Double Agent Mafia
Double Agent Mafia
Game information
Link Double Agent Mafia BTSC threads
Size Players (Medium)
Start Date 7.26.14
Winning Faction Tie Win Baddies/Goodies
Roster #, player (medium)
  1. Social Darwin
  2. Auramyna
  3. Araver
  4. Akaslickster
  5. Marquessa
  6. Coolkid
  7. Framm18
  8. Hirkala
  9. Onetruth
  10. Fat Tony
  11. GMaster479
  12. Akriti
  13. Nana
  14. ShadowAngel
First to Die Fat Tony
Last Remaining Aura, Coolkid, Onetruth, Nana
MVP Nana and Coolkid
Awards -

Double Agent Mafia was a game designed and hosted by Brandonb based on his own design.

It began on July 26, 2014 and ended in a Tie Win Baddies/Goodies in N6 (August 7).

Game Mechanics


  • Each day everyone votes for (1) a player to lynch, and (2) whether the actions will be included in the night post.
  • Tie lynch votes will result in all tied players being lynched.
  • There are 14 players. The game will start with 1 baddie, who will recruit another baddie from the roster, and they another, and maybe another, etc. until the baddie BTSC reaches up to 4 players with a balance weight of up to 20.
  • The game will start during the day phase (D0), during this time everyone will vote for whether to display the following night actions in the following night post.
  • If the actions are voted to be shown, then the they will be automatically omitted from the following night post following that. (E.g. if actions are shown N1, then they will not be shown N2, and a vote may then be held again to determine N3. If actions are not shown N1, then a vote may be held D1 to determine whether to include them in the N2 post).
  • If actions are shown, all targets of specified abilities will be revealed in the NP. If actions are not shown, only successful kills will be shown (though blocked players, saves, and spies will be informed via PM with their results). Vote manips will obviously be apparent if they make a difference in the outcome of a lynch.
  • Please post all questions and/or comments to the host in RED

Role Description

There are two of each of the following roles

  1. Block - Stops any player from using their ability (that same night or the following day if targeting a vote manip).
  2. Save - Saves any player from being killed at night.
  3. Kill - Removes a player from the game, unless killer is blocked or target is saved.
  4. Spy - Learns the faction of the targeted player
  5. Vote Manip - PMs the host to change a targets' vote during the day (lynch AND/OR NP content vote)
  6. Kamakaze Bodyguard (KBG) - Chooses a target each night to bodyguard. If the target is successfully targeted for a kill, the KBG dies in the target's place and also kills the attacker.
  7. ODTG - Choose any of the above 6 abilties, but must choose at night, and may only use a given ability ODTG.

Q: Is the "D0" vote only for the NP content?
A: It is for player confirmation, to vote for/against NP content on N1, and to allow time for baddie recruitment.

Q: Is baddie recruitment 1 per night?
A: Baddie recruitment starts on D0 and must be finished by the end of N1.

Q: How does recruiting work?
A: The baddies effectively just choose their recruits and build their own faction. Like I said, this is quite experimental.

Q: What happens if the NP Reveal vote is tied?
A: A tie NP vote will show the NP content.

Q: What actions show in the NP, when the content is revealed?
A: EVERYTHING is revealed. Save targets, block targets, the whole show (except the vote manips of course).

Q: What roles are given action results, when the content is revealed?
A: The spy result... That's the only thing that won't already appear in the NP.

Q: What roles are given action results when the content is NOT revealed?
A: If actions are not shown, only successful kills will be shown (though blocked players, spies, and successful saves will be informed via PM with their results).

Host's Summary

Game Idea and Expectations


Haha! It seems that Araver and CK have broken down a lot of the game analysis that I was planning on commenting about. But, here are my thoughts, even if some of it may be redundant…

Obviously inspired by Agent of Shield, I really wanted to create a game where, at least in my own mind, all of the players are on one big team. As if in one big spy unit they all know each other. They know each other’s identities, specialties, and behaviors. So how does an enemy faction infiltrate this organization? Or hide among them? Well, look at Agent of Shield and see that they bad guys tended to recruit the good guys. They lay low and tend to not take many aggressive actions, even though they may possess the capability. And they put their own actions on display with the rest of the good guys... that is, if they’re smart. Keeping this in mind is what really drove the development of this new experiment. But obviously one of the biggest hurdles to overcome when so much information is out in the open, is how to balance the roles

Role Balance:
First and foremost, from the very beginning, I was interested in how well this game would align with my role weighting method of game balancing. I was initially afraid that the spies were weighted too lightly since everyone knew who they were, that the saves and KBGs were too light because they knew the roles they were saving, and that the kills were too heavy because, given the potentially miniscule amounts of info if the baddies played properly, they might never be able to act for either side. But then I thought, knowing the roles doesn’t really help the saves or KGBs too much, since only the faction matters. And the issue with the kills didn’t matter either since, in a regular mafia game, if information gets out about a good target, the lynch almost always handles it before the killer can take care of it. As for the block, I figured that it has the advantage of knowing its target all the time, however, this means that it mostly loses its secondary spy effect since the only additional info it could reveal is the NK.
However, as for the spy… in a regular game the spy will not usually out themselves in order to clear someone, and sometimes may hold on to a baddie ID for a day or so just to analyze behavior. But in this game, where everything is visible and there’s actually an incentive for them to speak up about both outcomes, the spy was a potentially huge issue. So, I decided to try and alleviate part of the potential problem by forcing one of them upon the baddies. I flipped a coin between FT and Nana to see which one of the spies would be the first baddie to begin the recruiting process… I figured that if the spies were too light for the game they would be more beneficial to the goodies, and at least this way the baddies would have one of them. Would the game have worked if I had made the first baddie truly random? Looking back now I think so, because the recruitment impacted the game in a counter-balancing way that I didn’t initially expect to this extent. Given the amount of weight the baddies could take on… if the baddies took all of the aggressive killing roles, then they were neutralized by the saving roles and subject to the info roles. If they took saving roles then they would have less attack potential and again be subject to the info roles. And if they took the info roles, then they would be weaker all-around on attack and defense. Overall I’m very happy with the way this game played out.

Player Balance:
I was also a bit worried that this game might become unbalanced given that the baddie recruiters might focus on a particular type of player. So it seemed that Greeks were going to be necessary to help and alleviate this potential problem. But then I remembered that this was an invite only game for those players that I really trust to be active and meaningfully contribute to the game. In the end I was very happy with the activity of everyone, and I think over the course of the entire game I only had 1 night action that wasn’t submitted.
Though I have to say, I was a bit disappointed to see FT go so early (as he would’ve been very helpful for the goodies), but then again, he was the goodie spy and therefore a very very preferential N1 target. Overall I think that the player balance wasn’t a huge issue since it was clear to me that everyone, baddies and goodies alike, were very pessimistic. Both sides believed that they were going to lose. I do wish that everyone was optimistic to make for what I think would be a more enjoyable game, but the frequent pessimism from everyone did strangely calm my concerns. Tongue

Role Disclosure:
Clearly the separating feature of this game is that all roles were disclosed. It had a very interesting effect on the game. On the one hand, the baddies couldn’t manipulate the goodies as much by claiming other roles. But on the other hand, goodies couldn’t out themselves to sway the votes away from themselves. But back on the first hand, the level of NP disclosure meant that the baddies couldn’t be too cavalier with their night actions. Then back to the second hand, being conservative is how they go without being discovered, and really went towards my desired double agent theme for the game.

Night Posts:
When so much info is given out at the beginning, there’s really not much to go on for the rest of the game except for the disclosure of night post info. This was again a hurdle that presented itself. What abilities to put in the NP? Will it prejudice the goodies if the baddies recruit based upon those roles that don’t appear? If I make all of the roles appear then will it prejudice the baddies? So my answer was to put all AND none alternating. But then, will I prejudice the goodies if I assign them to show on even nights or odd nights? So I figured that I should let the players decide which nights were which. But if the sequence was effectively decided on the D0 cycle, why bother to continue to ask the players to vote in the upcoming days instead of strictly setting the NP info cycle? Well, the ideas was that because the D0 vote would likely be decided in what was perceived as the best interest of the goodies, a later failure to vote would likely help to counterbalance in favor of the baddies. Clearly that never occurred, and I think in the future it would be best to just do the D0 vote and then set it permanently from there.

I'm very happy with the way that this game played, and how it ended, as almost everything on both sides proceeded equally miserably for the players but also kept me on the edge of my seat! Haha, I don't know who had more fun, you all or me! But I do want to apologize AGAIN for the lack of production value. I really wanted to, and envisioned, making the posts much more fun. Now, as for MVP... Do I give one to both sides? Do I give more than one? Can I give several? Everyone played so :fbomb:ing well. And even though I can narrow it down somewhat I'm honestly having a terrible time between Aura for playing an aggressively convincing goodie in the face of her allegiance and her role, CK for running the goodie strategy and analysis so well, Nana for bending and/or creating the flow of information in the game thread to her advantage, Onetruth for her activity and for most important game-saving action (though she was directly saving a baddie, the result was that she saved two pivotal goodies), and Araver for completing the baddie btsc dynamic by arguing, predicting, and analyzing the potential outcomes for their planned actions each night.
But I need to make a decision, and I don't think 5 would be a good thing for various reasons. So I guess, given the game's deficiency of information, I'm going to give them to the two players that really controlled it in the game thread. Nana and CK. Clap

Thanks again everyone for playing! I really hope that you enjoyed this as much as I did! And PLEASE, if someone else decides to host this game format in the future, let me know! I would really like to try it!

By Brandonb

Special commendations

Host disappointments

Alternate endings

Winning Faction

Day and Night Posts

Intro N1 D1 N2 D2 N3 D3 N4 D4 N5 D5 N6 Post-N6

End of Game Roster

  1. Social Darwin - Block - Lynched D2
  2. Auramyna - Kill
  3. araver - ODTG - Lynched D5
  4. akaslickster - Kill - Lynched D1
  5. marquessa - ODTG - Lynched D3
  6. coolkid - Block
  7. Framm18 - Vote Manip - Killed N4
  8. Hirkala - Vote Manip - Killed N4
  9. onetruth - Save
  10. Fat Tony - Spy - Killed N1
  11. GMaster479 - Save - Lynched D4
  12. Akriti - KBG - Lynched D5
  13. Nana7 - Spy
  14. ShadowAngel - KBG - Killed N3


Ad blocker interference detected!

Wikia is a free-to-use site that makes money from advertising. We have a modified experience for viewers using ad blockers

Wikia is not accessible if you’ve made further modifications. Remove the custom ad blocker rule(s) and the page will load as expected.