaps no instance is truly balanced, the direction of balance changes enough from game to game that there is no way to accurately predict a given instance, and therefore should be played as if one has a chance of winning.
There is a way to break metabalance. If every game was, ultimately, unbalanced in favor of one faction, players who found that they were not in that faction would ultimately decide not to play, as they would have an unreachable goal.
Actually this allows for a lot of intriguing things to happen in mafia games. For one thing roles that aren't in a game can still affect it. Cop variations are a good example. Because cop variations have been used in other games, a cop in a game could be a variation, so he can't be automatically trusted.
Anyway, those are my main thoughts on the subject.
Sasaki Kojiro
04-15-2010, 05:22
I suppose this might as well be bumped so the discussions about pro town networks don't infect other games. I think we have two problems:
1) Pro town network takes over and totally dominates the game: I think this has been solved by a number of hosts. Allowing the mafia to take two actions means they can hide in groups with ease, and having attack/defend groups include a risk of death for those involved eliminates them as a townie confirmer as well. Ambiguous detective results, limited detective results, and results requiring multiple nights have also been shown to be very effective. Recruitement as a threat is the best of course, as well as godfather type roles and town vs town factions. I think any host can deal with the "totally dominates" problem.
2) The existence of even a small, inefficient network makes the game less fun for people "not in the know". Definitely true. The solution to this, I think, is to make it easier for people to form their own little networks and groups. Some feature where protection groups or vig groups that have a history of working together get an in game bonus--one of them can miss orders by accident and still appear in the group, or the vig group can kill with less people/have more input on the kill writeup, a distinctive calling card for example. Maybe something like CR's win conditions--your level goes up if your group completes a certain number of night actions. Maybe an established vig group's victory level goes up if they hit a mafioso, and if they hit 3 they get a "great victory" even if they are eliminated and the mafia win.
Reenk Roink
04-15-2010, 06:03
I'm not against the idea of Protown groups per se like many seem to be in reactionary terms to as I believe they aren't a problem because of a big handicap to the Mafia in general. Essentially, protown groups provide the town with collective information and target the voting power, so in general they help get rid of that uninformed and low influence nature of being town. However, the risks they bring are generally also amplified.
The protown groups have been of varying success, though the big to colossal failures (Settlement, Shadow fort) seem to be more marked than the average to moderate successes (PSM). Generally they are strongest in the small games (Zelda Mafia where everyone had a role, and there were only 2 Mafia being one where the protown group was very strong in the context of the game). Also, even where the protown groups succeed, it is generally due to being fortunate of some external event (such as the early townie play in Capo III or the Fh in PSM). Also, many hosts get rid of some of the more cheaper exploits that may be present (cover roles, ambiguity, unreliable detectives, no outside reveals/communication and so on).
I do agree that Protown groups generally are not much fun, however, this is because of a very specific reason. I tend to favor and like my ideas the most (btw Sigurd/GH/khaan, please begin deleting posts that don't agree with my POV...) and so I'm willing to go along with a protown group until it starts doing things I don't want or not doing things I want. And also as mafia, it is pretty fun to infiltrate a protown group. :beam:
By the way, by far the most devastating protown group was a really small one way back ago in Rise of the Mob when after Kage investigated Prole and got her lynched, he found the doctor Peasant Phill and revealed to him. Soon a network of dead and alive townies with those two arose, much to my chagrin. The greatest detective play ever, I thought it somewhat cheap at the time, but it was executed perfectly. :2thumbsup:
I've been talking with some people about this over at mafia scum, and the consensus seems to be:
1. Wow, you have very small numbers of mafia. 6 in a 53 man game? :O
2. Wow, you have HUGE games :O
3. You need to be less dependant on pro-town roles (I then explained Vanilla mafia to them and they were amazed)
4. Try some new formats. What about a mime game, where everyone can in addition to their vote only post two word posts? Or Haiku mafia, where everyone talks in haikus? The one proposal I really liked was the idea of al overs mafia, with there being say 8 lovers couples plus the 2 mafia.
Kagemusha
04-15-2010, 09:27
My 5 cents to the issue. I still believe that townie networks are not something Mafia cant defeat. In Capo II it took a similar networking for mafia to crush the town,when different mafia factions shared hands and took out the town in joint effort. In Capo III my mafia family was close to braking down the town. I was doing night actions with the detective and his group. (Johnhuthington and Tincow), while my Don Atheotes had infiltrated the pizza network. Our downfall was that disco the family Luca had revealed us all to Prole who was a CIA detective for some strange reason. Had that not happened.We could have ripped open both the detective faction and pizza network in few turns. In Shadow fort i asked myself a question who to recruit when i had the ability to do so. The qualifications were a player that would have been investigated early on. Being capable of networking in to the town and great personal skills. I decided to recruit ATPG and i think we showed that mafia faction could sit right in the main townie group, without the group having a clue that they were being manipulated and when the time came ripped apart.
To me what townie groups bring in the game is more challence.
EDIT: Reenk thanks for mentioning Rise of the Mob.It was great game! Also one of those rare instances i have been able to play a pro town role.Good memories!:bow:
Askthepizzaguy
04-16-2010, 02:20
Ambiguous detective results are the only way to go. I hate detectives.
Of the times I was mafia, the times I died were mostly because of detectives investigating me right away. Three times at CFC, once at TWC, and a couple times over here. And a couple host-influenced deaths. (Hosts leaving hints or changing the rules mid-game)
The times I won there were generally no detectives or my teammates were able to carry on without me. I am rarely ever lynched as mafia without a detective scan result. I think half the times I've survived games I've been mafia with no detective around.
Might I repeat that I hate detectives?
Try some new formats. What about a mime game, where everyone can in addition to their vote only post two word posts? Or Haiku mafia, where everyone talks in haikus? The one proposal I really liked was the idea of al overs mafia, with there being say 8 lovers couples plus the 2 mafia.
There have been a few Rhyming Mafia games at CFC. Every post in the thread had to rhyme.
Ambiguous detective results are the only way to go. I hate detectives.
I can only agree with this. Detectives should never be completely accurate, and there should be good chances for both false positives and false negatives.
In the last game I hosted (it's been far too long!), I had two semi-detectives, both of whom would get a positive scan on each other, but neither getting a positive scan on the Puppet Master (though on his henchmen).
Speaking of CFC, the detectives were one thing that civplayah did well in that game that just completed, Niklas. One guy able to check personal goals (plenty of red herrings and false negatives there), one guy able to check backgrounds (similar), and two players with what were effectively single-shot alignment scans. If not for me and Seon breaking the game by getting all but confirmed the first night (and so being able to gather most of the game's information in one place quite early on), it would have been very very difficult to put the pieces together in the right order.
seireikhaan
04-16-2010, 18:46
I can only agree with this. Detectives should never be completely accurate, and there should be good chances for both false positives and false negatives.
In the last game I hosted (it's been far too long!), I had two semi-detectives, both of whom would get a positive scan on each other, but neither getting a positive scan on the Puppet Master (though on his henchmen).
I disagree. Any detective that can have false results might as well as not exist- they have no purpose whatsoever. You can never know if they're right or not, so the results have absolutely no bearing on anything.
That said, I also dislike detectives. Watcher roles are much better.
Askthepizzaguy
04-16-2010, 19:39
I disagree. Any detective that can have false results might as well as not exist- they have no purpose whatsoever.
I disagree! :laugh4:
I challenge you on that point. They can be employed rather successfully in Dethy mini games, as well as provide negativity to the town's strength. Any bad detective could be used to hamper the town's efforts, to compensate for perhaps another, real detective, or to counter the fact that they have detectives and roleblockers and the mafia have none.
I also once designed a detective role that found everyone to be guilty, except for the main bad guy, who they found to be "unquestioningly the one you're looking for".
So it's a basic detective with misleading, but totally useful results.
Detectives who are wrong just to be wrong are a bit cruel to the people who are unfortunate enough to draw the roles (not to mention to everybody else). If there's warning of the possibility of false results, or if there is a way with enough effort and information gathering to figure out how to make use of the role properly, then I'm all for it.
Detectives who are wrong just to be wrong are a bit cruel to the people who are unfortunate enough to draw the roles (not to mention to everybody else). If there's warning of the possibility of false results, or if there is a way with enough effort and information gathering to figure out how to make use of the role properly, then I'm all for it.
I'd rather be a cursed detective than a vanilla townie, I imagine it would be interesting to play as, even if you weren't aware of it, it would still make for some interesting scenarios. In ATPG's example of a good and a bad detective, it would be one word against another, which would give us all plenty to talk about.
Best method is to remove investigation all together, and give clues and hints in private results and public results.
Sasaki Kojiro
04-16-2010, 20:37
I think the flavor cops work great. No really trickery, but it takes actual detective work to make use of the results. You have to piece together which type of weapons the bad guys have etc. And the multiple nights/can only find one type of mafioso thing works great too.
I'd rather be a cursed detective than a vanilla townie, I imagine it would be interesting to play as, even if you weren't aware of it, it would still make for some interesting scenarios. In ATPG's example of a good and a bad detective, it would be one word against another, which would give us all plenty to talk about.
Probably not -- in practice it would almost certainly be that we would lynch the first person the detective called out as scum, then as soon as that person flipped town, the detective would follow. Net result, two dead townies, no new information.
What do you mean by clues and hints, Beskar?
I've played two roles in the past year that were designed to mislead me and others, including one role that spit back guilty investigation results on some pro-town roles. While those roles were somewhat frustrating to play, in both situations I think they added more entertainment value overall to the games they were in. So, I consider them good roles... I just don't want a third one for a while.
Probably not -- in practice it would almost certainly be that we would lynch the first person the detective called out as scum, then as soon as that person flipped town, the detective would follow. Net result, two dead townies, no new information.
True.
Perhaps a scene of the crime should be described and then you need to piece it together, not get it handed to you on a plate, as I think Sasaki suggests.
seireikhaan
04-16-2010, 22:37
I disagree! :laugh4:
I challenge you on that point. They can be employed rather successfully in Dethy mini games, as well as provide negativity to the town's strength. Any bad detective could be used to hamper the town's efforts, to compensate for perhaps another, real detective, or to counter the fact that they have detectives and roleblockers and the mafia have none.
I also once designed a detective role that found everyone to be guilty, except for the main bad guy, who they found to be "unquestioningly the one you're looking for".
So it's a basic detective with misleading, but totally useful results.
So the purpose of a false detective is to harm the town? Sorry, I prefer not to mess around with negative numbers while doing mafia balance. Purpose of detective- find mafia. If they have unreliable results, the results cannot be relied on(duh). Ifso facto, quantum uno, detective with unreliable results has no purpose.
And a role like the latter is only ever purposeful in a pretty small game.
vBulletin® v3.7.1, Copyright ©2000-2025, Jelsoft Enterprises Ltd.