As was foretold, we've added advertisements to the forums! If you have questions, or if you encounter any bugs, please visit this thread: https://forums.penny-arcade.com/discussion/240191/forum-advertisement-faq-and-reports-thread/

How to Host a Phalla (mechanics & balance discussion)

NerissaNerissa Registered User regular
edited August 2008 in Critical Failures
This thread is dedicated to discussing variations on the Phalla / Wangshire / Mafia forum games. I'm going to start with basic rules and common variations / house rules, then add my own discussion in a separate post. As things come up in the thread that properly belong in this post, I'll edit it.

The Basic Rules

For a more in-depth discussion, see http://en.wikipedia.org/wiki/Mafia_Game

The "village" consists of all players in the game. Included in the village is a small number of bad guys. Normally, the bad guys all know one another at the beginning of the game, but nobody else knows anything about anyone other than themselves.

Each day (normally 24-hour period) the village discusses and publically votes for one person to be executed that night. At the same time, the bad guys choose one person to kill each night and PM the name to the narrator.

The good side wins if all of the bad guys die. The bad side wins if the number of good guys is less than or equal to the number of bad guys (on the assumption that at that point, they control the vote plus get their kill, so it's inevitable).

Usually, there will be one or more "special" roles among the good guys (and occasionally among the bad guys). Exactly who holds each role is only known to that person, although how many of each role exist is most often known.

Each night, the narrator gives the details of who died, and normally whether they were a bad guy. Some games also include clues or outright statements if someone who died is a good special.

Standard Special Roles

Roles will have different names depending on the theme of the game. Listed below are the usual "generic" terms for each role. Normally, specials exercise their power by communicating with the narrator via PM.

Seer -- once per night can "investigate" a particular person and find out from the narrator if they are a bad guy. Variations of seers may be able to determine a good special as well, although they may be unable to distinguish them from a bad guy.

Vigilante -- once per night can kill a person of their choosing.

Guardian -- once per night can protect someone from any death other than execution by the village. Whether they can protect against multiple attacks on one person in the same night is up to the narrator.

Masons -- a group of good guys, usually of similar size to the group of bad guys, who start the game knowing each other, and that they are all good.

Common Variations

Many larger games contain multiple groups of bad guys working at cross-purposes.

At least one game has included a council elected every other night who are able to overturn the popular vote. Normally, this shifts the execution to the person who got the second-most votes.

The typical size for a group of bad guys is 3, although an individual game may have more, particularly if there is only one group of bad guys in a larger game.

Common "House" Rules

You may not communicate regarding the game with someone who is dead, unless there is some specific mechanic in place to do so. In that case, communication normally goes through the narrator.

You may not use any anonymous method of communication to discuss the game. Forum-based PMs, external IMs, and e-mails are allowed as long as the person you are talking to knows who you are.

You may not use screencaps to "prove" anything to anyone.

Normally, everyone gets a "flavor" role in the game. Bad guys and good specials usually get a normal "villager" role in addition to their special role. This keeps people from using their initial PM to verify someone else's role. The two most common ways to do this are to (a) use the exact same message to everyone, or (b) give everyone a unique villager role.



OK, I think I got all of the basics. What am I missing? What variations have you seen? What did you like / dislike? How do you balance something like this?

Nerissa on
«13456738

Posts

  • NerissaNerissa Registered User regular
    edited March 2007
    Some games I've seen have had more than one of each special role. Usually they've got some limitation so that only one of them would be effective a night. Do you balance this out as equivalent to one full-time special, more, or less? It has the advantage (to the village) that if one of them dies, they're not completely out their special, but on the other hand, until they're networked, they may be working at cross-purposes.

    The early games had good specials showing as identical to bad guys for the seer. Good or bad? Should a seer actually know someone's role when they target them?

    Nerissa on
  • thorgotthorgot there is special providence in the fall of a sparrowRegistered User regular
    edited March 2007
    I really really want to do an incredibly short (5 people) game that I read about. It has

    1 mafia
    1 real seer
    1 naïve seer (always seers innocent)
    1 paranoid seer (always seers guilty)
    1 insane seer (always seers incorrect)

    Obviously, all the seers would think they are real seers, and the seer PM would be posted publically so the mafia can pretend to be a seer.

    The game starts out at night, but the mafia doesn't get a kill the first night. So the four seers get an investigation before anything else happens. They (including the mafia) reveal their investigations. From then on, it is a normal game. The village can vote to not lynch anybody.

    It's basically a giant logic puzzle. If played well by the village, the mafia doesn't have a very good chance, but that all depends on how smart the people playing are.

    thorgot on
    campionthorgotsig.jpg
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    A lot of the specials' power comes from working together. The guardian's power is somewhat useless before the endgame if he's not in contact with the seer or vig. The seer's visions don't do him any good if he can't get the monsters he finds killed, and while he can use members of his network to get a bandwagon started a vigilante is a simpler and more effective way to get the job done. The vigilante meanwhile is most effective if he knows who the seer and guardian are, since he can then kill people without hitting one of the specials, and can knock off a monster the night after the seer finds it.

    Hence, whether the seer can differentiate specials from monsters is a rather important decision that has a large effect on balance issues. Any game where he can't should have fewer monsters than a game where they can, and possibly more specials as well.

    Balancing games with roles split across multiple people depend on the above, but in general should be treated as a little more than a special. How much more depends on how much the seer(s) can discern; if they can see someone's role then split roles will have higher "special power" than if the seers can only determine innocence, and that in turn will have higher SP than if the seer views specials as monsters.

    Smasher on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    thorgot wrote: »
    I really really want to do an incredibly short (5 people) game that I read about. It has

    1 mafia
    1 real seer
    1 naïve seer (always seers innocent)
    1 paranoid seer (always seers guilty)
    1 insane seer (always seers incorrect)

    Obviously, all the seers would think they are real seers, and the seer PM would be posted publically so the mafia can pretend to be a seer.

    The game starts out at night, but the mafia doesn't get a kill the first night. So the four seers get an investigation before anything else happens. They (including the mafia) reveal their investigations. From then on, it is a normal game. The village can vote to not lynch anybody.

    It's basically a giant logic puzzle. If played well by the village, the mafia doesn't have a very good chance, but that all depends on how smart the people playing are.

    Oh man, this sounds like it could be so much fun.

    I take it the real seer only sees innocence or guilt and not what their role is?

    Smasher on
  • thorgotthorgot there is special providence in the fall of a sparrowRegistered User regular
    edited March 2007
    Yeah, all the seers do that.

    thorgot on
    campionthorgotsig.jpg
  • SquashuaSquashua __BANNED USERS regular
    edited March 2007
    Nerissa wrote: »
    What am I missing?

    Consistent naming conventions for the "mafia" aka bad guys or monsters or evil villagers or dissidents or whatever.

    :D

    Squashua on
  • NerissaNerissa Registered User regular
    edited March 2007
    jdarksun wrote: »
    Nerissa wrote: »
    What am I missing?
    The math. :) Likelihood of the different groups to win, etc.

    Feel free to provide it :)

    Really, that's where the balance comes in. If it's balanced properly, each group should have the same chance of winning at the beginning of the game. The question is, how many bad guys = how many specials, and which other variations can balance one another?

    Also, how does keeping certain mechanics from public knowledge (i.e. in one game it wasn't even known how many of each special were around) affect the balance?

    Here's a mechanic I'm looking at for my upcoming game.

    I'm going to give every person a secret. My "masons" will be pairs of people who know each other's secret, but don't know the other pairs. Because this makes them slightly less useful than the average "mason" I will probably have a few more of them than I would with a single group. My bad guys' secret will simply be that they are a bad guy. Everyone else will have a unique secret.

    I want to give the bad guys some advantage for finding out other people's secrets. I've pretty much decided on an extra kill for each secret they correctly report to me, however by itself that seems a bit too powerful. Only allowing them to actually use one bonus kill per day, and requiring that the person whose secret they know actually live through the night that they are reported will limit it somewhat, but I don't think that is enough.

    Does the bonus kill balance with, say, one less bad guy? Or should I give the good side more specials?

    Nerissa on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    As you have it now I don't see the secrets making any difference. What reason do masons have for giving their secret to anybody else, even if they're sure the other person is good? They can just vouch for each other and achieve the same effect without the risk of giving their secret to the monsters.

    Smasher on
  • GnastyGnasty Registered User regular
    edited March 2007
    I've heard people mention both thralls and some kind of necromancer character who creates zombies. What did those guys actually do, mechanics wise?

    Gnasty on
    i just wanna 'be myself'
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    Thralls are villagers who are working for the monsters. They have no powers and the seer sees them as innocent, but they win if the monsters do and lose if they don't. However, they don't count as a monster, so if all the regular monsters die they and he lose. Thrall may or may not start out in contact with the monsters.

    We haven't had a game with necromancers who can create zombies. There was one game with a scientist who can talk to dead people through the narrator and ask them questions, but he ended up not really being useful and hasn't been in a game since.

    Smasher on
  • NerissaNerissa Registered User regular
    edited March 2007
    Smasher wrote: »
    As you have it now I don't see the secrets making any difference. What reason do masons have for giving their secret to anybody else, even if they're sure the other person is good? They can just vouch for each other and achieve the same effect without the risk of giving their secret to the monsters.

    Well, two monsters can just as easily vouch for each other.

    Nerissa on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    Nerissa wrote: »
    Smasher wrote: »
    As you have it now I don't see the secrets making any difference. What reason do masons have for giving their secret to anybody else, even if they're sure the other person is good? They can just vouch for each other and achieve the same effect without the risk of giving their secret to the monsters.

    Well, two monsters can just as easily vouch for each other.

    Yeah, having the pairs will make the game play out differently. I meant that there's no incentive for anybody to ever tell anybody else the secret, so the extra kill mechanic will never get used.

    If anything, I think you'd need more monsters. The best strategy for the village is just to have everyone publicly mutually vouch for their partner. Then if one person in a pair dies and is innocent, the other is confirmed legit without even being seered, and vice versa for the monsters.

    Smasher on
  • NerissaNerissa Registered User regular
    edited March 2007
    Smasher wrote: »
    Nerissa wrote: »
    Smasher wrote: »
    As you have it now I don't see the secrets making any difference. What reason do masons have for giving their secret to anybody else, even if they're sure the other person is good? They can just vouch for each other and achieve the same effect without the risk of giving their secret to the monsters.

    Well, two monsters can just as easily vouch for each other.

    Yeah, having the pairs will make the game play out differently. I meant that there's no incentive for anybody to ever tell anybody else the secret, so the extra kill mechanic will never get used.

    If anything, I think you'd need more monsters. The best strategy for the village is just to have everyone publicly mutually vouch for their partner. Then if one person in a pair dies and is innocent, the other is confirmed legit without even being seered, and vice versa for the monsters.

    Wait, it sounds like you are thinking that everyone will be paired up, that's not what I'm saying. There will be like 3 pairs of people who know each other's secret. Going public is likely to get you killed by the bad guys, and your partner the next night when you are proved innocent, so I don't see them wanting to do that.

    Also, some people are going to get a bit of "extra" information about someone else, which will be keyed in some way to their secret. The intent is that if they use it to accuse someone, the response would be "and just how do you know THAT?"

    On the other hand, you might be right about nobody ever giving anyone else their secret. If that turns out to be the case, then anything I do to try to balance the bonus kills would end up overbalancing things toward the villagers. Hmm...

    Nerissa on
  • GnastyGnasty Registered User regular
    edited March 2007
    Just curious, but what do you mean by secrets? Is it some kind of special ability, or just some kind of factoid that the bad guys can use to kill you? Just because at the moment I'm not sure why anyone would ever choose to reveal their secret, especially if it didn't have any effect on the game mechanics (other than making them a target for the monsters).

    Gnasty on
    i just wanna 'be myself'
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    A simple (in principle) way to fix it is to make it so telling someone your secret gives some sort of benefit.

    The trick is figuring out what benefit would be good enough to take the risk of an extra monster kill, yet wouldn't be completely overwhelming if all the villagers did it.

    Smasher on
  • ElJeffeElJeffe Moderator, ClubPA mod
    edited March 2007
    I think it's going to be impossible to do the math once you throw specials in there, because for the numbers to be at all relevant, they'd have to figure in strategy on the part of both villagers and baddies. A smart person who uses his seer ability well is going to be far more effective than a dipshit who uses it randomly.

    ElJeffe on
    I submitted an entry to Lego Ideas, and if 10,000 people support me, it'll be turned into an actual Lego set!If you'd like to see and support my submission, follow this link.
  • AroducAroduc regular
    edited March 2007
    Basic balance is obtained at around 1 special per monster. This is a hell of a lot harder to control when there are multiple monster groups or monsters that get extra kills or the like, but it's a good starting point. A lot of the rules and roles were designed for the offline mode of play which has no way to privately communicate, so no groups can exist, which pushes the seer back into the same degree of power as the other specials.

    As for numerical stuff, it's kind of a pain in the ass to calculate since it's branching expected values for anything greater than 1 monster and -really- tough to get an accurate read on for any special and especially for forum games where you can communicate privately, any special can confirm himself to whoever he wants just by calling his shot as it were.

    It's a lot better (game balance-wise) to use specials that can do something once and then are done. One block. One kill. Etc. That still has the potential to massive affect the game (a guardian or vigilante alive with their power at the end slaughters the monster's chance to win), but the odds of them staying alive that long are pretty low as well. Alternatively, add a crippling handicap to it. Every time the vigilante kills, a random innocent is also killed. Guardian seers (and dies) as a monster. Etc.

    Also, this fascination the Phallas have with revealing the role of people as they die is really horrible. Half the fun is the intrigue and pretending to be something else. That's just personal taste though. It really swings things (again) back towards the village because it gives them more information and prevents monsters from lying.

    Aroduc on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    jdark, I don't know how he did his simulation but those numbers are wrong. Taking the 1 wolf/6 players example:

    The village goes first, and has no info. They stake a random player, who since the wolf has no friends to back him up has a 1/6 chance of being the wolf and 5/6 chance of not being the wolf. Assuming the wolf lives he then kills a villager that night. The next day the village has a 3/4 chance of staking the wrong person, then the wolf kills someone else, and he wins since it's 1v1. So, the chance of the village winning is one minus the chance of it losing, which is 1-5/6*3/4 = 1-.625 = .375.

    Smasher on
  • AroducAroduc regular
    edited March 2007
    Smasher wrote: »
    jdark, I don't know how he did his simulation but those numbers are wrong. Taking the 1 wolf/6 players example:

    The village goes first, and has no info. They stake a random player, who since the wolf has no friends to back him up has a 1/6 chance of being the wolf and 5/6 chance of not being the wolf. Assuming the wolf lives he then kills a villager that night. The next day the village has a 3/4 chance of staking the wrong person, then the wolf kills someone else, and he wins since it's 1v1. So, the chance of the village winning is one minus the chance of it losing, which is 1-5/6*3/4 = 1-.625 = .375.

    Those numbers are just 2 off. He's including the narrator and... I'm not even sure what else, but I think that the wolf gets a kill before day 1. It's easy to see by looking at the the 5 and 6 players having 1/3 and 1/4 chance for the village to win respectively, and the 3 and 4 players having zero chance of winning, which is obviously incorrect. It makes less of a difference as things increase though.

    Aroduc on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    Ah, Aroduc's got the right idea. The first person to get wolf'd and the narrator are included in the count, and as neither has any effect on the gameplay it's equivalent to our rules without counting the narrator.

    Smasher on
  • ElJeffeElJeffe Moderator, ClubPA mod
    edited March 2007
    I'm kind of wondering how the traditional, offline version of the game even works, regarding the seer.

    So you're the seer. You have a vision. Now what? If you had a vision of a good guy, you can't contact him without revealing yourself. If you had a vision of a bad guy, you can't convince anyone to stake him without revealing yourself. And the second you come out, you're dead. Unless you get lucky enough to have visions of 2 or 3 of the monsters prior to your death, so you can reveal them all at once before they eat you, it seems completely worthless. What am I missing?

    ElJeffe on
    I submitted an entry to Lego Ideas, and if 10,000 people support me, it'll be turned into an actual Lego set!If you'd like to see and support my submission, follow this link.
  • GrimmyTOAGrimmyTOA Registered User regular
    edited March 2007
    I've played a couple of live games -- and generally either the seer just tries to implicate the mafia all subtle-like, or comes right out, says that they're the seer, and hopes that the guardian believes them.

    Live games are pretty small, generally, and so killing off even one monster near the beginning of the game can make the self-sacrifice of the seer worthwhile.

    It's not the ultimate power that it seems to be in the online version, however.

    EDIT: Changed 'finger' to 'implicate' so as not to distract the prurient minds of oDaM.

    GrimmyTOA on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    ElJeffe wrote: »
    I'm kind of wondering how the traditional, offline version of the game even works, regarding the seer.

    So you're the seer. You have a vision. Now what? If you had a vision of a good guy, you can't contact him without revealing yourself. If you had a vision of a bad guy, you can't convince anyone to stake him without revealing yourself. And the second you come out, you're dead. Unless you get lucky enough to have visions of 2 or 3 of the monsters prior to your death, so you can reveal them all at once before they eat you, it seems completely worthless. What am I missing?

    The guardian? If I were him I'd protect the claimed seer while we voted to stake the claimed mobster. If he's innocent then obviously the seer is a fake, and if he's guilty then there's at least a decent chance he's not a mobster posing as the seer.

    Smasher on
  • AroducAroduc regular
    edited March 2007
    Often there isn't a guardian, and this is where it's fun to play with a lot of experienced people who are great at lying. A villager may have no actual clue, but just want to make sure that someone gets staked, so they claim to be the seer and go after staking X. Or maybe on the second to last turn, the monster says "I've seered X and Y and they're clean, therefore the monster is M or N, so if we stake both of them, we win," or the seer does the same as either of those just out of good intentions.

    I've played a game where no fewer than 4 people claimed to be the real seer, all for different reasons entirely.

    Aroduc on
  • DevoutlyApatheticDevoutlyApathetic Registered User regular
    edited March 2007
    Aroduc wrote: »
    Often there isn't a guardian, and this is where it's fun to play with a lot of experienced people who are great at lying. A villager may have no actual clue, but just want to make sure that someone gets staked, so they claim to be the seer and go after staking X. Or maybe on the second to last turn, the monster says "I've seered X and Y and they're clean, therefore the monster is M or N, so if we stake both of them, we win," or the seer does the same as either of those just out of good intentions.

    I've played a game where no fewer than 4 people claimed to be the real seer, all for different reasons entirely.


    I've also seen it played over a longer period of time interspaced with other activities. This requires an environment were a group of people are generally around each other most of the day but could meet at specific times. (Say meal times were the "Nights") In that time exists for secret meetings and what not but are still somewhat visible to the other players.

    DevoutlyApathetic on
    Nod. Get treat. PSN: Quippish
  • SquashuaSquashua __BANNED USERS regular
    edited March 2007
    ElJeffe wrote: »
    I think it's going to be impossible to do the math once you throw specials in there, because for the numbers to be at all relevant, they'd have to figure in strategy on the part of both villagers and baddies. A smart person who uses his seer ability well is going to be far more effective than a dipshit who uses it randomly.

    You also have to count for attrition.

    Squashua on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    I think it'd be constructive to go through all the past games and see what separates the evil victories from the village victories.

    Newham (Evil victory): There were 3 of each special, with the seers and guardians splitting one use per night across all of them and the vigilantes each getting a kill every night if desired but progressively becoming more insane with each villager they killed.

    There were two main evil cults who started out seperate but were on the same side, and who got a kill and an evil/not evil seering each night. They ended up hooking up on day 2. The two cults could also give up their kills on one single night to convert someone to either cult; this ability was used at the very end, but didn't really have an impact on the game.

    There was also a third hidden cult that nobody else knew about. They didn't get a kill, but could deflect stakings on even nights to a target of their choice, so long as the original target was not a member of their cult. They also seered/died as innocents, though there were subtle clues to the contrary in the narration that most people didn't pick up on.

    The non-guardian specials all got slaughtered in relatively short order; however, this was largely due to them not hooking up and trying to act on their own, which drew monster attention quickly. If the specials had played better the game would have been fairly significantly more balanced, if not tilted towards the village due to the large base network. At the same time, the third cult succeeded in deducing and killing three of the main cultists, which isn't a factor that can be counted on when balancing games.

    Communist game (Village victory): Everyone was some sort of special in this game. Due to that, it's hard to apply any lesson from it to a normal game, so I'm not gonna bother here.

    Sicily (Evil victory): This small game had 3 mobsters with 2 kills a night, 1 seer, and that's it. It went south quickly for the village with the seer getting killed the first night, but even without that was unbalanced due to the 2 mob kills a night with no vigilante. The only reason this came close was that the narrator killed 2 of the 3 mobsters when they didn't send in a kill one night.

    New York (Evil Victory): Another small game run at the same time as Sicily. The seer got two visions a night, while the mob got one kill. There were two vigilante groups, the FBI and the NYPD. Each group got one kill a night and had 3 members, but also had one mole who worked with the mob. Unlike the Communist game, this mole was an actual member of his group and could exercise the power. The third and last member of the mob was the don. If he got killed the village won immediately.

    This game was well balanced IMO. The seer's two visions and mob's vulnerability in the don were offset by the moles being able to influence the two vig kills. This should have been a tight villager victory, but one of the moles (the last in his group) managed to convince the villagers only the other group had a mole in it.

    Councilors (Evil victory): Another small game about the same size as Sicily and New York. This had the standard vig/seer/guardian, as well as two competing rogue groups with 2 and 3 members respectively and kills on alternating nights. The interesting twist in this game was a mercenary with seer and guardian powers (could do both every night), and who could work with one or the other of the rogue groups. The loss of the seer the first night was offset by the death of two rogues from opposite teams who also died the first night. This came down to a very tight finish, with the rogues pulling out the victory due to the mercenary convincing a villager to work with him and the rogues instead of the village.

    Balancing this game would be difficult without making it too short, but I would reduce the rogues from 5 to 4 and give each rogue team a kill per night. I would then either take away the mercenary's guardian power or inform the village of the mercenary role.

    Castle (village victory): A medium size game, with seer/guardian/vig along with three one-time use items that conferred similar respective abilities to whomever used them. Three servants served as a standard stonecutter group. There were four assassins with one kill a night, and a wraith with no clear victory objective who could kill once per night, as well as retaliate against anyone killing him. The wraith proved to be a big wild card, killing both the guardian and two of the assassins. The items proved to be the biggest balance issue, as they could be traded in such a way as to prove a significant number of nobles innocent to each other, which more than made up for the seer dying the first night. Fixing this would probably be easiest by making them untradable.


    Not listed are the first four phallas. These were afaik all villager victories except for rishiri, but I didn't participate in them and don't remember them well enough to give an accurate accounting.

    Smasher on
  • ElJeffeElJeffe Moderator, ClubPA mod
    edited March 2007
    I'd like to point out that in Newham, I intended to make it so that the villagers were screwed, and it was basically a battle between rival cults with the villagers caught in the crossfire. It still was a tight game, and the villagers came close to winning in the end. Oh, and a secret:
    I fucked up the end of that game. I ran the numbers and thought that the cultists were guaranteed victory. Later on, I realized that the villagers actually had like a 15% chance of winning. They were probably screwed, but not definitely. Whoops.

    I think the moral, here, is that if it seems like the bad guys are going to completely kick the village's ass, the game is probably balanced.

    ElJeffe on
    I submitted an entry to Lego Ideas, and if 10,000 people support me, it'll be turned into an actual Lego set!If you'd like to see and support my submission, follow this link.
  • NerissaNerissa Registered User regular
    edited March 2007
    In the first Phalla, we learned that people will compare PMs, thus giving the good guys all one PM and the bad guys all a different PM is a bad idea. I don't think there were any vigilantes or guardians there, and only one group of bad guys (vampires).

    In the second Phalla, we learned that if you are going to give out separate PMs, then make them ALL different. Again, people were comparing and finding their own "mason" groups based on the PMs. We also learned that Richy's towns have more than one mayor.

    Rishiri (#3) was the first (only? I missed a couple) time we saw a council. We also had an unknown number of specials. Also, Shinto outright lied about some of the mechanics. Specifically, he said that all of the monsters were chosen randomly, when in reality he had chosen a head monster for each group and allowed them to choose their cohorts. IMO, this is kinda cheap. If you don't want to give out the information, just say so, but don't outright lie. The narrator is supposed to be impartial. This was the first one I played, and I know the oni got taken out relatively early, I believe the ninjas won? I think it was close, though.

    Nerissa on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    There are some mechanics I either dislike or am wary of. I'll post my thoughts about them, and other people can comment if they want.

    1) Thrall. The seer is useful for two things: finding bad guys and verifying good guys. Thrall severely compromise the seer's ability to do the latter, and if the thrall gets into the seer's network he blows the whole thing wide open.

    One might argue that the seer network tends to be too powerful, and depending on the setup of a particular game this can certainly be true. However, I would prefer accomplishing a fix for that by giving the seer fewer visions (say, one every other night) than using a thrall. The reason for that is simply that the effect of the thrall is too random; if the seer happens to vision him there goes the network, and if he doesn't then the network is still too powerful. This is even more true if the evil group gets more than one kill per night.

    The thrall can be defended against to an extent by dividing the network into cells, but what if the thrall is the first one seered? The seer has to trust somebody to get the network started, and the thrall makes that a roll of the dice. That leads us nicely to the next issue:

    2) Uncontrollable randomness. Obviously there are random elements to the game; the distribution of roles is the biggest one. However, other than that I feel most if not all of the other randomness in the game should be created by players' choices. Phalla is meant to be a game of strategy and deception, and adding metaphorical dice rolls doesn't improve that in any way I can see.

    If a player commits a certain action, and that action isn't affected by the action of another player, then the result of the first player's action should be the same every time. If a guardian role is split across multiple people, then on a given night a guardian should be able to depend on having or not having his ability work. If two people tie for the staking vote, they should both die (or some other consistant result, if you can divise a better one). If a player has a deflection ability, give him the ability to control where it goes or don't give him the ability at all.

    When things happen randomly, it feels like I'm playing monopoly instead of Phalla. A useful analogy is poker, where the cards you get are random but how you play them is all you and your opponent. That takes skill; rolling dice doesn't. I don't know about anyone else, but I don't play Phalla to pass Go.

    3) Hidden game mechanics. A good part of playing Phalla is controlling information. The monsters don't want everyone else to figure out what they are, and simultaneously want to confuse things for everyone else as much as possible. The specials want to figure out each other's identities without also letting the monsters know them. Being good at Phalla means being able to both hide your information and pick out other peoples' information through their lies.

    Hidden game mechanics throw a wrench in things. Hiding the number of monsters or something like that is fine, as long as it's made clear to the players this is the case. However, hiding important roles undermines players' abilities to make informed decisions, and in a sense leaves everybody playing a slightly different game. It makes the game unpredictable in a (to the player) totally random way, which ties in with my earlier point. The game becomes less about figuring out your oponents and more about figuring out the narrator.

    I won't say that hidden game mechanics should never be used. I do think that they're currently overdone, and should only be used when a role requires players not to know about it in order to be effective.

    Smasher on
  • NerissaNerissa Registered User regular
    edited March 2007
    Smasher wrote: »
    There are some mechanics I either dislike or am wary of. I'll post my thoughts about them, and other people can comment if they want.

    1) Thrall. The seer is useful for two things: finding bad guys and verifying good guys. Thrall severely compromise the seer's ability to do the latter, and if the thrall gets into the seer's network he blows the whole thing wide open.

    One might argue that the seer network tends to be too powerful, and depending on the setup of a particular game this can certainly be true. However, I would prefer accomplishing a fix for that by giving the seer fewer visions (say, one every other night) than using a thrall. The reason for that is simply that the effect of the thrall is too random; if the seer happens to vision him there goes the network, and if he doesn't then the network is still too powerful. This is even more true if the evil group gets more than one kill per night.

    The thrall can be defended against to an extent by dividing the network into cells, but what if the thrall is the first one seered? The seer has to trust somebody to get the network started, and the thrall makes that a roll of the dice. That leads us nicely to the next issue:
    The thrall is a mechanic that started to be used in the games I missed, so I've never actually seen one in action. It's an interesting idea, but I think you're probably right that it skews things toward the monsters pretty strongly.
    2) Uncontrollable randomness. Obviously there are random elements to the game; the distribution of roles is the biggest one. However, other than that I feel most if not all of the other randomness in the game should be created by players' choices. Phalla is meant to be a game of strategy and deception, and adding metaphorical dice rolls doesn't improve that in any way I can see.

    If a player commits a certain action, and that action isn't affected by the action of another player, then the result of the first player's action should be the same every time. If a guardian role is split across multiple people, then on a given night a guardian should be able to depend on having or not having his ability work. If two people tie for the staking vote, they should both die (or some other consistant result, if you can divise a better one). If a player has a deflection ability, give him the ability to control where it goes or don't give him the ability at all.

    When things happen randomly, it feels like I'm playing monopoly instead of Phalla. A useful analogy is poker, where the cards you get are random but how you play them is all you and your opponent. That takes skill; rolling dice doesn't. I don't know about anyone else, but I don't play Phalla to pass Go.
    I agree with you here, Phalla isn't about randomness. Not that games based on randomness aren't fun, but randomness is appropriate for some games and not for others. That's something that we saw in Rishiri with the multiple seers / guardians who never knew whether they were going to be effective or not.
    3) Hidden game mechanics. A good part of playing Phalla is controlling information. The monsters don't want everyone else to figure out what they are, and simultaneously want to confuse things for everyone else as much as possible. The specials want to figure out each other's identities without also letting the monsters know them. Being good at Phalla means being able to both hide your information and pick out other peoples' information through their lies.

    Hidden game mechanics throw a wrench in things. Hiding the number of monsters or something like that is fine, as long as it's made clear to the players this is the case. However, hiding important roles undermines players' abilities to make informed decisions, and in a sense leaves everybody playing a slightly different game. It makes the game unpredictable in a (to the player) totally random way, which ties in with my earlier point. The game becomes less about figuring out your oponents and more about figuring out the narrator.

    I won't say that hidden game mechanics should never be used. I do think that they're currently overdone, and should only be used when a role requires players not to know about it in order to be effective.
    Hidden mechanics have their place, but the fact that they are hidden needs to be considered as part of the balance. However, there's a limit. If you're going to give the bad guys specials, for example, but don't want to make it public how many of what kind, at least make it clear that there ARE bad specials.

    Nerissa on
  • DogDog Registered User, Administrator, Vanilla Staff admin
    edited March 2007
    I am interested in how you weight the roles to ensure balance, or is it just trial and error?

    Unknown User on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    Rygar wrote: »
    I am interested in how you weight the roles to ensure balance, or is it just trial and error?

    Experience plays a large part in it, since seeing how multiple games with different roles or mechanics play out gives you a pretty good idea of how things will work in future ones. For simple games a simulator can be used as a guide to show you what raw statistics indicate will happen with a particular setup, but these fail to take into account the human aspect of the game (which as I noted in my earlier post is a rather crucial part of it) and so can't give an exact answer.

    Smasher on
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    Aroduc wrote: »
    I've played a game where no fewer than 4 people claimed to be the real seer, all for different reasons entirely.

    1) Real seer
    2) Monster trying to deflect vote
    3) Villager trying to ensure someone gets staked
    4) ?

    Smasher on
  • AroducAroduc regular
    edited March 2007
    Smasher wrote: »
    Aroduc wrote: »
    I've played a game where no fewer than 4 people claimed to be the real seer, all for different reasons entirely.

    1) Real seer
    2) Monster trying to deflect vote
    3) Villager trying to ensure someone gets staked
    4) ?

    Actually, none were the real seer. First person did it to try to save themself. And it didn't work. Next person more or less the same, but it worked and got some innocent staked, and then themselves the next round. Then one monster claimed to be the seer and implicated the other, and the other reciprocated claiming to be the seer and one was staked, then the last round the remaining monster admitted that he was lying last round and got some other innocent staked to win.

    Aroduc on
  • SUPERSUGASUPERSUGA Registered User regular
    edited March 2007
    Aroduc wrote: »
    Also, this fascination the Phallas have with revealing the role of people as they die is really horrible. Half the fun is the intrigue and pretending to be something else. That's just personal taste though. It really swings things (again) back towards the village because it gives them more information and prevents monsters from lying.
    Have to say I agree with this a little, but then I think when you take this game onto an online environment it's going to be a lot more methodical and "gamey" than if you're playing with a bunch of friends at a party or something.

    Playing this at a party is something I've always wanted to try. Private communication is doable but is clear for everyone to see and this breeds delicious intrigue. Hey, Steve and Mike are both in the kitchen alone... Stake!

    They're two very different games in each environment really.

    Edit: So I somewhat misread Aroduc's quote, thinking he was talking about people being quick to set up the Seer network and such, but I still agree with his point.

    SUPERSUGA on
  • ElJeffeElJeffe Moderator, ClubPA mod
    edited March 2007
    Smasher wrote: »
    The thrall can be defended against to an extent by dividing the network into cells, but what if the thrall is the first one seered? The seer has to trust somebody to get the network started, and the thrall makes that a roll of the dice. That leads us nicely to the next issue:

    The effectiveness of the thrall is pretty heavily reduced if he's not initially in contact with the bad guys. If there's one thrall, and he doesn't know who the bad guys are, then he can be largely countered by contacting two people on the first two nights and keeping them separate. By dividing the network in two, you can be certain that one of them is thrall-free. I think thralls are a good way to counter the village's inherent advantage, and also a good way to force the players to think hard about how to set up their network.
    2) Uncontrollable randomness. Obviously there are random elements to the game; the distribution of roles is the biggest one. However, other than that I feel most if not all of the other randomness in the game should be created by players' choices. Phalla is meant to be a game of strategy and deception, and adding metaphorical dice rolls doesn't improve that in any way I can see.

    I mostly agree with this. In Newham, with multiple people in each special role, I randomly selected which person operated each night. It was sort of in theme with the setting, where everything is uncertain and chaotic, but it was a bitch to maintain. It was extra work for me, and I occasionally had to cheat the die roll to make it non-sucky for the actual specials. Otherwise, one of the seers, for example, would've gone 6 nights before he got a vision, because my random number generator didn't like him.
    3) Hidden game mechanics. A good part of playing Phalla is controlling information. The monsters don't want everyone else to figure out what they are, and simultaneously want to confuse things for everyone else as much as possible. The specials want to figure out each other's identities without also letting the monsters know them. Being good at Phalla means being able to both hide your information and pick out other peoples' information through their lies.

    Hidden game mechanics throw a wrench in things. Hiding the number of monsters or something like that is fine, as long as it's made clear to the players this is the case. However, hiding important roles undermines players' abilities to make informed decisions, and in a sense leaves everybody playing a slightly different game. It makes the game unpredictable in a (to the player) totally random way, which ties in with my earlier point. The game becomes less about figuring out your oponents and more about figuring out the narrator.

    I won't say that hidden game mechanics should never be used. I do think that they're currently overdone, and should only be used when a role requires players not to know about it in order to be effective.

    Mostly agreed, though it depends on the game, and the flavor thereof. With mine, the mood was meant to be one of uncertainty, confusion, and horror, and so I think it worked. In others, I think that full transparency is definitely the way to go. Speaking of Shinto's for example, I actually liked that he cherry picked the roles, and I like that he lied about it. If he'd said, "Yes, I picked each person specifically for his role," that would lead into a lot of meta-analysis about who Shinto would be likely to pick for each role. Thinking they were random made us abandon that line of thought, and just take people's actions at face value. I also cheated there a little in mine - I claimed they were totally random, but in truth I made sure that no special roles went to people I'd never heard of who had seemingly signed up just for the game. I didn't want to wind up with a couple of seers deciding never to post, or something, because that can ruin the game for everyone. (For the record, I won't be doing that for the next game I'm running, if for no other reason than that I've never heard of a good third of the participants. Also, one of the people who was originally a seer never posted more than once and wound up being offed for non-participation.)

    So yeah, I think a little calculated deception or opacity in game mechanics can be a good thing.

    ElJeffe on
    I submitted an entry to Lego Ideas, and if 10,000 people support me, it'll be turned into an actual Lego set!If you'd like to see and support my submission, follow this link.
  • SmasherSmasher Starting to get dizzy Registered User regular
    edited March 2007
    ElJeffe wrote: »
    3) Hidden game mechanics. A good part of playing Phalla is controlling information. The monsters don't want everyone else to figure out what they are, and simultaneously want to confuse things for everyone else as much as possible. The specials want to figure out each other's identities without also letting the monsters know them. Being good at Phalla means being able to both hide your information and pick out other peoples' information through their lies.

    Hidden game mechanics throw a wrench in things. Hiding the number of monsters or something like that is fine, as long as it's made clear to the players this is the case. However, hiding important roles undermines players' abilities to make informed decisions, and in a sense leaves everybody playing a slightly different game. It makes the game unpredictable in a (to the player) totally random way, which ties in with my earlier point. The game becomes less about figuring out your oponents and more about figuring out the narrator.

    I won't say that hidden game mechanics should never be used. I do think that they're currently overdone, and should only be used when a role requires players not to know about it in order to be effective.

    Mostly agreed, though it depends on the game, and the flavor thereof. With mine, the mood was meant to be one of uncertainty, confusion, and horror, and so I think it worked. In others, I think that full transparency is definitely the way to go. Speaking of Shinto's for example, I actually liked that he cherry picked the roles, and I like that he lied about it. If he'd said, "Yes, I picked each person specifically for his role," that would lead into a lot of meta-analysis about who Shinto would be likely to pick for each role. Thinking they were random made us abandon that line of thought, and just take people's actions at face value. I also cheated there a little in mine - I claimed they were totally random, but in truth I made sure that no special roles went to people I'd never heard of who had seemingly signed up just for the game. I didn't want to wind up with a couple of seers deciding never to post, or something, because that can ruin the game for everyone. (For the record, I won't be doing that for the next game I'm running, if for no other reason than that I've never heard of a good third of the participants. Also, one of the people who was originally a seer never posted more than once and wound up being offed for non-participation.)

    So yeah, I think a little calculated deception or opacity in game mechanics can be a good thing.

    I don't really consider either of those examples of game mechanics; those are more about role distribution. I don't have any problem with what either of you did in that regard, since while being semi-guaranteed to not get a special role is kind of a drag, it'd be worse for the village as a whole if an inactive player got it.

    I don't remember, did we have to eliminate the Cult of Hastur in Newham in order to win?

    Smasher on
  • ElJeffeElJeffe Moderator, ClubPA mod
    edited March 2007
    Smasher wrote: »
    I don't really consider either of those examples of game mechanics; those are more about role distribution. I don't have any problem with what either of you did in that regard, since while being semi-guaranteed to not get a special role is kind of a drag, it'd be worse for the village as a whole if an inactive player got it.

    Well, that's why I never mentioned it before, and why I won't do it now that I've copped to doing it before. It's only a drag if you're aware of it. ;)
    I don't remember, did we have to eliminate the Cult of Hastur in Newham in order to win?

    Yup. And they never had concrete evidence that the cult existed. I was really hoping that they'd kill the last known cultist, and then the game wouldn't end, and they'd be all, "wtf?"

    ElJeffe on
    I submitted an entry to Lego Ideas, and if 10,000 people support me, it'll be turned into an actual Lego set!If you'd like to see and support my submission, follow this link.
  • Aroused BullAroused Bull Registered User regular
    edited March 2007
    ElJeffe wrote: »
    I also cheated there a little in mine - I claimed they were totally random, but in truth I made sure that no special roles went to people I'd never heard of who had seemingly signed up just for the game. I didn't want to wind up with a couple of seers deciding never to post, or something, because that can ruin the game for everyone.
    I did the same when I randomised the roles for the commie game. It didn't matter very much, since everybody was a special anyway, but I made sure the most powerful roles - communists, reporters, moles, etc - went to people with good posting records, so that they wouldn't disappear and throw off the balance.

    I have an unrelated question: How many votes is too many? I'm working on a phalla involving elected roles, which means the players have to vote for who they want to die and also for who they want to put into power. Two votes a day players can handle fine, as we saw in Rishiri, but I'm thinking that yet another vote every third day or something would be too confusing. I'd like opinions on the matter.

    Aroused Bull on
This discussion has been closed.