Micro 643: Desperation Day (Game Over)

[xyzzy]

that was a very poignant speech, Comparing Realities. very moving.

of the two meaningless wagons so far, this is the one that feels like it has slightly more behind it (plus, that dabbing panda is weird):

VOTE: keyenpeydee

[xyzzy]

and go into day 3 with basically no info besides who died nights 1 and 2 and which players agreed to no lynch? that seems like a wildly bad idea

like how are you proposing that we'd scumhunt without actually lynching anyone? that seems difficult at best. that'd basically be giving scum free reign to set up day 3 however they like

I'm gonna check the math real quick but I'm pretty sure even just based on random choices that's a subpar choice for the town to make. also I'm guessing that callforjudgment already did that math and came to the conclusion that a ruleset allowing for a no lynch to happen explicitly does not make the game mathematically more breakable that way??? probably.

[xyzzy]

I'm not sure whether I did this math completely correctly (the first set of numbers add up to 1.002, which might be a rounding error, or it might be that I missed something???? not sure) but the numbers mostly match the numbers the original thread for this setup had, which maybe I just should've checked there first,

but whatever

anyway here's stats for town odds if we lynch d1 and d2 vs no lynch d1 and d2. I didn't check the numbers for if we only lynch one of those days because I'm pretty sure the general rule that town odds are better with an even number of players still applies with how d3 works:

7/9*5/7*3/5 = .333 (no scum ever lynched, scum win)
7/9*2/7*4/5*2/3 = .119 (scum lynched d1 or d2, scum win) (2 ways this can happen)
7/9*5/7*2/5 = .222 (no scum lynched d1 or d2, scum lynched d3, town win)
7/9*2/7*3/5*2/3 = .089 (scum lynched d1 or d2, town lynched d3, scum lynched d4, town win) (2 ways this can happen)
2/9*1/7 = .032 (scum lynched d1 and d2, town win)

43.1% chance of town win if we lynch day 1 and day 2

5/7*3/5 = .429 (town lynched d3 and d4, scum win)
5/7*2/5*2/3 = .190 (town lynched d3, scum lynched d4, town lynched d5, scum win)
5/7*2/5*1/3 = .095 (town lynched d3, scum lynched d4 and d5, town win)
2/7 = .286 (scum lynched d3, town win)

38.1% chance of town win if we no lynch day 1 and day 2

so it's objectively worse to no lynch

also I don't trust your home site as a good source of info because that's entirely dependent upon the site meta. does your home site focus on mostly setups with lots of power roles, for instance? if so, those numbers are completely irrelevant to this game.

[xyzzy]

that being said I think it would be wise to lynch our second-highest scum suspect d2. that's kind of a gimmick, but it's one that still focuses on having a worthwhile lynch d2 while giving us an ideal d3

[xyzzy]

also the only bad part about lynching our second highest scum d2 I can think of is that it lets scum plan ahead for who is likely to be lynched d3 more than what is totally optimal, but I think the payoff is worth it

[xyzzy]

also I don't trust your home site as a good source of info because that's entirely dependent upon the site meta. does your home site focus on mostly setups with lots of power roles, for instance? if so, those numbers are completely irrelevant to this game.

Likewise, I dont trust your math at all because it hasn't been tested by practical application.
Theory math sucks.

~Mas Y

the mathematical probability of certain events happening is literally half of the entire basis on which the balance of mafia games is based (the other half is that town is, on average, not that good). you're arguing that town should do something that is objectively not in their best interest when you have direct evidence for why that's the case right in front of you

VOTE: Mas y Menos

[xyzzy]

there's nothing theoretical about math. it's concrete fact.

[xyzzy]

are you saying math isn't a source of factual information? like........ that's just.......................... objectively wrong

this has nothing to do with whether human action is predictable. it has to do with the fact that our choices in this game can be modeled by concrete mathematical statements and that certain actions can be ruled out as suboptimal. of course there are exceptions where human interaction overrides that in actual games, but this early into the game isn't the correct time to be making those decisions. like if we get to September 21st and we decide that, based on what's happened during day 1, no lynch makes the most sense, then sure, that's possibly a route worth taking. by asserting that we should do that from the very beginning, you are explicitly saying that you want the town to not work in its own best interest.

[xyzzy]

Mas y Menos I am having a

bit

of a dilemma here, and I need your help. I need to figure out which of these numbers is bigger, but I just can't figure it out! can you help?

a)43
b)38

any assistance you can offer would be greatly appreciated! thanks~

[xyzzy]

thnx 4 the help <3

[xyzzy]

that's fair

[xyzzy]

No, in fact, we likely have an even better chance if we lynch, as we can get vote reactions. This guy's not getting it at all.

A vote reaction to a mislynch train usually leads to very little info.
Can you pinpoint games in which mislynches helped you solve the game by how people went on the wagons?

Cause I can show you were mislynch trains compound on each other because town misreads town and other town thinks that other one was being scum pushing that, far more often then the opposite.

~Mas Y

Missed this Gamma

oh yeah I meant to respond to this too

this feels like a question deliberately set up to force the answer you want, because the truth is, in 99% of games, it would be impossible to pinpoint one early game action that was truly decisive in determining which faction eventually wins. maybe mislynches are, on the whole, harmful to the town (I disagree with that, but whatever). however, whether mislynches are a beneficial source of info is super subjective. also, as I mentioned earlier, apparently your opinion is formed partially by experiences on another site, which are frankly totally irrelevant to this discussion. that'd be like me trying to datamine the in-person mafia games I played back in high school at church camp with people who didn't realize that there was actual strategy involved in the game. in other words: totally irrelevant

[xyzzy]

1: What is your preferred faction?
2: What would you say you bring to the game that others don't?
3: Tell me a bit about your username, signature, and/or avatar.

1. I haven't really played enough games since coming back to the site in April (after being away 8 years) to really answer this question reasonably. I've been playing in one game at a time, and in all the games I've been in that are complete, I was town.
2. I have a tendency to go back and re-read the thread and ask questions about things that only seem out of the ordinary in hindsight. I feel like that's something I've done more than others in the games that I've played, whereas a lot of people mostly focus on re-reading as a way of documenting who is scum and who is not.
3. my username is a word that shows up in many text adventures. my avatar is art of the character Kay Faraday from the

Ace Attorney

series drawn by a friend of mine. my signature is a link to my GTKAS thread, which you should go to and make posts in.

[xyzzy]

it's definitely possible that Mas y Menos isn't scum (I mean, statistically, it's likely), but they're proposing harmful actions by the town, and they've yet to back down from that, and that's worth putting pressure on.

[xyzzy]

oh hey here's these

4. I played a bunch of games 8 years ago, but I left the site for a long time and only came back in April of this year; since then I've played in one game at a time.
5. Firebringer's alt replaced into a game I was in. I have also played with BlackStar. I'm pretty sure there aren't any others, although I might be forgetting.

[xyzzy]

Comparing Realities, you misunderstood my reason for RVS voting keyenpeydee; my reasoning was that the other wagon happening at the time was just a wagon for the sake of wagons, whereas keyenpeydee had done one very slightly scummy thing. he had a minutely higher chance of being scum than rb did at the time. if I didn't feel like pushing a wagon of some sort, I would've voted for someone without any votes on them.

other than that, I really like that post.

[xyzzy]

I think it's just meant to provide a common baseline of information about everyone? it doesn't seem too useful to me, but I can see the appeal it might have to some.

[xyzzy]

Xyzz, could you please answer CR's question about your 37?

UhhhhhmmmmmActually, I kind of like that...Would we all individually vote our second-highest FOS, or vote who is generally the second-most FOSed? Because the first scenario lets scum be more creatively manipulative, which is, well, bad, and the second is tough to live by and easily screwed up by a lone rebel.

tbh I'm not sure what the best way to handle figuring that out is, and it's complicated by the fact that, ideally, it should happen without the scum being able to fully determine who everyone thinks the most scummy person is (which could be used to make night kill decisions). my instinct is to say the former without anyone directly revealing who their primary suspect was for the day until day 3. basically, if we're all sufficiently cagey about the precise order of our reads lists throughout the first two days, I don't think it'll be as possible for scum to manipulate the results.

[xyzzy]

I agree, except I'd suggest just listing people in those categories in alphabetical order (that way, there's no possibility that the randomization isn't truly random), and maybe get rid of the null category, since otherwise the number of people in each group will be really tiny.

[xyzzy]

some people communicate best through walls of text; I know that at times that's the best way for me to meaningfully express my thoughts about the game. refusing to engage with certain players because their posts take a bit more effort to digest isn't good for the town.

[xyzzy]

I wanted to hear what rb actually had to say about this before I did this, and now I've heard what rb has to say about it, and... like... yikes

I'm gonna keep my vote on Mas y Menos because I feel like they're trying to fly under the radar so that we'll forget anything scummy they've done, but rb is definitely a solid lynch candidate at the moment

BTD6_maker has hardly said anything at all, and hasn't really said much of anything of substance. not a huge fan of that.

[xyzzy]

No.

I found where you said that. it took a lot of work to find. I had to scroll up a whole 7 posts

also, this is a game that rewards bussing a bit more than usual, so the fact that we're voting for the same person isn't really that strange at all.

[xyzzy]

callforjudgement, just so you know, you have 2 votes listed for rb and none for me—it looks like you removed me from the list of people voting for Mas y Menos instead of rb.

Fixed — callforjudgement

[xyzzy]

BTD6_maker has hardly said anything at all, and hasn't really said much of anything of substance. not a huge fan of that.

BTD6_maker became a whole lot better since I made this post

[xyzzy]

Care to explain why rb is a strong lynch candidate? Make it detailed.

sorry for missing this question earlier

I don't feel as strongly about rb now, because I think the explanation given for the whole skimming wall posts thing was reasonable (and I'll be the first to admit that I completely missed post 108). rb hasn't done anything that really screams "town" to me but also doesn't seem so much like scum now.

this isn't related to the actual game, but rb, you don't have anything listed under the "pronoun" section of your account profile--what do you prefer?

[xyzzy]

@BlackStar

It's because she says rb is a strong lynch candidate. I asked her why? She says, "I don't feel as strongly about rb now.. etc etc". Which It's questionable that her strong read on him just disappeared.

@xyzzy, Why did your strong read disappear?

because my read was based on the whole issue of actually reading wall posts vs not reading them at all vs skimming them; although I felt that rb's initial explanation for this was really scummy, they turned it around and justified those previous statements really well.

[xyzzy]

it's good to clarify that it's intended as an edit rather than a shift in your thought process. (i.e. your opinion hasn't suddenly changed)

[xyzzy]

I need to reread keyenpeedee's ISO, because for the most part he's someone I've not been paying a lot of attention to.

also, as I said earlier, I feel like this is a setup that strongly improves the benefits of bussing, except on day 3, where it's a little worse than usual. however, days 1 and 2 we should focus on the possibility of bussing strategies for sure.

[xyzzy]

I need to reread keyenpeedee's ISO, because for the most part he's someone I've not been paying a lot of attention to.

also, as I said earlier, I feel like this is a setup that strongly improves the benefits of bussing, except on day 3, where it's a little worse than usual. however, days 1 and 2 we should focus on the possibility of bussing strategies for sure.

Doing a reread on me to sheep BlackStar and so that you can hammer me and my suspicions on you will be left out.

I can understand why you'd think that, but if that's what I were doing, I'd be applying way more pressure than saying that I need to do an ISO reread.

[xyzzy]

I've read over keyenpeydee's ISO a couple times and this feels more like a townie who is at times playing poorly then scum to me

[xyzzy]

I mean ending day 1 a week early without general consensus that that's an acceptable move is a bad thing to do regardless of one player promising a policy tunnel

[xyzzy]

just unvoting and not voting anyone else isn't applying your vote elsewhere.

[xyzzy]

scum gambit? really shitty town? ugh

[xyzzy]

he was at L-1 as of the last vote count; Gamma Emerald moved his vote elsewhere and then back to keyenpeydee, and then keyenpeydee self-voted. unless I completely missed someone else unvoting, that was a self-hammer.

[xyzzy]

hey pro-tip to everyone else since it apparently needs to be stated: don't self hammer as town wtf

anyway here's my suggestion based on what I said yesterday: we each put together a reads list with 3 people each in two categories (scum/not scum) with our lists alphabetized so that it's not possible for scum to know precisely which members of the town are perceived best and worst. then we figure out roughly who is the second scummiest person and lynch them, saving the scummiest person for tomorrow. to avoid allowing scum to manipulate the results too much, we'll choose the order in which we do this by popcorning it. are other people down for this plan?

(obviously we'll spend plenty of time actually discussing things—I'm not proposing that we just immediately lynch someone the moment we have the two candidates chosen—but I think this method will produce a strong guide for what direction to move in that is suited particularly well to this setup.)

[xyzzy]

I'd say it's probably best to not explicitly state how your reads have changed until there's been a couple of shifts—otherwise it's fairly easy to tell which people in your reads list are actually near the middle of the list.

in general it's safe to assume that the person who is perceived as the most scummy has a higher chance of actually being scum; assuming that we lynch both of those players over the course of the next two days, it's better to lynch the scummiest person tomorrow, since that increases the odds that we simply win tomorrow.

[xyzzy]

that's a fair critique of it—it is a pretty fragile plan where things can go wrong. however, I feel like the likelihood of town winning at 2:3 lylo are much, much better than the odds of town winning at 1:2; technically, when you factor in the odds of lynching scum on days 2 and 4, our odds are marginally better when we lynch scum day 2, but the complexities involved in scum coordinating a lynch at 2:3 are pretty substantial. it's way easier for scum to just hammer at 1:2 and win.

[xyzzy]

I meant to quote 407 there in case it's not clear what that's a reply to

[xyzzy]

hey, people who aren't Alchemist21: what do you think of the plan I proposed in 398?

[xyzzy]

wait, if you think it's possible to rig, why do you have no qualms about doing it? that doesn't make any sense and is... unsettling, to say the least.

BlackStar, the idea is to maximize the likelihood of hitting scum on day 3 while minimizing the odds of hitting town on both days, and to plan this in such a way that it's not feasible for scum to break it (either by manipulating the results in their favor or by choosing night kills with full info on how every member of the town feels about every other player).

there are four ways days 2 and 3 can play out: 1)we lynch scum both days, town wins, 2)we lynch town both days, scum wins, 3)we lynch scum day 2 and town day 3 and go to day 4 with 1:2 lylo, or 4)we lynch town day 2 and scum day 3. scenarios 1 and 4 are ideal, scenario 3 is manageable, and scenario 2 if obviously bad. my plan is designed to maximize the odds that we get scenario 1 or 4. a big part of avoiding the bad scenarios is by denying scum the full set of information they would need to maximize the value of the night 2 kill.

[xyzzy]

a reads list with 3 people each in two categories (scum/not scum) with our lists alphabetized

also, I feel like there should be consensus on whether this is a good idea before anyone actually posts a list. also, popcorning.

[xyzzy]

well, I guess the whole concept of popcorning kind of went by the wayside, but whatevs

(scum)/(not scum): (Alchemist21, Gamma Emerald, Mas y Menos)/(BlackStar, BTD6_maker, Comparing Realities)

[xyzzy]

okay so here's everyone's scum lists for easy reference:

Mas y Menos: (Comparing Realities, Gamma Emerald, xyzzy)
BlackStar: (Alchemist21, Comparing Realities, Gamma Emerald)
BTD6_maker: (BlackStar, Comparing Realities, Gamma Emerald)
xyzzy: (Alchemist21, Gamma Emerald, Mas y Menos)
Alchemist21: (BlackStar, BTD6_maker, xyzzy)
Comparing Realities: (Alchemist21, BlackStar, Mas y Menos)
Gamma Emerald: (Alchemist21, BlackStar, Mas y Menos)

and the number of times each person shows up as scum in lists:

Alchemist21: 4
BlackStar: 4
BTD6_Maker: 1
Comparing Realities: 3
Gamma Emerald: 4
Mas y Menos: 3
xyzzy: 2

so we should be looking at Alchemist21, BlackStar and Gamma Emerald.

Mas y Menos and Alchemist21 are the only ones who only have one person from that list on their lists; I'd be interested in hearing what each of them thinks about the other two. I'd also be interested in knowing what the three people on those lists think of each other.

[xyzzy]

oh, duh, you're right.

okay, so according to the methodology I'm using, this is the correct choice:

VOTE: Mas y Menos

[xyzzy]

I've never been in a game that didn't have one, so I just assumed that this one did too

this is a really bizarre thing to not notice, but it feels like a mistake that scum wouldn't make.

[xyzzy]

I generally doubt that things are that simple w/r/t night kills, Comparing Realities; while it's certainly possible that the whole thing was just an effort to eliminate the person who suspected then most, guessing what that reasoning might be is just that—a guess.

like, here's a few more possibilities: 1)scum were specifically trying to set up lynches on town players who rb was town reading who he might get in the way of lynching, 2)scum were trying to eliminate someone who wasn't as universally town read to cast suspicion on someone who everyone sees as town, 3)scum picked someone basically at random so that reasoning through why rb was chosen would lead to false conclusions, 4)some kind of setup-based gambit that is impossible for us to guess, 5)scum specifically wants us to look at rb's reads and derive bad info from them.

while it's certainly plausible that scum simply picked someone who was scum reading them (and perhaps even the most likely option), I think it's a mistake to just assume that that's the case.

[xyzzy]

BTD6 has got to be scum if only because nobody thinks he's scum.

this is some really invalid logic. it's possible that this is the case, but it certainly shouldn't be treated as implied.

[xyzzy]

• 7 Vanilla Townies
  
• 2 Mafia Goons (with a compulsive factional nightkill)
  
• If scum is lynched on Day 3, town automatically win. Otherwise, factions win or lose as normal.

The factional nightkill is compulsive to stop scum no-killing on N2 to push town onto evens.

EV is calculated as follows:

• There's a 2/9*1/7 = 2/63 chance of lynching scum D1 and D2, town win.
  
• There's a 7/9*5/7 = 35/63 chance of lynching town D1 and D2. You have a 40% chance of triggering the instant win, and a 60% chance of mislynching and losing.
  
• The remaining 26/63, there's a 4:1 situation going into D3, which has a (1/5 + 4/5 * 1/3) = 7/15 chance of a town win.

Total EV is (30 + 210 + 182) / (63 × 15) = 422 / 945, or about 44.66%.

Strategy here is fairly interesting, given that town may want to lynch a less likely scum candidate D2 (or even D1) in order to save a likely scumflip for their "desperation day" on D3, at the cost of scum getting a nightkill which they can use to influence how easily that player gets lynched (or spout WIFOM everywhere). It also makes bussing more viable early and less viable late than is usual in a 9p; I'm not sure whether people consider that a good thing or not.

Thanks to xyzzy, whose setups got me thinking along these lines.

Based on the cited thread and the credit to xyzzy in the OP of that thread seen here, it's probably more likely than not that xyzzy had already started working out some numbers for the game already, maybe even before the game even started, and only brought it up because No Lynching was discussed (which was also quick to come up in the cited thread). So just bringing up the math for it is probably null, and I certainly wouldn't make associatives based on it.

QFT. I went into the game already motivated to think about it in that light before the game even started; I would've been ready for that discussion regardless of any in-game factors.

I tend to focus a lot of energy when playing open setups to figuring out how to optimize town play in them; that's what i was doing yesterday, and it's what I've done today.

[xyzzy]

xyzzy, would you care to respond to my accusations instead of flat-out ignoring them like every other player I've raised objections to?

can you clarify what specifically you're asking me to respond to? I don't really see anything in your ISO directed at me that I've not responded to, but if there's something you'd like a more specific response to, feel free to let me know.

[xyzzy]

I personally seriously doubt that BTD6_maker is scum, so the theory that both scum are on the wagon is off the table imo

[xyzzy]

welp

[xyzzy]

welp

Any last thoughts to share?

nope.

[xyzzy]

welp

Any last thoughts to share?

nope.

I'll take that as a scum claim.

I just realized that "nope" counts as last thoughts. so does this post. shit, I'm real bad at not having last thoughts, huh

[xyzzy]

just realized that "nope" counts as last thoughts. so does this post. shit, I'm real bad at not having last thoughts, huh

Naah just bad.
-mas y

wait are you saying that I'm just "bad" as opposed to "real bad"? I guess that's technically a compliment, relatively speaking

[xyzzy]

[xyzzy]

That guy is creepy.

how dare you say that about Elijah Wood? have you no shame?

[xyzzy]

can any of you solve the Mystery of Elijah Wood? truly he will haunt you as you ponder the possibilities

[xyzzy]

the meaning of it is I was responding to Mas y Menos

[xyzzy]

what if the real scum is the friends we made along the way?

[xyzzy]

I'm way too lazy to do this edit myself but imagine if this image said "five tildes" instead of "two wieners":

[xyzzy]

what if the real scum is the friends we made along the way?

That's not a very nice thing to say about your friends.

friends, romans, scumbuddies, lend me your ears

[xyzzy]

what if the real scum is the friends we made along the way?

That's not a very nice thing to say about your friends.

friends, romans, scumbuddies, lend me your ears

Scum claim.

[xyzzy]

Release the scum PT, I want them to see our pre-game plan as soon as possible.

same

we didn't end up following said plan at all, but: that Elijah Wood gif was totally something we planned as a "please bus me, thanks" signal