Postgame: Part 1-General Thoughts and Commentary
Okay, so this part is going to go into my personal mafia philosophy so if you want individual opinions/critiques, that will come in the second part. This will go more into how this game was designed and specific opinions on mechanics/play (disclaimer: these opinions may end up being asserted in a way where they come off as trying to be facts. My way of thinking of things is not the only correct way to play mafia, it’s just something that I think could help players in general, and me asserting things that way is not intentional.).
The Setup:
The main philosophy of setup design was trying to reduce investigative power and confirmable townies, the two things that most easily bust a game, to low levels, so that scumhunting was by far the largest focus. Even if everything goes right, only about 4/14 townies can possibly clear themselves, and those townies are not given any real power to go along with it. In terms of investigative power, you have Tracker, which if it targets exactly the right player, can out a nightkill, but can’t do much else, Roleblocker and Killblocker, which is the same as Tracker but can’t even confirm that player being the nightkill, and Cel’s name seer, which takes a lot of time and carries a low chance of very meaningful results (note that all of these roles have between a 5-20% chance of getting a useful result as the game progresses, instead of something like Watcher, which is close to 50% since it’s not too hard to predict the nightkill, and seer, at 100%). Instead, power roles tend to be protective, which pushes the game in a more fun direction, since it allows players to survive in the game for longer. Other roles are notably absent of town killpower and almost fully absent of town vote manipulation, instead focusing on communication.
Certamen awards were built to be useful for both alignments, but especially important for town to win to make up for their overall lack of investigative power (and killpower I suppose). The competitions also had the obvious activity purposes, since they forced you to check into the thread fairly often or risk missing potential power for your team.
My conclusion from this is that the idea of reducing the amount of confirmable townies was desperately needed and something we need to see much more. Additionally, reducing investigative power was a large success and something that I think hosts in the future here should 100% try to do. That being said, this balancing was not without its failures. The truth is that by removing investigative roles while not resorting to annoying vote weight manipulation or unfun killing roles, we ended up creating way too many protective roles and the amount of outside communication was insane when including the certamina. This issue could have likely been solved by finding roles that are obscure but still fun.
The Mechanic:
First things first, thank goodness there was only one mechanic in this game because two definitely would have been too insane to have. The whole idea of certamina is that they first of all would convince players to pop into the thread more often, so they wouldn’t miss one, and secondly could provide some pretty useful powers. Unfortunately, I don’t think people wanted these powers nearly as much as they should have, since every power conveys a sizable advantage to the player who receives in regardless of alignment (I’ll justify this part more in Part 2). These events I thought were fun and easy to participate in, not taking up more than 10 minutes of your day usually, yet people clearly didn’t want it very much at all (this definitely isn’t everybody, but ~50% of people weren’t participating in them at all and some of them like Venator Celebrus and Historianus Bonus had incredibly low turnout). I’m honestly not sure at all why this happened besides prizes not being OP and perhaps laziness. Yet it seems that people enjoyed having them around, so I suppose that combined with the potential views of the thread that we got from it are worth it?
I suppose I also ought to touch on anti-claim mechanics here. I was honestly really shocked how town didn’t really seem to understand that claiming was a good option if you were about to die (it appears this may have been caused by confusion that us hosts would be the ones stopping people from claiming; not scum?) or if you got a one-shot ability like Cel’s certamen award. Instead, the town lived in fear of this strange magical presence that the OP even said couldn’t kill you. Literally the worst thing that could happen short of death is being vanillized, and there were certainly times that people should easily have claimed when that was really the largest possible risk. I think Yog’s mistake was a large cause of confusion, but also there was a distinct lack of logic applied to anti-claim compared to stuff like PMJ’s ridiculous theory involving Robin and certamen blocking. That being said, at the end of the day the fault for this mechanic not working is ours as the hosts. I’d suggest any hosts that want to include anti-claim mechanics in the future should spell them out in detail for the players, so that this does not happen again.
The Hosting:
The game was actually surprisingly devoid of your typical hosting errors, which was really satisfying from our part. Roles processing was done with barely no corrections needing to be issued to the results, vote counts were correct the far majority of the time, and we were able to quickly resolve most issues that came up or questions that we were PM’d. And while we weren’t able to post them often enough near the end of the game, I think the vote counts were still posted often enough to be perfectly acceptable. That being said, there still were a couple of major mistakes:
-The largest one was me potentially TMI’ing on Jabber. It in reality wasn’t indicative of alignment, but that’s still a really bad mistake which invites in unwanted and ridiculous speculation that luckily didn’t matter too much (hosts, keep it in 3rd person hypotheticals next time k thx).
-There were things often missed in the start of day post, such as noting the number for a majority and the MYLO reminder, that should have always been included there. This is a simple mistake but a large one.
-We basically always weren’t there to warn everybody about players being on L-2 or L-1.
-We weren’t harsh enough on activity some times and too harsh other times, making our words seem quite hollow at times (which they often were because it’s hard to modkill people without murdering game balance).
In being careful to not repeat constant hosting mistakes, we ended up missing a few simple things that shouldn't be missed. Still, I’m pretty proud of our hosting in general because mistakes are inevitable but there were quite few this game.
The Play:
I was actually really happy for the Loyalists at first this game, as Town got one scum outed by end of Day 1, Cel caught a perspective slip from Samwise on Day 2, and town lynched two scum by Day 3. Yet then, suddenly town got into a terrible mode: reads not based on motivation. PMJ and Cel attacked eachother for what was really playstyle (in their individual critiques, I’ll note why I found what they did to not be from a scum mindset) and things started to slowly go downhill when instead of building a core/pressuring more players to better evaluate them/creating a PoE/looking back at the past from interactions with scum and evaluating with the hard evidence that is votes, town started to just get a consensus scumread for the day and then stop evaluating for then (the main exception to this is Yog, who did in fact try to clear/be suspicious of people based on wagon formation, but actually got attacked of all things for doing so). Meanwhile, Drac just sat happily in his Venator Celebrus chat turning town into two sides that only scum really managed to stay out of (mostly since every player’s thoughts this game consistently led back to ever-centralizing PMJ).
I think the largest generic takeaway to the town from this is to not get complacent and to do any of those things I listed above because this whole consensus scum thing really got nobody anywhere, but the second largest is simply to use actions along with words. When I can get better reads from analyzing the D3 t/s wagons (PMJ, Jabber, Yog, jplap town; Samwise, Cel scum (though in reality with a bit more thinking, they’re never s/s so really if you hit right the first time you don’t need to lynch the other); scattered, Drac likely scum for being offwagon), than town could give me due to all the hysteria, there’s a problem. Or when PMJ townread Samwise for seriously voting what he knew to be town over scum, there’s a big problem.
I guess the main thing I’m trying to say here is that when we have so few posts in a game, it’s not too large of a challenge to do some research on the past instead of purely focusing on the present. If you keep up the desire to scumhunt that you have early game, you can get really good results lategame by doing any of those things that I suggested. Just please please please don’t get into back and forth fights that are basically OMGUS, because wall tunneling messes up much more often than it succeeds.
Tl;dr:
-Reducing investigative power is very good thing. Replacing investigative roles must be done carefully, however, to make sure other types of roles don’t get overemphasized.
-The certamina were rough around the edges, but generally a good idea. Prizes, however, could have been better.
-Anti-claim mechanics needed to be more transparent so people didn’t create an elaborate fantasy of what they were.
-Hosting went well overall, but we made some simple mistakes that should be avoided in the future.
-Play started good, but players didn’t look back at all or take any of the simple steps that could have helped town keep up their momentum.
