Postgame
Apologies to those that this postgame disappoints - it will primarily be focused on the design of the game, philosophies behind things, an analysis of what worked and what didn't, and then a brief rundown of what happened and some notable plays that occurred.
Design Philosophy - General
I wanted this game to allow for the maximal possible skill expression regardless of role. I know that this became a meme, but I did honestly want to set out towards this aim. Towards this end, I settled on an (almost) totally split mafia similar to Oberon in the game Resistance. I felt that this setup allowed for the most possible plays, both from a village and from a mafia POV.
I also knew that I wanted the role strength in the game to be fairly high, and resilient to bad luck on all fronts. I knew that comeback mechanics had worked well in Metroid Prime 2 and decided to implement those pretty strongly in this game.
After last game, I liked the idea of including recruits so that the Cartel and Psycho could patch whatever holes they felt they had in their lategame planning / role strength. I also thought this would make the village play a little more flexibly and be a bit easier to mole / spread paranoia amongst, which were both themes I wanted to explore.
Finally, I wanted there to be some additional neutral roles that had to perform a balancing act between the major factions (being the Big City, Cartel, and the Psychopath). As such, I made three other neutrals: The Assassin (the go-between between the Village and Cartel), The FBI Agent (Village and Psycho), and The Anarchist (Cartel and Psycho). In the end, I made the FBI Agent and the Anarchist overly friendly towards the main factions and not enough towards the Psycho, so neither really fulfilled their original intent in that regard. The Assassin played pretty much as I had hoped it would, thankfully, and ended up pretty actively playing both the mafia and village without ever taking too strong of a side with or against either one.
Design Philosophy - Big City
I had seen most big villages win simply by having a large number of non-interactive, uninteresting roles that acted as an amorphous blob that overwhelmed bad mafias. If those villages ever lost majority or got into a tricky endgame situation, they invariably lost due to having zero unknown / unexpected role power and no ability to make a comeback whatsoever. I thought that was horrible and set out to design pretty much the opposite of that. I really tried to make the village roles strong and interactive, and have a lot of cool / unique connections for players to explore.
There was a lot of consternation, however, about how I balanced that - I set up the numbers to be balanced around the village having effectively 11/6 kills per cycle (1 vote + 1/3 jail + 1/2 vigilante) at base, and the mafia + wolf having 2.5 kills per cycle. I then assumed village would have one misvote and two recruits against it and then calced a ratio of 2.5 / 1.83 from the result. This is all a long version of explaining how I originally (when the game had 50 players) arrived at 31 villagers (including Ditto) and 17 mafia (including Celever) - 31 -> 28 villagers minus two recruits and a missed vote, 17 + 3 (Psycho and two recruits) = 20, and Assassin I figured would be about 2/3 of a Cartel and 1/3 of a villager - the result was a ratio of 20.66 to 28.33, which is a ratio of 1.37, or almost exactly 2.5 / (11 / 6). I threw in one goon and one villager with the change to 52, not a huge difference.
Now admittedly the 1/3 jailkeeper wasn't a 'real kill', and the village suffered from a number of undercooked roles that I needed to put a bit more time / thought into. I'll list my design concerns with the other factions later, but strictly from a village perspective the changes I would make if I could make a 'patch notes' would be:
- Include one additional villager and one less Cartel goon
- Have that additional villager be a backup Jailer called the executioner, who kills jailed goons after N cycles (maybe 3 or 4)
- Include the steadfast and miller text in the OP so that other factions could more effectively fake claim
- Update the Lawyer role to be a forger so that the village would have a way to hide their best roles from mass inspection by the mafia
- Move the one-shot redirect back to the Mayor, which is where it started and never should've left
- Update some of the shitty village roles to be stronger at base form (Conspiracy Theorist, Secretary pre-upgrade, most of the cops, and The Poll Worker all needed to be stronger / more interactive)
- Make a real role as the 52nd role in the game instead of The Mime, I honestly just winged it because we had two extra signups
- Tone down the banker and make the extra life an active, to make it clear that the wolf (or others) would be able to absorb it
- Probably some other stuff I'm forgetting
Actual Play - Big City
The Big City played well in general, though not as well as they think they did. DLE did a great job of moling the Cartel early and getting some valuable information, and Yeti / Dak did a great job of rounding up the villagers and keeping things organized without sheeting any moles.
Their problems lay more in execution than anything else, however. Yeti / Dak got fooled by the Psycho kill frame, which admittedly is a tough thing to guess but in my opinion they had better options than blowing a vigilante shot on it. They also weren't explicit enough about keeping LonelyNess in the dark, and so LN got a LOT of village info for free just from talking to people that he probably should not have been able to get.
Finally and most importantly, their night actions left a lot on the table. I mean this from the village as a whole's perspective, not from the perspective of what the village leaders were ordering. Bus driving Blazade into the kill from A Fairy on Night 3 was the most crucial mistake they made in this regard, but they underutilized a lot of roles that had much more potential than was realized, including:
- The Sacrifice (Redless)
- The Curator (Martin, who probably should've been targeted by The Sacrifice)
- The Journalist (LightWolf, who primarily was used to just check goons when he could've been Psycho hunting)
- The Mime (idled) and The Mimic (waited three nights then just copied M2H anyway, who was available way earlier)
- The Escort (roleblocked Ditto / The FBI Agent, and only roleblocked / strength sapped Pennsylvania on the Cartel over the course of like six nights alive)
Basically, it felt like the village night actions were never anywhere close to being maximally efficient. Some of this is just the nature of being a Smogon village, but I do think that this one had an unusually high amount of missed opportunities.
I do want to call out one role that I was particularly happy with how it functioned / was played:
- The Psychic / Pidge - I said this on the postgame chat, but I honestly think that nobody on the village was having fun until this role and The Summoning Circle discord server was activated. This role basically salvaged the village interest in the game and made it so that a LOT more people got invested / involved than otherwise would have. Pidge played well from a player perspective too, so just really well done all around here.
I also want to give props to the villagers who gutted it out post-Yeti / Dak and played through to the end, including billymills, LightWolf, zorbees, Mekkah, Patos, theangryscientist, Whydon, Realiti, Blakers, Redless, and ChaosNinjaGaming. Special props (bet you thought I forgot about him) to Bluedoom, who imo had a really rough first cycle or two but who I watched grow as an OC player to a massive degree over the course of this game as part of the village leader core. He went from claiming to LN on N0 to literally submitting all of the village night actions by N11, including the action set that won them the game. Spectacular turnaround.
Design Philosophy - The Cartel
I've heard the Cartel described as an 'amorphous blob' and I would generally agree with that sentiment. It was meant to be a faction that was not reliant on any particular players and that was susceptible to village misinformation / moling, but that was incredibly resilient and flexible while having sufficient numbers to be able to tank hits for a while and potentially even win a straight firefight against the village with some assistance from the Psycho.
I know that a lot of the mafia felt that the 'vanilla' nature of the goons made fake claiming harder, and that complaint is not unjustified. However, there were a lot of tools available to the Cartel to help with fake claiming that weren't utilized properly / at all, so I think this is a combination of design and Cartel faults on this one.
For the patch notes / changes I would have made:
- As mentioned above, add in the steadfast / miller text to the OP to help with fake claiming.
- Add in some additional situational tools for the late game with prohibitive strength costs, including a vote stop (cost ~60), an ensure (~20), and a bus drive (~10-15).
- Make it clear 100% that the FBI Agent is a neutral that needs the Cartel dead, and add in that The Anarchist is a neutral that needs the Big City dead. As a side note, I would've also given the Anarchist an anonymous line to both the Psycho and the Cartel to help it coordinate the anti-village forces (and probably remove the Psycho anon line to the Cartel as a means of forcing the Anarchist to be the coordinator between these anti-village forces).
- No, I would not give the Goons individual abilities.
Actual Play - Cartel
LN did a great job of blitzing out during N0 and getting a TON of free info. Sadly, the Cartel gathering of goon claims did not go as well, and they lost a number of claims to DLE that they really should not have. I think that timezones played a big role in this as SB. and Tommy were often not around at deadline, which is a pretty lame way to lose claims and have goons outed over. I do think that this was kind of the nature of the game though to some extent, and was good play by DLE primarily.
The Cartel night actions tended to be a bit more timid than I'd have personally gone for (not promoting Night 0, going for 26 tracks instead of a handful of full inspects, choosing to kill their inspection results rather than killspecting in the early going), but very few of them were objectively 'wrong'. The only two that were pretty clearly overly greedy were not Negative Safeguarding Utah the night it was open to a kill, and not beneficial safeguarding LN to protect from the Martyr when they knew Fairy was killing him.
The Cartel united around N3 or so, and started playing a lot better once they reached the state of being a cohesive unit. I think the individual goons were really struggling with a lack of direction or information prior to that, which is sort of the nature of the game but they had options to keep claiming later open (rather than just the open gooning that a few of them chose).
Their negotiations with the Psycho struggled initially due to being overly cagey with sharing village information and just generally being unwilling to deal with the Psycho as an equal, but it seemed like LN (on the anon account 'The Mafian') was able to establish a stronger rapport with Bass and they worked together very well until LN died.
Ultimately the Cartel nearly won, and had a few 50/50 guesses needed in order to win. Considering how badly they got clapped early, this was quite the accomplishment, and a testament to their resilience (both in role power and player fortitude). Respect the attempt and the fight they displayed, even if the large number of tactical and deception errors meant that I am not sure if I'd say they 'deserved' to be so close to the win.
Design Philosophy - Psychopath
Well this role was referred to as 'my baby' by more than one player in this game, and I will admit that I spent a lot of time on it. I think it a neat / clean combinations of abilities and information that gave it a good amount of versatility without having a stupid number of roles - indeed, on N0 it had a grand total of 0 actions! Obviously this changed (and in quite a hurry) down the line, but I do think that people forget how weak this role was in the early going.
Overall the hope was that this role would be rather strong and able to balance the two factions against one another, then straddle the line between the two and ride the snowballing nature of the role to a victory.
Alternatively, this role had a lategame win con of making a hidden recruit that then gets a sneaky upgrade related to faking it's own death and a 'defibrillator' revive, as a callback to the Mafia Mafia 2 defibrillator temporarily reviving the wolf.
Patch notes / changes I would make:
- Include goon strength as an option for what it can absorb, to encourage hitting the Cartel more.
- Give it a straight-up LPV but say that you cannot survive more than one vote (plus the Minion swap, if they opt for that).
Actual Play - Psychopath
Bass played really well this game, even though I think he is beating himself up over his handful of (relatively minor imo) mistakes. I think he did a great job of getting the millers on his side and squeezing some good info for free out of that, while sliding under the radar of the town for about as long as he reasonably could.
I think he was a bit of a pain for the teams to negotiate with, unfortunately - I think he was a little too focused on antagonizing the village with auxiliary abilities that he could have used as a negotiating tactic. He ultimately bought himself enough ill will from them that they were legitimately considering just throwing to the Cartel to make him lose, and he had to go hard enough in the other direction to stay alive that he ended up actually pushing them too far into the 'strength' column when the village got an upgrade that he didn't expect.
Ultimately wolves are really REALLY hard to win with in big mafia games, and he came really close to doing so. Now maybe he knows how Snype felt after LoTR mafia. Either way, surviving to the last ten alive is a significant accomplishment and he ALMOST pulled it off.
Awards
Best Big City: Yeti
Best Cartel: LonelyNess
Best Neutral: Bass (A Fairy played quite well as well, but imo had the easier path to victory)
Thanks to everyone for playing, and hopefully
Blazade can post the (six hour!) video postgame at some point in the next few days.