|
![[image loading]](http://imgur.com/jYef5.jpg)
What I learned from this game in terms of modding:
1. Majority Lynch is the way to go. Too many times a player can't vote because of an event, a real life occurence, or a legitimate excuse. I propose a new rule for inactivity modkills. Voting is done by majority lynch and a player must post at least once during each full cycle (night and day). This gives each player a very long time to post. Inactivity with votes isn't a problem to the game, but people who don't post in the thread hurt the game. A player who almost never posts in the thread but always votes is less valuable to a player who is active but forgets to vote IMO.
Any player who does not vote once during the game will be put on probation and cannot play in the next mafia game.
48 hour days are simply WAY too long late in the game when the most active posters are dead. People lose interest as no content is posted for a long 2 days and games start to flatten out. Haunted Mafia had a HUGE amount of modkills. I think majority lynch, revamped inactivity rules, and a bigger crack down on spamming will help reduce this number significantly. There should be regular micro mafia games for newbies (with a very simple set-up such as 6 town 1 cop 2 mafia) that gets them warmed up to the game. Newbies jumping into a huge game like Haunted Mafia might be intimidated and it leads to modkills.
Majority lynch also opens up the possibility of roles concerning the "Hammer" which is the vote that pushes a player to the majority and ends the day, killing them. This will make the way voting works more interesting and make vote analysis a much more relevant part of the game.
2. Communication with co-host is the key to success as a mod. My communication with Meeple was essentially non existant. My suggestion is for mods to use QuickTopic to share information (post all night actions, voting records, and relevant information to the forum). QuickTopic is very difficult to penetrate with google and it is a relatively safe way for mods to share information easily. I also suggest QuickTopic for scumteams.
Every host should also have a back-up cohost who they send all information to in the case that a cohost is inactive. Mods should not use pastebin to store information.
3. Cracking down on spamming/flaming is necessary. Modkilling is never fun but it is necessary. I propose a simple policy. Double/triple posting ONLY in the case that the posts are game relevant post. Consecutive non relevant posts would constitute a warning, a second offense would constitute a ban. In games that move very quickly, players have a hard time catching up if they aren't very active on TL. A lot of information gets buried under pages of spam and players are discouraged to read everything they missed when there is too much irrelevant posting.
4. Smurfs must confirm their real identity to the mod. In Haunted Mafia, I gave BloodyC0bbler information I should not because I was not aware he was using the smurf name "LastArgument" in my game. Annul also confirmed himself as Count Dracula to BloodyC0bbler in irc because of smurfing. If you're going to smurf, smurf responsibly. Tell the mod.
Players in the game, don't reveal information to players outside of the game under any circumstance due to the possibility of smurfs.
5. Sixty-one people is too much. I think somewhere around 40 should be the hard limit for a game. It's too easy to make mistakes in counting, it's too many people to keep track of, and it welcomes a lot of inactivity. It was a bad idea to hold a game that big. I bit off more than I could chew.
6. PM's are a good thing. They don't make town OP, they make mafia very powerful if they understand and use the godfather right and I think mafia can too easily take advantage of a town that can't communicate reliably. I don't like the "No PM" trend.
7. Clues bad. Elections good. Clues rely on players to spice up their profiles well. Some players are too hard to make difficult clues for, others have profiles that are too obscure. Look at the two extremes in haunted mafia. Masq who was very obvious and SouthRawrea who wasn't anywhere close to being caught on a clue at any point o the game.
Elections give a lot of discussion for Day 1 and they prevent the game from having an RNG stage with IMO is really lame. I think every game should have a gimmick that creates Day 1 discussion beyond "which inactive should we lynch". You could get creative with this.
Haunted Mafia Analysis:
Mafia started off very well. NB presented himself as a somewhat capable leader behind the scenes and their hits were pretty spot on. Ghosts blocked 2 of the vampires hits and a vampire was lynched on the first day.
The problem with Mafia is they had no real leader. BrownBear had the capability to carry Mafia but his life was cut down fairly early and eventually it all fell on the shoulders of the novice SpydR. While I'm impressed with SpydR's play he didn't quite have what it took to lead a scum group.
Zerroth's errors ended the game for the mafia. He roleclaimed at time when it was basically guaranteed vampires could kill him in the night by stacking hits. Inactivity was a huge problem for the mafia as well, they had more modkills than the vampires did. Really everything was stacked against the vampires over the course of the game. Both major mistakes I made hurt the vampires badly and they had a nearly disastrous early game.
What saved the Vampires? Annul understood his role as Count Dracula, NB did not properly play the Godfather. The Godfather role is a role that is mostly powerful in games with PMs. No PM games heavily nerf the godfather role as the main advantage the godfather has is that he can reliably infiltrate town circles. NB didn't infiltrate the town circle or use his role for manipulative purposes, annul was very calculating and believable in his fake jack o lantern role.
I set up the game so that Mafia would have the advantage in manipulation and that vampires would have the advantage in killing abilities. Poison, an extremely useful role (as it can be used to both manipulate and get safer hits in) was neglected largely by the vampires who wrote it off too quickly as useless (mostly because QuickStriker was confused to how it worked).
The game ended up being the opposte. Vampires outkilled the mafia and outmanipulated them. BrownBear didn't really have the time necessary to take full advantage of his framer role (a role that is EXTREMELY deadly late game), but managed to get a frame off on Pandain which was crucial as he was checked by a DT and visited by several other roles that night.
Zerroth was hit and miss with his roleblocks, mostly hitting weaker roles like boogeymen. Speaking of boogeymen, the boogeyman role is a town circle role. The boogeyman must be heavily involved in the town circle. The boogeyman can not be tricked or bypassed unlike the DT and can be used to confirm whatever bullshit that godfather roles/fake mafia spew. If annul said "i placed a bomb on this person" and one of the boogeymen knew for a fact that he hid in annuls closet and that didn't happen he has very strong evidence that annul is a mafia.
Detective play wasn't excellent in this game. Aztrorisk had some good checks but KtheZ was an inactive DT, which is a useless DT.
What strikes me as interesting is Veldrils lynching and this spoke to me a lot. I had never been town at this point and it wasn't until I saw the game from a neutral position that I really saw how easily town can make such awful mistakes. Veldril defended Masq very very very lightly, his initial defense of Masq was barely anything.
Now all things being considered Veldril was a brand new player. Brand new players are typically scared to take a strong position and are scared/wary of things like bandwagoning and the intensity with which mafia posts/accusations often are often rife with.
Afte rlynching Masq, it was VERY easy for mafia to say "oh well that means Veldril is red too." One thing I suggest town be wary of it is alliance type posts. Really analyse them. It is legitimate to point out scummy defenses/possible alliances but mafia also want to create fake alliances to incriminate a player with a bus, to take someone down with an inevitable lynching that would occur (this is what happened in the vampires case), or to make two townies look like a group of 2 mafia or a group of 1 townie and 1 mafia.
Look at how a player defends another player. Does that player defend the player by demonstrating how the attacks against him are unreasonable? Or does he defend through methods that are not based in reason (attacking the attacker, appeal to emotion, etc.). Ask yourself "What is the defending players goal here? Is this a defense a townie might make?"
Another interesting thing was the insistence that blue roles were really in any danger. Role claims should have been happening much more in Haunted Mafia than they were. Mafia don't want to kill blue roles they want to USE the blue roles AGAINST the other scumteam. Blue roles help scum as much as they help town if the scumteams conduct their business well. Annul was the one who realized this and his focus on the mafia and using the blue roles as a crutch led them to victory. The mafia's strategy seemed more scattered and their insistence on hitting problematic townies was a mistake.
Mistakes the town made? There was one big one. Lynching SpydR. I mean seriously what the fuck? You know there are 2 mafia left. You know 1 of the mafia is SpydR. But you lynch him, making sure the vampires are GUARANTEED to win. The town created a scenario for themselves in which they couldn't win and they are the ones to blame in this case. SpydR shouldn't have roleclaimed either IMO. The fact that he did makes the towns actions all the more confusing.
The towns clue analysis was much too literal/rigid. I don't like clues anymore, it's too hard to see how different mods make clues and it's just far too subjective.
It was an interesting game and there is probably a lot more to be said about it (particularly from people involved in the PMing action).
|
I don't like PM games not because they make town OP but because they make scum OP.
There's always the people that have the right intention but wrong execution who try to make a PM circle or join one (this is you Pandain). They take discussions that SHOULD in every sense be public into PM land and hide a lot of what they say. This ends up being stupidly suspicious if it's revealed but they're just plain townies in the end. Also, because of these proactive players, it's very easy for scum to join a PM group with the absolute trust of everyone.
PM land is stupid and detracts from the game. I like the private type of PM land that's in Insane Mafia where not every idiot gets to PM others. Maybe you could try a system where Mayor gets to pick a person per night to join the circle or something? Idk, it would spice up the Mayor role too.
|
On November 01 2010 15:01 youngminii wrote:
I don't like PM games not because they make town OP but because they make scum OP.
There's always the people that have the right intention but wrong execution who try to make a PM circle or join one (this is you Pandain). They take discussions that SHOULD in every sense be public into PM land and hide a lot of what they say. This ends up being stupidly suspicious if it's revealed but they're just plain townies in the end. Also, because of these proactive players, it's very easy for scum to join a PM group with the absolute trust of everyone.
PM land is stupid and detracts from the game. I like the private type of PM land that's in Insane Mafia where not every idiot gets to PM others. Maybe you could try a system where Mayor gets to pick a person per night to join the circle or something? Idk, it would spice up the Mayor role too.
I understand where you're coming from.
I have an idea from a game in where there are separate PM circles each with an even amount of mafia that aren't aware who is mafia in the other PM circles and they all choose their kills separately.
That is actually a really interesting consideration that you brought up with the mayor role ;o I've always felt it's a really boring and underpowered role considering how much work someone has to go through to get it.
|
Qucktopic would've definitely been a good idea... I was too cautious about stepping of DrH's toes about asking about actions and all that. It definitely would've helped tie us together.
|
Even prospect of managing & playing with 40 people is personally terrifying in my opinion -_-;; idk, doesn't seem to find any good small <9 roster setups being run recently.
Edit: Oh wait, it just got buried. Either way I really don't like massive roster setups, not only it increases the commitment required for individual members it just introduces too much complication I think. Of course, you really wouldn't know & pay attention to anyone in real town either, but this is game we are playing anyway xD
|
Well... in hindsight we might've split this game into two games, one run by DrH and one run by me. And clues can be handled if the start date of the game is delayed until everyone has a suitable profile. There were several people that never really had good enough profiles, despite PM's from me.
|
Even if I didn't participate in this game I still think it's really great for DrH to take the time to write this out and share his experience.
If there ever is another Haunted Mafia game, I think I will join now that I have spectated one.
|
I'm already planning christmas mafia with some really unique roles ;o
|
I had a long post all typed up but the file is corrupted!! 
Anyway, I guess I'm too lazy/frustrated so I'll just post an abridge version of my thoughts here.
Part 1 discussed why I think majority lynch is not the answer. The basic points are that majority lynch favors impulsiveness at the expense of good analysis. You make the point that Majority lynch gives people a long time to post. This is true; it gives people as much time to post as in a normal game. However, the issue is that your posts may not count. Majority lynch means that a day can be decided before someone gets a chance to speak. Decisions about a lynch can be made before some people have a chance to get online and present their information. Since this is an international forum, this problem can become a big issue because of time zones. Secondly, requiring posting but not voting doesn’t solve the problem of real life issues. It is highly unlikely that a person who can’t appear to vote will have time to appear to post. Majority lynch doesn’t really solve the modkill problem.
It is true that the game slows down later in the game where active players are dead. However, I think 48 hour days are still necessary given time gaps. As mentioned above, people need a chance to speak and get their ideas out there, even if there will be less discussion about it. Just because there is less discussion doesn’t mean that there’s nothing worth discussing. All discussion should be valuable to the town.
Part 2 discussed the problems with regulating post behavior. Spam posting is a legitimate scum tactic, while a debatable town tactic. If this view is taken, regulating post behavior gives away information. It is true that spam causes disillusionment as people cannot keep up with the thread. However, that is probably a problem with people thinking they need to respond to every criticism thrown at them rather than any problems relating to being able to post freely. Not sure if there is a real solution to this other than to convince people that it is unnecessary to respond to every point thrown at them.
Part 3 discussed game size. I mention that in practice (through experience), optimal game size is around 30, which is a good balance between activity and role balance. Once you get fewer players, there are less options for role variety (unless it’s a crazy theme game) that maintain a good balance for both mafia and town. Once you get any more players, inactivity goes rampant even though theoretically game balance is easier. Plus, its harder to mod a large game rather than a smaller game. I can say more on this if people are curious. Otherwise, I’ll just keep that in the balance team world since its not very applicable to playing the game.
Part 4 discussed PMs. There is a big debate to this on both sides. I think DHs analysis is slightly off here. From a practical/realistic point of view, PMs are a double edged sword and benefit skill. Skilled town players and skilled mafia players can both use PMs to their advantage in order to get information from those who do not know how to use them. Contrary to belief, PMs benefit town immensely in terms of analysis. PMs allow townies to pressure other players in ways that cannot be done in thread. PMs give players the ability to circumvent the game atmosphere, whether it is inactive or spammy, in order to get people to respond. Give PM powers to a town-aligned Ver or Qatol, and stuff happens. Quickly.
A godfather with PM powers is potentially powerful. However, I think its overstated. In a 30 player setup with traditional roles, it is difficult for a godfather to pull off a blue claim effectively. Claiming DT is full of traps because in order to pull it off, the GF must accurately guess a players role, whether DT, medic, vig, etc., or townie. Claiming Vig isn’t very powerful because typically vigs aren’t given blue role information. They’re usually treated as pet that must be order to perform an action. Mafia can’t pull off the vig hit without making themselves look suspicious. Hatter is very difficult, especially if there is a real hatter in the game. Claiming hatter in a 30 person game should be immediately countered by the real hatter placing a bomb on the claimed hatter. There’s no way for the GF hatter to pull it off, because the real hatter should be asking to get lynched. GF can’t repeat that feat because getting himself lynched would result in no gain, while a hatter getting themselves lynched would net the town the GF. Medic is an easier claim, since you can get away with understandably not proving your role for a long time. Problem is, if the town can direct the medic’s actions without telling the medics the blue list, mafia just lost a hit option. Town directing the GF-medic’s hit does nothing for the mafia except prohibit them from hitting the target they are supposedly supposed to protect. Which effectively gives town an extra medic save and prevents mafia from killing high profile town players. If town plays correctly, faking medic is not a good idea for the mafia.
In essence, the power of PMs is magnified by variances in player skill. PMs allow the strong to prey upon the weak. Regardless, PMs help increase strategic depth for both sides. In a high skill game, however, PMs favor town more than mafia. Reality, on the other hand, is very different.
Part 5 discusses the assertion about clues and elections. I’d agree that clues have huge volatility based on profiles. They do provide a good discussion point for day 1. However, I’d argue that this discussion is highly misleading for the town and is favorable for the mafia, given that it allows them to be active while not contributing. Overall, clues are pro-mafia in the early game, and pro-town in the later game when clues stack up. Given that high profile players die in the earlier stages of the game, the existence of clues clouds up the early game and likely benefits the mafia.
Do elections really provide an adequate discussion point for day 1? Upon examination, elections usually revolve around plans revolving around blues. It is very rare that an election will revolve around good scumhunting. If it does, it is more based on name experience than actually day 1 scumhunting. Regardless, an election that revolves around a plan usually needs an element of creativity. An element that is present not because of the election, but because of some game mechanic. Which certainly can be discussed without an election. Election simply gives a platform for people to discuss plans. Plans have been known to be crafted even without elections, which makes me think elections are superfluous. Games can exist without an RNG stage. You just need to have people focus on analysis or plans. Preferably the former. Regardless, RVS occurs more frequently in smaller games where there is less strategic variety and there is nothing novel to talk about. Large theme games already give enough discussion points where elections are unnecessary because day 1 revolves around plan and setup discussion rather than random votes. Basically, elections are not what solve the problem. Hard core analysts and novel setups/roles provide things to talk about.
|
clues are awesome
|
hey i remember this game. it was back when glasse wasnt completely useless
good times good times.
|
actually i think the way glasse is playing might be very smart if hes doing what i think he's doing
;;;;;oooooo
|
On November 01 2010 20:35 Coagulation wrote:
hey i remember this game. it was back when glasse wasnt completely useless
good times good times.
I say useful stuff once in a while
im like 3 faced
1/3 normal cat
1/3 crazy cat
1/3 serious cat
|
i think we should keep insane mafia to insane mafia ;o
|
true
:3
haunted was fun though
i think 2 red faction = awesome
|
what if everyone was a scum faction thinking they were the only scum faction fighting against a town but it was actually just 4 scum factions or something and there is no town at all
how long would it take to figure that out
|
well i could come and quote this post
|
On November 01 2010 21:28 Glasse wrote:
well i could come and quote this post
fair enough
|
|
|
On November 01 2010 14:51 DoctorHelvetica wrote:
1. Majority Lynch is the way to go. Too many times a player can't vote because of an event, a real life occurence, or a legitimate excuse. I propose a new rule for inactivity modkills. Voting is done by majority lynch and a player must post at least once during each full cycle (night and day). This gives each player a very long time to post. Inactivity with votes isn't a problem to the game, but people who don't post in the thread hurt the game. A player who almost never posts in the thread but always votes is less valuable to a player who is active but forgets to vote IMO.
Any player who does not vote once during the game will be put on probation and cannot play in the next mafia game.
48 hour days are simply WAY too long late in the game when the most active posters are dead. People lose interest as no content is posted for a long 2 days and games start to flatten out. Haunted Mafia had a HUGE amount of modkills. I think majority lynch, revamped inactivity rules, and a bigger crack down on spamming will help reduce this number significantly. There should be regular micro mafia games for newbies (with a very simple set-up such as 6 town 1 cop 2 mafia) that gets them warmed up to the game. Newbies jumping into a huge game like Haunted Mafia might be intimidated and it leads to modkills.
Majority lynch also opens up the possibility of roles concerning the "Hammer" which is the vote that pushes a player to the majority and ends the day, killing them. This will make the way voting works more interesting and make vote analysis a much more relevant part of the game.
I disagree completely. While majority lynch may speed up the pace of the game more, it does nothing to fix the inactive/lurker problem. If a player can't vote because of an event that they should have seen coming, or they just don't post, they shouldn't have played in the game. It is downright disrespectful to the mod and every other player that they DID sign up for the game, and they deserve to be modkilled and banned. If there was no way they could have foreseen their problem, they should just PM the mod when they can. In most games, a mod PMs the players who are in danger of being modkilled. They really shouldn't be able to forget to vote. Real life happens, but I don't think going to majority lynch does anything but bring back the atmosphere of general inactivity that was rampant in the games before we started modkilling for inactivity.
Also, Incognito brings up a good point. What if the Australian Detective rolechecks a red but the cycle happens to start at 3 am his time? He may not even get to post the results of his rolecheck by the time the lynch is decided, especially if there is momentum to lynch someone based upon the previous day (think about Veldril's lynch).
If you want to speed up the game later, consider changing things to a 24 hour day. You can also use the LSB model and not have a night phase at all.
Haunted mafia had more new players than any game since Incognito/dreamflower and BC hosted Mafia XVI (which had approximately the same amount of players and 2/3 as many modkills combined between the 2 games). It is natural to see more modkills because you have people who just aren't used to how mafia works. You also have people who decide that mafia isn't for them and they move on.
2. Communication with co-host is the key to success as a mod. My communication with Meeple was essentially non existant. My suggestion is for mods to use QuickTopic to share information (post all night actions, voting records, and relevant information to the forum). QuickTopic is very difficult to penetrate with google and it is a relatively safe way for mods to share information easily. I also suggest QuickTopic for scumteams.
Every host should also have a back-up cohost who they send all information to in the case that a cohost is inactive. Mods should not use pastebin to store information.
Agreed, especially in the bigger games. It really helps to have a second pair of eyes on everything, and splitting up the modding work makes the game less of a constant obligation for the host.
3. Cracking down on spamming/flaming is necessary. Modkilling is never fun but it is necessary. I propose a simple policy. Double/triple posting ONLY in the case that the posts are game relevant post. Consecutive non relevant posts would constitute a warning, a second offense would constitute a ban. In games that move very quickly, players have a hard time catching up if they aren't very active on TL. A lot of information gets buried under pages of spam and players are discouraged to read everything they missed when there is too much irrelevant posting.
We have been talking about this for a while now, but haven't been able to quantify it well. Part of it is it can sometimes be hard to figure out what is a game-relevant post (obviously it is usually easy, but sometimes people are trying to hint at something game-related with the hope that it looks like spam to most players). The idea I have liked the most for this is to say you can't post more than 3 times in 20 minutes (or change up the numbers a bit if you don't like them). But it obviously doesn't stop spam posts at all.
Flaming modkills should be relatively few and far between. People who do it will either learn their lesson or get banned and move on.
4. Smurfs must confirm their real identity to the mod. In Haunted Mafia, I gave BloodyC0bbler information I should not because I was not aware he was using the smurf name "LastArgument" in my game. Annul also confirmed himself as Count Dracula to BloodyC0bbler in irc because of smurfing. If you're going to smurf, smurf responsibly. Tell the mod.
Players in the game, don't reveal information to players outside of the game under any circumstance due to the possibility of smurfs.
Agreed. This was dumb on BC's part. But I'm just not a fan of smurfing anyways.
5. Sixty-one people is too much. I think somewhere around 40 should be the hard limit for a game. It's too easy to make mistakes in counting, it's too many people to keep track of, and it welcomes a lot of inactivity. It was a bad idea to hold a game that big. I bit off more than I could chew.
It was nice to see how a larger game would work now that we worry about things like activity. We haven't had a game this big since Mafia VII. That being said, I agree that this game was too large.
6. PM's are a good thing. They don't make town OP, they make mafia very powerful if they understand and use the godfather right and I think mafia can too easily take advantage of a town that can't communicate reliably. I don't like the "No PM" trend.
Agreed. I think PMs are incredibly useful tools if used correctly. The game I played without PMs was incredibly frustrating. I think it helps a lot to be able to coordinate things outside of the thread and to follow up on suspicions with people so you have a stronger argument when you post. Plus I feel that it cuts down on the spam a bit because people can get it out of their system via PM.
7. Clues bad. Elections good. Clues rely on players to spice up their profiles well. Some players are too hard to make difficult clues for, others have profiles that are too obscure. Look at the two extremes in haunted mafia. Masq who was very obvious and SouthRawrea who wasn't anywhere close to being caught on a clue at any point o the game.
Elections give a lot of discussion for Day 1 and they prevent the game from having an RNG stage with IMO is really lame. I think every game should have a gimmick that creates Day 1 discussion beyond "which inactive should we lynch". You could get creative with this.
I disagree. I think clues have their use as long as they are handled responsibly by the town. The way I see them, they give people something to talk about on day 1 and provoke some discussion by would-be accusers and would-be accused. That is usually enough to allow behavioral analysis and catch a few scum early in the game. After the early discussion, I basically don't think clues should be used much until they start repeating, which is usually day 3. After that, I feel like the arguments for a lynch based upon clues become legitimate.
The main reason why I don't like elections is because they usually mandate some sort of bodyguard role. I personally think the bodyguard is the most useless/ boring role ever invented. I guess I would be okay with elections if they didn't involve bodyguards, but I'm still not entirely sure it wouldn't still be preferable discussion-wise and environment-wise to have clues.
|
|
|
|
|
|
EPT
