Mini 2266: GnG's Upick Mafia [GAME OVER]

[mastina]

For the record: I was interrupted before I could make the post which would detail this, but I think that the setup was a lot less townsided than people are giving it credit for being.

The masons were not masons, which I was more or less figured out by my questions to Ginngie.

Pooky's base role could have killed a town player outright.

There were multiple powers that the town had access to which could hurt the town, including all the uses of the miller.

And the biggie: every power the town had access to, the scum had access to as well. And literally all of them were useful.

Bulletproof to shut down town vigs.
Rolestop to roleblock town actions, including vigs.
Ninja to avoid the tracker.
Doctor to nullify the vigs.

Their 1x strongman could be turned into a 3x strongman by using ability thievery on it.

They could gain access to

three

vig shots, allowing for

four

scum-controlled kills.

And they could revive scum, who if they had a bulletproof, could only be re-eliminated with the town using another elimination on them.

These are all very very very powerful tools.

That, not even going into scum upgrading themselves, and the scum being able to have more XP overall.

Everyone's treating the game as if it were like 95:5 town:scum, but as I was going to say:
For a bunch of people who have modded games theoretically balanced that ended up hard-swinging towards one direction and ended up looking X-Sided in favor of that direction...
...Y'all are kinda bad at recognizing that

that's what happened

here. The scum had a LOT of tools at their disposal, especially if they had managed to live longer than they did.

They just didn't.

And the town powers were used with a higher rate of success than they should have been, for instance both masonries being town in spite of them being neighborhoods.

Does that make the game balanced as a 10:3? Well, no. I'm not saying it wasn't

townsided

, but I AM saying it wasn't townsided

as much as you're assuming

. As a 10:3, I'd rate it at ~60-65:35-40 town:scum. Townsided, but not as egregiously so...

...Especially given that there was a chance for it to

not

be a 10:3.
I could have become a role that was not only not town, but was also explicitly antitown. It would make me act against the interests of the town, actively harming them at every stage.

Roden could have more heavily focused on the 3p win and actually obtained it, leaving the game. It would make Roden be not a member of the town and cost the town a valuable mislim. Not as actively harmful as if I had become scum, but still bad.

Given all of what the scum had,
AND,
That the game could potentially become 8:3:1 +1...

...I'd rate it a little less townsided than the above. Given the chance for optional 3ps to become 3p, it'd be closer to ~52.5-57.5:42.5-47.5 (discounting the 3ps there numberwise), which is still a

little

townsided, but not nearly as egregiously so.

The scum could have four kills go through; the scum could have three strongman kills bypassing all the protectives; the scum could have scum immune to town vigs requiring they be re-limmed; the scum could roleblock almost any action this game.

Those, in tandem with the potentially lower number of town players and the antitown players explicitly being antitown, would have given the scum much much better odds.

...It's just that everything that

could

go wrong for the scum, did, and everything which could go right for the town, did.

So I wouldn't be too harsh on Ginngie balance-wise. It wasn't perfect, it was slightly townsided especially with the town-optional-3ps not pursuing the option of 3p, but it wasn't a guaranteed town victory.

[Dwlee99]

I think that the setup was a lot less townsided than people are giving it credit for being.

Stop

[PookyTheMagicalBear]

Also the reason for having so many investigations is because masons were a lie and were just a hood, and miller and ninja existing so there would be a lot of potential for fake guilties fake innocents. Oh and I thought a succubus was a vampire when i googled it so shiro was also on the block for Rodens coin.

i dont get this

u mean you lied about the masonry or the players lied about the masonry? lol

[jjh927]

I mean, it theoretically could have swung that way, but lolbalance is about averages and if you give abilities to everyone then town get to use them more than scum

It really just wasn't swingy enough for it to wind up swinging towards scum on a probabilistic level

Your example for 4 exp to get the rolesteal only to use it to duplicate a strongman is A LOT when a bulletproof is 1 exp, and it's totally unexplained exp that goes missing, and only duplicates it once because you can't steal a stolen ability. Same goes for the vig. If town gave scum enough exp to get vigs on n1 and they all just happened to go into the warrior tree for the vigs, that'd involve town giving a huge amount of resources to scum.

At the end of day 1, which would have been the most productive day for production of exp, there was a total of 16 full integer exp points floating around. 6 of those were in the sneak hood which scum didn't have access to unless reinhardt moved over there.

A vig was worth 4 points. And they didn't even know it was a vig.


Overall, it is predictable that things went how they went, with individual players being willing to pass a small amount of exp to achieve something early on, but generally with people they trusted based on day 1 and in a manner that scum would have been held fully accountable for their actions if they misused the exp.

[mastina]

I think that the setup was a lot less townsided than people are giving it credit for being.

Stop

No.

I will attack a setup for being X-sided regardless of the outcome of the game, when that is true. If I win in a townsided game I will call it out; if I win in a scumsided game I will call it out; if I lose in a scumsided game I will call it out; if I lose in a townsided game I will call it out. Regardless of who won or lost the game.

When I say it was not as townsided as people are saying it is.

I mean it is not as townsided as people are saying it is because in theory, it wasn't.

Pooky killing a town player could happen.
The two town players with 3ps could have 3p'd instead of townsided.

Those alone reduce the town from 10 to 7.
A town death D1 could reduce them down to 6 (Dunnstral D1).
A town nightkill from the scum strongman could reduce them down to 5.

There was the potential for there to be

only five town alive

on

day two

--and that's

without the scum accessing a Vig

(which I think they mathematically were able to do?).

5:3:1 on

day two

would be five town to four nontown. Literally

lylo on D2

, IF the scum didn't have a successful vig. Add in a successful vig and in the worst case scenario?

The town would be in KINGMAKER on D2

. As in, it was possible for the town to effectively be in an unwinnable situation on D2, a 4:3:1 where the town doesn't have majority and needs to rely on a SK to get a scum elimination at all.

Now, that didn't happen. It was pretty much the opposite of that which happened.

But that's an example of why you don't judge how a game did play out for the theoretical balance of the game.

Was this game perfectly balanced? No, I do think it was slightly townsided.

But I will maintain that it was not absurdly townsided because in basically any world which wasn't this one, the scum have a lot more ability to play the game effectively.

[Dwlee99]

Mastina there were literally revival abilities. Scum pretty much never win this setup if it's repeated 100 times.

[jjh927]

Basically, because there wasn't a lot of exp floating around and so big ticket stuff required either highly telegraphed actions over multiple days or significant levels of cooperation, players were fully accountable for exp related actions. The crazy scum plays you're picturing just wouldn't ultimately lead to a scum stomp because of how clear the misuse of exp would be

Also, once the reviver became public knowledge it was clearly the +town option, barring a reasonable level of paranoia

[jjh927]

Also, town can really snowball in this whereas scum cannot. If scum lose a player, they're down a huge amount of options in terms of exp related actions and exp generation. Once it's down to the last scum they are literally never making another nightkill again.

[mastina]

Mastina there were literally revival abilities. Scum pretty much never win this setup if it's repeated 100 times.

Revivals which the scum could also get and which the scum could stop.

It costs 4 XP to revive a player.
It costs only 2 XP to get a rolestop to block the revive.

The scum also had access to more XP than the town, via Enchant. So they could secretly use their XP in places that the town would be (mostly) unable to track them on. After all, aside from Roden, the town would have no way to prove someone did or did not spend XP somewhere. So scum could just outright lie basically freely. Especially in a game labeled as bastard.

Now I realize this is me talking and I come from a school of honest scumplay where telling the truth is usually a great weapon. But in this game specifically, I'd actually argue that while that was a

possible

strategy, the

optimal

strategy was just lying about your XP spending. It'd be known how much you made outside of Enchant, but it wouldn't be known how you spent it outside of Roden.

[Ginngie]

So I wouldn't be too harsh on Ginngie balance-wise. It wasn't perfect, it was slightly townsided especially with the town-optional-3ps not pursuing the option of 3p, but it wasn't a guaranteed town victory.

I thank you for the post but even I admit this was the worst balance I've created and said it in the set up thread as well

[Dunnstral]

It costs 4 XP to revive a player.
It costs only 2 XP to get a rolestop to block the revive.

Town has significantly more xp due to having more members

[jjh927]

Scum had SLIGHTLY more access to exp than individual town players, but still massively less overall. If they have to burn everything on counterplay that might not work to prevent revives, how will they get anything done at all

This also doesn't address how if scum get limed day 1 that is a direct path to basically town autowin

[jjh927]

Remove a single scum player, especially earlygame, and scum lose a huge amount in the exp side of things proportionally against town. And once there is only one scum player left, they will lose at almost any point in the game because they will never succeed in another nightkill.

[Ginngie]

oh I forgot to mention in my game end post that another critical error is that in my previous game where I had a market, it was balanced because I calculated how much possible money could be generated, then changed the cost of items so the amount of XP possible equated to how much XP the total cost of the items were. This game I didn't run those numbers which was a big mistake. I actually added challenges because I didn't think the XP earned was enough. Simply checking the math I more than likely would have gotten rid of challenges all together or eliminated the wagon XP. Oh well, lessons for future games

[mastina]

Also, town can really snowball in this whereas scum cannot.

Um.

I literally demonstrated how scum could have effectively won the game by D2?

I'd call "game is literally 4:3:1 on D2" to be a scum snowball.

And I don't really see the argument for how a player MUST be accountable for their XP spending this game when there was a grand total of one mechanic (one player's role) that could track that. Yes, you know how much XP a player has

earned

. How do you know how they have spent it? You have only your word that they did so. What stops the scum from lying about how they've (not) used their XP? What stops the scum from buying an ability they shouldn't have due to Enchant's bonus XP accumulating? If they buy an ability they shouldn't have the XP for, how does the town know they have exceeded their XP? Roden's legit their only method to discover that from the scum.

Why can't the scum do those things? Lie and say they have more XP than they do, or lie and say they have less XP than they do? What can the town do to

force

a player to demonstrate the amount of XP they have? None of the actions which could be bought would produce a provable result that there was no failure condition for. Forced to demonstrate a vig? Target BP or bg'd or doctored or they were rolestopped. Forced to revive? Rolestopped. Forced to rolestop? Rolestopped, so the action they're blocking goes through anyway. Forced to steal a role? Fakeable but even if not, could be claimed to be rolestopped.

What ability could the town pressure the scum into forcing they can/can't buy, with it being proven that they can/can't buy it?

I don't see how the scum couldn't just lie.

[Dunnstral]

Scum snowballing relies on town playing poorly, not on scum playing well. The town has to misplay to allow that to happen

They also don't know that there is a vig at the start of the game

[jjh927]

If your idea of scum snowballing involves town literally giving scum all of their exp then sure

Any setup is balanced because the entire scumteam could be locktowned by the entire playerlist on day 1

[mastina]

Remove a single scum player, especially earlygame, and scum lose a huge amount in the exp side of things proportionally against town. And once there is only one scum player left, they will lose at almost any point in the game because they will never succeed in another nightkill.

I mean, yes--but this is true of any game. An early scum death reduces scum power and increases town power. That's inherent to the design of a mafia game.

That's swing.

While you can say it's more likely to swing towards town (because it is, the game

was

townsided), it's not guaranteed to swing towards the town even remotely.

How different does this game look if there's two town mislims in a row especially with the two potential 3ps going for 3p?

Two town mislims in a row is something I would say is entirely not unreasonable to have happen in a game. It didn't happen since we eliminated Gypyx on D2, but we still had players defending Gypyx on D2 so it was fully possible that Gypyx could have lived through that day. How different does the game look if Gypyx is still alive going into D3? That, in of itself, means that there's still two scum alive minimum going into N3 (possibly 3, depending on if the guilty manifested or not since the guilty on Dwlee was at least in part followthrough from the Gypyx elimination), which gives them a lot more agency.

[jjh927]

Also I did think of a way of verifying someone's exp that I kept in reserve in case I wanted to become the fun police.

It costs 0.25 to retry a challenge. If someone claims to have X.Y exp, then give them (1-0.Y) exp, get them to give all their exp to someone in the same hood as them, fail the challenge, and attempt to retry it.

[Ginngie]

you never cease to amaze how you break my fucking game

[jjh927]

Remove a single scum player, especially earlygame, and scum lose a huge amount in the exp side of things proportionally against town. And once there is only one scum player left, they will lose at almost any point in the game because they will never succeed in another nightkill.

I mean, yes--but this is true of any game. An early scum death reduces scum power and increases town power. That's inherent to the design of a mafia game.

That's swing.

While you can say it's more likely to swing towards town (because it is, the game

was

townsided), it's not guaranteed to swing towards the town even remotely.

How different does this game look if there's two town mislims in a row especially with the two potential 3ps going for 3p?

Two town mislims in a row is something I would say is entirely not unreasonable to have happen in a game. It didn't happen since we eliminated Gypyx on D2, but we still had players defending Gypyx on D2 so it was fully possible that Gypyx could have lived through that day. How different does the game look if Gypyx is still alive going into D3? That, in of itself, means that there's still two scum alive minimum going into N3 (possibly 3, depending on if the guilty manifested or not since the guilty on Dwlee was at least in part followthrough from the Gypyx elimination), which gives them a lot more agency.

We still discover that the mage 4th tier is a reviver in night 2.

[Ginngie]

BTW this was game was townsided and I designed it to be townsided given my games have always been scumsided but I overcorrected.

Mafia theory gives scum a 60:40 chance of winning mountainous so I went with more town power to try and even it over, which I overcorrected.

[jjh927]

Oh, and the night 2 nightkill still fails

[jjh927]

so like, if we mislimmed in d2 as well then town still comes out of it with a conftown revived LLD and only 2 town dead. 11p, 8-3, reviver now public knowledge

[Lady Lambdadelta]

so like, if we mislimmed in d2 as well then town still comes out of it with a conftown revived LLD and only 2 town dead. 11p, 8-3, reviver now public knowledge

A reminder that even if the BG night was resolved correctly, I chose that explicitly because I would just revive myself again the next day.

So nothing changes from that point on. I was never shot at and took no more actions.