Mafiascum.net

In post 2602, Menalque wrote:I could be slightly wrong here bc I didn’t follow that closely after I got banned but I don’t think it was his *actions* that falsely mechcleared him as it was bad setup spec, maybe S_S can help out on this tho, or lili herself?

You are correct.

In post 2612, Menalque wrote:My priority there would always be denying town the cop

The problem with doing this is then you pretty much have to claim cop. And skitter can tell you how well that goes :3

Mena, do you know if skitter would be more open minded to lilith!town as town here? I can relate to the idea of knowing how to read someone and trusting your read in them 100%, but if there are other people in the game that also know how to read lilith, it doesn't make a of sense to hold on to the read this long right? Idk, we're getting into the problem of people knowing each other's meta better than me, and generally being better scumhunters, but I wanna know if I wrote skitter off as town too early. Especially if this is true:

In post 2616, Menalque wrote:normally you call out when people are TRing you for bad reasons but I don’t think I’ve seen you do that here

I think my reasons for TRing skitter were somewhat lazy tbh.

jailkeeper's a strong role and has obvious benefits for scum in acting as a roleblock while denying town a protective, it took me looking at 2 old games to find an instance of scum taking it high in the draft (pick number 2). i don't think the role selection is necessarily clearing and don't townread her on content but i'm getting the vote i want more right now.

kind of sort of feel like skitter would not pursue a target so singlemindedly as scum but that's more on the idea that most people would pivot off if it became clear it wasn't viable, don't have an idea if that thing would be in her wheelhouse or not

it is ~interesting~ that there is basically one wagon with no opposition but there is a big chunk of players right now that are passive to say the least. i like the people i'm voting with, though, even if they do not trust me

In post 2567, Menalque wrote:VOTE: dunn

this is de facto G-1 with NM in the game, let’s get an intent on dunn

Wtf is G

Peta why does scum!lilith claim when she does? Also what happens if dunn is town and how likely do you think that is?

In post 2626, Something_Smart wrote:

In post 2612, Menalque wrote:My priority there would always be denying town the cop

The problem with doing this is then you pretty much have to claim cop. And skitter can tell you how well that goes :3

yeah 2/10, would not recommend

In post 2536, Infinity 324 wrote:she's been really transparent and the interaction with mena during d2 felt really T/T to me

@mena @infinity this bit is why i don't feel like it's a super generic/bad townread on me
there's a bit more nuance here than just like generic ~super obvtown, has nuance~ etc

and i can understand objectively how someone might come to that conclusion about the me/mena interactions ^

In post 2628, petapan wrote:kind of sort of feel like skitter would not pursue a target so singlemindedly as scum but that's more on the idea that most people would pivot off if it became clear it wasn't viable, don't have an idea if that thing would be in her wheelhouse or not

it is, but i don't think i would be doing it here as scum

In post 2630, Not_Mafia wrote:

In post 2567, Menalque wrote:VOTE: dunn

this is de facto G-1 with NM in the game, let’s get an intent on dunn

Wtf is G

Guillotine minus one

I'm torn between following two basic principles.

One basic principle is that you don't pressure suspected TPRs because they have greater utility on average than non-PRs. Especially in a game like this, you give them room to operate.

The other basic principle works like "duck typing" in programming. If it walks like a duck and quacks like a duck, then it's a duck. Someone who isn't scumhunting, especially those making an effort to be present but not scumhunt, are typically scum. My gold standard for this is people either not expressing scumreads, or saying things about alignment opinions but never giving a real reason.

By the TPR principle I'm explicitly against pressuring the top 5-6 draft picks early in the game, . They can be the last elims because the ones left standing are the ones most likely to be scum. If we have a doc we may get lucky and have more no-kill nights, extending the game to the point where a pure mechanical solve is possible. If we have 1-shots that are more useful the smaller the player list gets, we give them room and don't out them so that scum don't know who the real danger is.

By the not scumhunting principle, I'm being pinged by Dunnstral, Something_Smart, and Best Bird. 2/3 of those are especially annoying because people explicitly townread them for things that I would consider to be NAI at best and explicitly scumtells in most players. Not_Mafia will take his slot from a "weird but could be town" read straight to this category. The conflict between this principle and the TPR principle is obviously largest in Dunnstral.

I tend to SR people who don't acknowledge the TPR principle. Hoopla is a prime candidate for this award. I really don't give a shit what the numbers say about scum doubling or not doubling. The TPR principle overrides that. If you don't have an explicit reason to scumread a presumed TPR, you don't yeet them, period. One thing gives me pause here, and that's the "protect via suspecting" strategy. Scum will choose not to NK people that they think are easy miselims. I have to acknowledge that some players purposely "suspect" the players they have identified as likely TPRs so that scum won't kill them as universal town reads.

What this all boils down to is thinking for today (and possibly tomorrow) that we should eliminate either the soft guilty from N1 to rule it out, or an explicitly scummy or not scumhunting player from the later part of the draft. The only scenario that can actually burn us with this strategy is all 3 scum in the early part because they all picked low unique numbers and town got extremely unlucky to not double any of the scum low numbers. I'm not unwilling to eliminate Dunnstral because I can see him playing this way as scum, but I'm concerned about it being too early from a pure mechanical point of view.

Forcing scum to claim PR early is good.

In post 2637, davesaz wrote:By the TPR principle I'm explicitly against pressuring the top 5-6 draft picks early in the game, regardless of behavior.

*checks top 5-6 draft picks*

c o n v e n i e n t

What role are you concerned dunn has that would be such a great loss here, Dave?

That’s also an awful lot of words to say “let’s do mena, Best Bird, or NM today”

Like he could be cop, doctor, UB (so 1s PGO), or 1s watcher

If he’s a cop then outing him gives us 2 innos, if he’s a doc then yeah be not ideal but basically any of the other roles are not so amazing that they’re likely to radically shift the balance of the game and none of them justify him playing like this

In post 2637, davesaz wrote:Not_Mafia will take his slot from a "weird but could be town" read straight to this category.

If you know this is going to happen, doesn't it make it NAI?

Talking about it at all helps scum. Nice job.

we all like collectively have ignored the possibility of 1s commuter being the reason the kill went missing n1, i'm noting

i'm not super vibing dave!scum tbh but idk where to vote rn

VOTE: hoopla let's try this

In post 2637, davesaz wrote:I tend to SR people who don't acknowledge the TPR principle. Hoopla is a prime candidate for this award.

You literally just saw her do this as town though, so whether you agree or disagree with it, it shouldn't have a bearing on her alignment.

In post 2638, Infinity 324 wrote:Forcing scum to claim PR early is good.

But forcing town to claim PR early is bad.

In post 2645, skitter30 wrote:we all like collectively have ignored the possibility of 1s commuter being the reason the kill went missing n1, i'm noting

I did so because commuter has literally never been successfully taken in any run of this setup before.