The Thing

[rogue_alchemist]

My understanding is that we vote on two tests simultaneously, and at EoD we find out if either of the people we voted is a Thing or not. If at least one of the people voted was a Thing, they are killed; if there are still Things alive after that, we get another vote, and if that vote kills a Thing we get another, and so on until a test fails. Is that correct?

Just seeing the thread now, so I'll have a post with my thoughts later today. Still wrapping my head around the rules but a quick question on the voting extension: assuming we get a Thing and have another round to vote, do we get two votes in the extra round or one?

AvatarVecna is correct. You have 2 tests, if either one of them hits a thing, then you get 1 more test, assuming there is another thing alive to hit. Otherwise the game is over and town wins. If the extra test hits another thing then, then you get 1 more extra again. This pattern continues until either all things are tested (and thus town wins) or you hit a scientist.

[AvatarVecna]

I'm working on a big math post, but for the moment I'll say that it's extremely unlikely we've got just one Thing, but we're allowed to make two tests. If that were the case, there's a 10/55 chance the game ends at EoD1, compared to a 28/550 if there's 2 Things.

[CaoimhinTheCape]

Caerulea's suggestion

Spoiler

Show

Spoiler: This is the player list in alphabetical order.

Show

(Player list)

I think JeenLeen (I think it was him) was on to something when he suggested that we just test people in a set order, at least the first go around. We are, in <=10 rounds, guaranteed to catch at least all the original things. In addition, based on who the thing is, we will have a decent idea of which people have been corrupted. After all, it makes little sense for the thing to pick someone who is about to be tested.

That's been a long-winded way of saying I vote to cleanse Apogee1 of corruption. May he remain unharmed and human.

- - - Updated - - -

We get two tests, right rogue_alchemist? Do you want us to vote for both tests at the same time, for instance like this: AvatarVecna?

- - - Updated - - -

Rereading the OP I have realised that we only get one test per day. That makes the 'in order' plan not quite as effective, but I'm not sure what a better one is.

So, from the original post, Cae suggests that we vote in a set order, saying we'll get a thing guaranteed by round 10. This math seems like it's based on one test though, since two votes gives us a thing guaranteed in the 6th round or earlier.

I don't like the pre-decided plan since we're telegraphing what we're doing to the Things, but with two votes per round it technically works as we'd get to a correct lynch at LyLo and then could chain a couple kills to keep the game going (unless Valmark happened to be the original thing).

So we have, in the worst case with only 1 thing, 5 tests to run (more if we ever hit a thing). And to win, we have to double up on successful thing tests a number of times equal to the number of things at the time of the test that hits, so if they ever get to three things we probably just lose unless they make each other really obvious.
I think having a more "optimal" mechanical strategy than pure chance is impossible, because since no private communication the things can always plan against what the plan would be. Or you know, you and/or someone else proposing a plan is a thing

Fair point about the plan, but this is when we assumed that it was one lynch per day.

I don't think this is a good idea. For one thing, if any initial Things are within the second half of the list, we're just giving the game to them, because they can just infect the person we just tested every night, and by the time we reach them over half the players will be infected and they'll control the game. So this gives the Things a 50-50 chance of auto-winning, or a 75% chance of winning if there are two Things to start with. That doesn't sound like a very good plan.

Also, if we agree to test alphabetically every day, that will most likely kill a lot of the discussion, which will make it really hard to find suspicious people, so even if we don't lose by the end of the first set of tests, we'll have much less information to go on after that.

As such, Caerulea. This isn't a very strong suspicion, since of course townies can come up with bad plans, but I prefer this to voting randomly.

I like this post overall, even if part of the analysis is from when we thought 1 vote. Like the point about this killing discussion.

Elenna and Apogee1 raise excellent points, about my plan being a guaranteed lose for town if the thing(s?) are anywhere in the second half of the list I provided. To mildly reduce how dumb I look, I proposed it thinking that we got two shots each day, more if we killed one, which would allow us to clear the list in 5 days maximum. I also planned (if it was followed) to propose we test AvatarVecna and Apogee1 again the next day, as they would suddenly have become the most likely targets for thingification. In general I think repeat tests aren't a bad idea.

It was obvious that she thought we'd have two votes from her first post, but the second part I could see either as genuine town or a Thing backtracking.

Caerulea, any specific reason for why you didn't just use rogue's list that would have led to you getting tested Day 1 according to the double voting theory but rather made another one?

Cause alphabetical lists look nicer. Not really any other reason besides that. If y'all want I'm fine with getting tested day one. I would prefer someone else only because I know I'm town, but from y'alls perspective you don't have certainty about my alignment.

Reading this as null. Yes, Caerulea moved herself off Day 1 of a list... only to put herself on Day 2?

EDIT: OK, this argument makes sense. In a game with 1 Thing game the list change is a difference between auto-losing and having a shot. In a 2 Thing game the list change doesn't make as much of a difference. I'm assuming 2 Things, given that we have 2 votes per day and a 1 player team that could lost randomly on Day 1 wouldn't be fun.

Note that the above only works for a double test game. If we only get one test, we have to do things way differently. That said, voting in a strict order isn't essential to the plan, so much as the act of announcing in advance who we're going to test next. As long as the Things know who we're eyeballing, they can't convert that person without throwing us a free Thing kill. We could get similar extra testing strength by having everyone submit a ranked preference list of who they think is suspicious, or by pairing people off and announcing that (absent any good leads) we'll decide who to test by flipping a coin and randomly testing someone from Group A, then from Group B, then from Group C, etc. until we find a Thing or get a good lead. Either way, it narrows the range of possible conversion targets for the Things by excluding some of the Scientists from the pool of reasonable conversion targets.

Cutting down the post to the main analysis part, but I like this so far. The ranked list is helpful for analysis and I like the idea of making groups and picking someone random from them, rather than a straight order. This assumes we don't get anywhere with analysis (which, ya know, is the best option) but I like the different ideas floated.

As it relates to Caerulea's post, it's a good analysis of the idea which works OK for a two vote game but not for a one vote.

The Outsider

Spoiler: The Outsider

Show

Hmm. Ordinarily, I'd RNG vote from the whole list to start things off. But since the vote only kills evils, I honestly feel a bit better about starting a wagon early. So I'm going to vote for the first person who received a vote: AvatarVecna.

Completely ignores discussion on Caerulea's plan.

I'm going to switch my vote to JeenLeen. I see the point you're making about AV and agree in theory, but I can't deny that it would be an excellent way for a Thing to protect another Thing. This seems like it might be a stretch, though, so I might change it later. On another note, I'll gladly vote for The Outsider as my second vote.

Also, I like the idea of ranked lists. I'll make one after I've had a bit more time to observe, because I'm very slow when it comes to these games.

It's only the beginning of the day so I'm not worried about the ranked list yet, but there was other discussion going on that you didn't touch on at all.

Jeen Leen

Spoiler

Show

I have an argument against testing AvatarVecna. Namely, she so often gets wagoned early in a game that I could see the Narrator moving the Thing from AV if she was randomly chosen for it.

I also have an argument against testing Caerulea. With ten players, we likely have 1 Thing. It seems very risky to draw heat to oneself by proposing a voting scheme if you are that Thing, especially this early in the Day (when you could be waiting for someone else to do such and draw heat.) I like Valmark's read that Caerulea put herself not in the top spot, but I also doubt Caerulea would have made those posts if she were the Thing.
I could be tricked by a strong WIFOM hope -- maybe Caerulea is the Thing and hopes we don't test her since the Thing obviously wouldn't risk making a plan -- but I kinda doubt it.

I'll start with Valmark. His questioning of Caerulea's motives is a good and legitimate one any Town could make, but it's also a good way for the Thing to try to last until D2.

I disagree with the AvatarVecna reasoning from an out of game perspective. That's going on an assumption of the narrator giving out roles in a way that isn't random which (unless the game is explicitly says this) isn't exactly fair.

I understand the defense of Caerulea (townie suggesting a flawed plan over her being a Thing) but something about the wording bothers me.

The vote on Valmark confuses me. If I'm reading your reasoning right... Valmark's questioning is good and Townish, but anyone could do it so Valmark is a Thing?

Self-voting is an interesting strategy in this game. Since it doesn't lead to death, there's less joking-feel about it and less risk to it (assuming you are Town). It is also handy to verify oneself as Town D1... although it's not that meaningful in a cult-type game like this one.

Also, want to stress that we probably shouldn't discuss how to alter our voting strategy based on who is tested yet... though, well, I can think of one potential idea where discussing it early could help, but stating that ahead of time kinda ruins the idea, and I'm not sure it's a good idea anyway. (If someone does think of a good reason to discuss it early, I could see it, but it seems unwise akin to how talking at Night is usually an unwise idea.)

Self Voting is something I'll touch on below. For the second part, we do need to thread the line of not giving the Things too much info vs getting the town on board with good plans by explaining them.

Apogee

Spoiler

Show

I think Elenna's post was very natural and flowed between ideas well enough so I'd say she's a scientist for now

- - - Updated - - -

Wait I think I misremembered something re: Elenna. Still probably a scientist but not as much as I thought rip.

I would, however, point towards TheOutsider and JeenLeen

I'd like more explanation for the votes (even if I agree with them) and what you misremembered regarding Elenna?

AvatarVecna

Spoiler

Show

AvatarVecna and Caerulea

I'd like more words from AV. I could guess your reasoning for the votes but won't put words in your mouth.

Book Wombat

Spoiler

Show

First thing I see when I open the thread is a bunch of analysis, something I'm horrible at. Welp. Well I think RNG votes on Day 1 are fine.

EDIT: A little later will be my normal daily posting time.

Even if you're not great on analysis, there should be enough here for more than a random vote, yes? Looking for a little more when you have a chance to post, even if it isn't a full analysis.

---

Now that I'm caught up. In almost all cases I don't see how voting yourself is helpful to the town.

• As anyone, we don't get as much analysis from who you're willing to vote.
• As a townie, this is a known wasted test.
• As a townie who is suspected and wants to prove your innocence... well, you're suspicious and people are already voting you.
• As a Thing, this can be done to get brownie points if you're not in danger of being tested.

As for my list:

1. JeenLeen - Has given decent reasons not to vote the people he's voting.
2. The Outsider - Ignores Caerulea's plan and the discussion of it. Posted a few times, so avoiding that seems suspicious.
3. Caerulea - Made a plan under the idea of 2 votes a day (which wasn't perfect). Points for trying, but the arguments for her being a Thing are also decent.

4. PartyOfRouges - No posts.
5. Book Wombat - Don't love completely random votes at the moment, especially when they line up with the vote leader anyway.
6. AvatarVecna - One post on rules clarification and another with only votes. Don't like it but AV has been quiet early games as town in the past.

7. Apogee1 - Not as much to go on, but OK for now.
8. Valmark - Not as sure as the people below but nothing pinged me as evil yet.

9. Grek - Good analysis, town read so far.
10. Elenna - Good analysis, town read so far. Has a few extra posts so gets the bottom spot.


I separated it out into groups just to give people a little separation. I'm happy with any of the top three as a vote for now, but I think it's ideal to have at least 3 competing wagons since 2 people are tested each day.

Could be convinced to vote for 4-6 today, depending on how it goes. Don't see myself voting for anyone I ranked 7+.

---

Vote Count:

Avatar Vecna (2): Valmark, AvatarVecna
Caerulea (6): Elenna, Grek, JeenLeen, Valmark, AvatarVecna, Book Wombat
Valmark (1): JeenLeen
CaoimhinTheCape (2): Caerulea, Book Wombat
The Outsider (4): Caerulea, Apogee1, The Outsider, CaoimhinTheCape
JeenLeen (4): Apogee1, The Outsider, Elenna, CaoimhinTheCape

Not voting: PartyOfRouges, PartyOfRouges, Grek

[Apogee1]

I'd like more explanation for the votes (even if I agree with them) and what you misremembered regarding Elenna?

Okay I have no idea how this happened but I somehow got it in my head Elenna like, switched votes between three people in succession

Which apparently no one else did in this game either so I just fabricated an entire post in my head I thought existed? Idk

I still like that post

JeenLeen was because the townreads were maybe forced or TMI'd

Outsider for reasons you've described

yeah

[Book Wombat]

Even if you're not great on analysis, there should be enough here for more than a random vote, yes? Looking for a little more when you have a chance to post, even if it isn't a full analysis.

Hrrrmmm, since The Thing wants to infect those who are least likely to be (re)tested, from N1 onwards one test could to an untested (suspicious) person and randomly pick one of the previously negative tested? On D3 or D4 both votes could go to previously tested if 0-1 were found and if more votes would go as normal?
Just a random idea, probably not thought through properly.

[Valmark]

There is to say that proving one's own innocence is kind of pointless since you can get converted over night.

[AvatarVecna]

Spoiler: All possible days

Show

A: 10v1

1. 45/55 it's still 10v1, and becomes 9v2 in the night.
2. 10/55 it's 10v0. Town victory.

B: 9v2

1. 360/550 it's still 9v2, and becomes 8v3 in the night.
2. 162/550 it's 9v1, and becomes 8v2 in the night.
3. 28/550 it's 9v0. Town victory.

C: 8v3

1. 1260/2475 it's still 8v3, and becomes 7v4 in the night.
2. 864/2475 it's 8v2, and becomes 7v3 in the night.
3. 312/2475 it's 8v1, and becomes 7v2 in the night.
4. 39/2475 it's 8v0. Town victory.

D: 7v4

1. 7560/19800 it's still 7v4, and becomes 6v5 in the night.
2. 7056/19800 it's 7v3, and becomes 6v4 in the night.
3. 4032/19800 it's 7v2, and becomes 6v3 in the night.
4. 1008/19800 it's 7v1, and becomes 6v2 in the night.
5. 144/19800 it's 7v0. Town victory.

E: 6v5

1. 18900/69300 it's still 6v5, and becomes 5v6 in the night. Thing victory.
2. 22680/69300 it's 6v4, and becomes 5v5 in the night. Thing victory.
3. 18480/69300 it's 6v3, and becomes 5v4 in the night.
4. 6930/69300 it's 6v2, and becomes 5v3 in the night.
5. 1980/69300 it's 6v1, and becomes 5v2 in the night.
6. 330/69300 it's 6v0. Town victory.

F: 8v2

1. 252/405 it's still 8v2, and becomes 7v3 in the night.
2. 128/405 it's 8v1, and becomes 7v2 in the night.
3. 25/405 it's 8v0. Town victory.

G: 7v3

1. 1512/3240 it's still 7v3, and becomes 6v4 in the night.
2. 1176/3240 it's 7v2, and becomes 6v3 in the night.
3. 483/3240 it's 7v1, and becomes 6v2 in the night.
4. 69/3240 it's 7v0. Town victory.

H: 6v4

1. 3780/11340 it's still 6v4, and becomes 5v5 in the night. Thing victory.
2. 4032/11340 it's 6v3, and becomes 5v4 in the night.
3. 2625/11340 it's 6v2, and becomes 5v3 in the night.
4. 756/11340 it's 6v1, and becomes 5v2 in the night.
5. 126/11340 it's 6v0. Town victory.

I: 7v2

1. 84/144 it's still 7v2, and becomes 6v3 in the night.
2. 49/144 it's 7v1, and becomes 6v2 in the night.
3. 11/144 it's 7v0. Town victory.

J: 6v3

1. 210/504 it's still 6v3, and becomes 5v4 in the night.
2. 189/504 it's 6v2, and becomes 5v3 in the night.
3. 90/504 it's 6v1, and becomes 5v2 in the night.
4. 15/504 it's 6v0. Town victory.

K: 5v4

1. 140/504 it's still 5v4, and becomes 4v5 in the night. Thing victory.
2. 175/504 it's 5v3, and becomes 4v4 in the night. Thing victory.
3. 135/504 it's 5v2, and becomes 4v3 in the night.
4. 45/504 it's 5v1, and becomes 4v2 in the night.
5. 9/504 it's 5v0. Town victory.

L: 6v2

1. 630/1176 it's still 6v2, and becomes 5v3 in the night.
2. 432/1176 it's 6v1, and becomes 5v2 in the night.
3. 114/1176 it's 6v0. Town victory.

M: 5v3

1. 420/1176 it's still 5v3, and becomes 4v4 in the night. Thing victory.
2. 450/1176 it's 5v2, and becomes 4v3 in the night.
3. 255/1176 it's 5v1, and becomes 4v2 in the night.
4. 51/1176 it's 5v0. Town victory.

N: 5v2

1. 60/126 it's still 5v2, and becomes 4v3 in the night.
2. 50/126 it's 5v1, and becomes 4v2 in the night.
3. 16/126 it's 5v0. Town victory.

O: 4v3

1. 30/105 it's still 4v3, and becomes 3v4 in the night. Thing victory.
2. 40/105 it's 4v2, and becomes 3v3 in the night. Thing victory.
3. 28/105 it's 4v1, and becomes 3v2 in the night.
4. 7/105 it's 4v0. Town victory.

P: 4v2

1. 30/75 it's still 4v2, and becomes 3v3 in the night. Thing victory.
2. 32/75 it's 4v1, and becomes 3v2 in the night.
3. 13/75 it's 4v0. Town victory.

Q: 3v2

1. 6/20 it's still 3v2, and becomes 2v3 in the night. Thing victory.
2. 9/20 it's 3v1, and becomes 2v2 in the night. Thing victory.
3. 5/20 it's 3v0. Town victory.

R: Town has won.

S: Thing has won.

I've done a good deal of math today, figuring out exactly how this would go if we let the dice decide how things play out. The long and short of it is that if we started 10v1, there's a ~3/8 chance village wins through random guessing. If we started with 9v2, or we fail to find the 1 today and we're at 9v2 tomorrow, we're now at ~1/4 chance village wins. That it's so stacked against us to start out just means that we need to be really on-point with our tests, if we want a good chance at winning. The good news is that, if we do manage to pop a Thing, we've got their vote history we can analyze to figure out who might be one of their buddies, and their placement in the order. The sole benefit of the "rolling randomly to determine tests" approach is that the Things are incapable of anticipating or influencing the result, but that just means that they can't really change our chances of losing anymore than we can if we do that method, and we really shouldn't so it was kinda all just a waste of time. >.>

A few other interesting tidbits:

1) Town always has a chance, no matter how far behind, to win outright on any given day. This is fairly obvious from the basic set-up, but it's amusing to me.

2) If a new day starts, we have at least 2 Things left.

I'm gonna start analyzing arguments and figuring out my own approach to playing the game from both sides.

[AvatarVecna]

Spoiler: Reactions in the order they occurred

Show

Spoiler: This is the player list in alphabetical order.

Show

Apogee1
AvatarVecna
Book Wombat
Caerulea
CaoimhinTheCape
Elenna
Grek
JeenLeen
PartyOfRouges
The Outsider
Valmark

I think JeenLeen (I think it was him) was on to something when he suggested that we just test people in a set order, at least the first go around. We are, in <=10 rounds, guaranteed to catch at least all the original things. In addition, based on who the thing is, we will have a decent idea of which people have been corrupted. After all, it makes little sense for the thing to pick someone who is about to be tested.

That's been a long-winded way of saying I vote to cleanse Apogee1 of corruption. May he remain unharmed and human.

- - - Updated - - -

We get two tests, right rogue_alchemist? Do you want us to vote for both tests at the same time, for instance like this: AvatarVecna?

- - - Updated - - -

Rereading the OP I have realised that we only get one test per day. That makes the 'in order' plan not quite as effective, but I'm not sure what a better one is.

First off, it doesn't help that Caerulea out of the gate is casting doubt on how many tests we have. This isn't a good look and it's almost certainly why Grek only has a single vote.

Okay so we've got somebody reordering the list in a way that takes themselves out of initial testing, but puts them in the second round of testing so it's not that great a delaying action. Only useful if Caerulea is the 1 in 10v1 and is desperate to not get tested first (and thought town would be using this plan). I'm not so happy about testing in a set order for five days though. That's 10 base tests, then at least one that catches an original Thing, so we're looking at 11-12 tests across five days.

D1: Test A/B.
N1: conversion
D2: Test C/D.
N2: conversion
D3: Test E/F.
N3: conversion
D4: Test G/H.
N4: conversion
D5: Test I/J.
N5: conversion

At that point, we've tested 10/11 people. If we started 10v1, we've either found no original Things (in which case K must be a Thing), or we found an original Thing, go an extra test, and tested K as well, so going into D6, everybody's been checked. The problem is, there's been five conversions in that time.

If we started 9v2, we've definitely tested 12 people over the course of 5 days, which gets the two original Things and has a 4/10 chance of getting a third. If we took out 2, we're going into D6 5v4, which is LYLO. If we took out 3, we're going into D6 6v3, which gives us some breathing room.

In conclusion, the plan put forth for the five-day forecast is only a good idea for town if you're absolutely sure we're 9v2 right now, and even then it's not great unless we're lucky. If we're 10v1 to start, this plan ruins us unless we get lucky testing the 1 person today, or if the Things panic and convert somebody we haven't tested yet. This feels like a Thing suggesting a bad plan on purpose far more than a townie accidentally suggesting an awful plan.

Both pieces together make a solid argument for Caerulea being the 1 in a 10v1 situation. None of this is particularly new observation, of course.

---

I have an argument against testing AvatarVecna. Namely, she so often gets wagoned early in a game that I could see the Narrator moving the Thing from AV if she was randomly chosen for it.

Since I'm currently a townie, I'll say that I agree with this conclusion ("we shouldn't test AV yet"), but not with the logic behind it ("because the narrator would've avoided giving AV the scumrole"). Given that everything is public in this game, and how close to home this hits, I think I'll keep my reasoning to myself on this one.

I also have an argument against testing Caerulea. With ten players, we likely have 1 Thing. It seems very risky to draw heat to oneself by proposing a voting scheme if you are that Thing, especially this early in the Day (when you could be waiting for someone else to do such and draw heat.) I like Valmark's read that Caerulea put herself not in the top spot, but I also doubt Caerulea would have made those posts if she were the Thing.
I could be tricked by a strong WIFOM hope -- maybe Caerulea is the Thing and hopes we don't test her since the Thing obviously wouldn't risk making a plan -- but I kinda doubt it.

Bolded/underlined for emphasis. "We have ten players, therefore we probably have one Thing." We have 11 players. That still means there's a good chance we have just one Thing, but this feels like a scumslip to me - this honestly looks like JeenLeen had "11 players, and I know only one is scum cuz it's me" in his head, and he intended to say something like "Im guessing ten townies and one Thing", but when he wrote it down he accidentally said ten players instead of eleven. That feels more like a mistake scum would make than a townie.

I'll start with Valmark. His questioning of Caerulea's motives is a good and legitimate one any Town could make, but it's also a good way for the Thing to try to last until D2.

I'm not the first to point out how weird this logic is, and that makes three weird takes in one JeenLeen post.

---

Note that the above only works for a double test game. If we only get one test, we have to do things way differently. That said, voting in a strict order isn't essential to the plan, so much as the act of announcing in advance who we're going to test next. As long as the Things know who we're eyeballing, they can't convert that person without throwing us a free Thing kill. We could get similar extra testing strength by having everyone submit a ranked preference list of who they think is suspicious, or by pairing people off and announcing that (absent any good leads) we'll decide who to test by flipping a coin and randomly testing someone from Group A, then from Group B, then from Group C, etc. until we find a Thing or get a good lead. Either way, it narrows the range of possible conversion targets for the Things by excluding some of the Scientists from the pool of reasonable conversion targets.

In order to put my money where my mouth is regarding those ranked preference lists, here's mine:

1. Caerulea, for what should be obvious reasons.
2. AvatarVecna, both because of the existing votes and because Caerula's list also removes them from immediate testing.
3. TheOutsider, for bandwagoning.
4. CaoimhinTheCape, for being quiet.
5. PartyOfRouges, another quiet one.
6. Book Wombat, for being quiet with a possible excuse.
7. JeenLeen, because reverse WIFOM.
8. Valmark, because WIFOM.
9. Elenna, for pointing out that the fixed order thing is a bad idea with one test per day.
10. Apogee1, because he did the same as Elenna, but first.
11. Grek, because I'm obviously not a Thing.

I'm liking the ranked-choice voting approach. I will say that 2 only makes me look suspicious if you assume the setup is 9v2 with AV/Caerulea as Things and that we only get one vote per day. If we get two votes (as Caerulea has said she thought was the case when the post was written), I'd still be tested D1. And more importantly, as you alluded to, if we started 9v2 with 1 vote per day, we're basically screwed no matter what we do, especially with how untalkative half the players are.

---

In this case, my preferance list is as follows:
1. CaoimhinTheCape (Haven't spoken yet. I feel like a thing would most likely not speak as much early, to avoid being the topic of discussion. They get a very large advantage by flying under the radar, and I don't think we generally vote people who are quiet early.)
1. PartyOfRouges (see above)
2. BookWombat (see above, less so because new player)
3. The Outsider (Made a very small contribution who's only effect was to wagon someone who isn't them. Quite suspicious.)
3. AvatarVecna (As above, but it's AV and reading her is hard)
4. Apogee1 (Pointing out that what I suggested is bad (and it is) under the one test per day system, but no investigation really. Easy way for a thing to appear without risk.)
4.5. Elenna (Pretty much same as Apogee1, but slightly less)
5. Grek (Suggestion of ranked preference lists seems good, even if I don't completely understand her proposal)
6. JeenLeen (I like their thoughts, and they are doing some investigative stuff. I don't agree about AV, she might get wagoned but I don't feel like she gets voted off day one that frequently. Also rogue_alchemist might want to give the starting thing to a skilled player)
10. Valmark (Excellent point regarding my list, and is being investagative. I don't think a thing would be that bold.)
11. Caerulea (I am not a thing, and so voting me wastes a test)

Voting AV because I would prefer testing her to me, and Cao in case anyone else wants to try and vote someone who's quiet.

EDIT: Switched vote to The Outsider because they seem quite thingy.

Wow what a list. Okay so first off:

1) Caerulea's numbering of the list is off, and gives the impression they were going here and there to edit it. Normally that kinda obvious mid-post edit is a telltale scumsign since scum tend to deliberate on their posts a bit more to avoid giving stuff away, but with a list like this it's kinda expected that it won't be written in the perfect order the first time, so idk.

2) I think it's really weird that Caerulea initially wanted to vote me D1 (which I still think is a bit of a bad idea, but I won't say why), but also...this is Caerulea's personal list of "here's the order I want to test people in"...and she's not voting for numbers 1 and 2? She's voting from the middle of the list? And both votes were on quiet players (at least, quiet at the time) that wouldn't provide anything to analyze afterwards regardless of how they flipped?

3) Caerulea's thoughts on JeenLeen's stuff feel really weird to me.

---

Self-voting is an interesting strategy in this game. Since it doesn't lead to death, there's less joking-feel about it and less risk to it (assuming you are Town). It is also handy to verify oneself as Town D1... although it's not that meaningful in a cult-type game like this one.

I do have some thoughts (which Grek already posted in recruitment) about how Things are likely to react to who gets tested, but I don't feel like rehashing or thinking through them would be beneficial to Town at this moment. I'm sure the Thing(s) are thinking about it, but no reason to add extra thoughts they can bounce ideas off of or get inspired by. (I'm a little hesitant saying even this much, but I think it's obvious that the Thing(s) would be thinking about it, or at least would once Night starts, so no harm in stating that. Also, want to stress that we probably shouldn't discuss how to alter our voting strategy based on who is tested yet... though, well, I can think of one potential idea where discussing it early could help, but stating that ahead of time kinda ruins the idea, and I'm not sure it's a good idea anyway. (If someone does think of a good reason to discuss it early, I could see it, but it seems unwise akin to how talking at Night is usually an unwise idea.)

Part of the reason talking at night is usually frowned upon is because there's private communication available to set things up so discussing things openly should be done as an attempt to throw off the enemy, or it shouldn't be done. I'm not sure how fond I am of the sentiment that we should have our usual "don't talk at night" rule for day-talking, especially since there's no private comms. Our options are to sit here quietly and just hope that all the other non-Thing players are all reaching the same conclusions you are for how to go about locating all Things before they overwhelm us, OR we can talk things out publicly. See who makes fishy arguments, or fishy counterarguments. See who tries to shift the vote in weird directions for unexplained reasons. Communicating might clue the Things in to our plans, but a lack of communication will doom us even more thoroughly.

---

Because it's different from saying that you aren't Town and wouldn't lynch yourself with a subjective list. I'm voting Caerulea exactly because their original list (back before Grek's idea) made them not the day's lynch while they seemed to be ignoring the list we all had that made them the day's lynch target.

Gut read on this argument is that Valmark is town. I feel like scum wouldn't push to make sure the miscommunication was untangled in everyone's heads.

---

Now that I'm caught up. In almost all cases I don't see how voting yourself is helpful to the town.

• As anyone, we don't get as much analysis from who you're willing to vote.
• As a townie, this is a known wasted test.
• As a townie who is suspected and wants to prove your innocence... well, you're suspicious and people are already voting you.
• As a Thing, this can be done to get brownie points if you're not in danger of being tested.

Analysis of who you're willing to vote is only a small part of the analysis this game should include. Talking through plans and seeing who makes fishy arguments is a big part of that too, and if somebody wants their arguments to be taken seriously, self-voting and actually making the wagon happen is a good way to show people your arguments can be trusted and taken seriously - self-voting is a way to take you out of the middle-zone between "I trust this person's judgement" and "I think this person is so shifty that of course I'll vote them". Being untrusted and yet unvoted is a frustrating position that I'm unfortunately rather familiar with, so I'm trying to get ahead of the curve a bit. Granted, I still think I'm a bad person to test today, for multiple reasons.

1. JeenLeen - Has given decent reasons not to vote the people he's voting.
2. The Outsider - Ignores Caerulea's plan and the discussion of it. Posted a few times, so avoiding that seems suspicious.
3. Caerulea - Made a plan under the idea of 2 votes a day (which wasn't perfect). Points for trying, but the arguments for her being a Thing are also decent.

4. PartyOfRouges - No posts.
5. Book Wombat - Don't love completely random votes at the moment, especially when they line up with the vote leader anyway.
6. AvatarVecna - One post on rules clarification and another with only votes. Don't like it but AV has been quiet early games as town in the past.

7. Apogee1 - Not as much to go on, but OK for now.
8. Valmark - Not as sure as the people below but nothing pinged me as evil yet.

9. Grek - Good analysis, town read so far.
10. Elenna - Good analysis, town read so far. Has a few extra posts so gets the bottom spot.

I like this list. It feels solid, and they're actually voting their top two suspects, unlike some people.

---

Hrrrmmm, since The Thing wants to infect those who are least likely to be (re)tested, from N1 onwards one test could to an untested (suspicious) person and randomly pick one of the previously negative tested? On D3 or D4 both votes could go to previously tested if 0-1 were found and if more votes would go as normal?
Just a random idea, probably not thought through properly.

Even if it works, it doesn't. At that point we're essentially randomly voting two people with extra steps, complete with re-testing people who've already been tested (maybe even a lot). Randomly testing people in this manner is a recipe for town losing, although it's at least not quite as bad as fully-random test assignment, nor is it as disastrous as Caerulea's idea in a 10v1 setup. I'm not sure how to feel about this, but my gut says it's just a less experienced player making a slightly bad call.

Hard Scum Lean
1) Caerulea, most every post feels weird and makes a fishy argument. If there's anybody I'd feel comfortable testing today because I think they're actually scum, it's Caerulea.

2) JeenLeen has several moments that feel off and weird and I'd rather get them tested.

3) The Outsider's lack of comment on the Caerulea stuff is really weird, and it flew under the radar for me until Cao mentioned it.

Soft Scum Lean
4) PartyOfRouges is inactive so far. I dislike this, especially for this game, and especially for personal strategic reasons.

5) Apogee1's weird argument with Valmark is giving me bad vibes.

Null
6) AvatarVecna. That's all I'm saying about it at this point - either you get it or you don't.

7) Book Wombat's only had a few posts, and the one with actual gametalk rather than RNG feels like a bad argument, but more an innocent mistake than malicious deception. Still, far less solid than my town leans.

Soft Town Lean
8) Elenna's posts feel solid, and she's voting JeenLeen and Caerulea from the looks of it, which feels a-ok in my book.

9) Grek's single post here has some solid analysis, and including an argument from the recruitment thread is a great way to get around the "but how do we know we can trust their arguments" issue I brought up earlier.

10) CaoimhinTheCape's analysis is solid as usual.

Hard Town Lean
11) Valmark's posts feel genuinely frustrated, and like he's actually trying to solve things without showing all his cards. I like it a lot.

[Grek]

List updated for the realization that we all get 2 votes:
1. JeenLeen, CaoimhinTheCape's points regarding the weirdness of Jeen's voting patterns have firmed up my resolve that Jeen is suspicious right now, to the point where I think I rate him above Caerulea. I also personally feel that discouraging discussion about future voting plans is a very Thing-strategy in this game.
2. Caerulea, because I still find the alphabetical vote list suspicious.
3. PartyOfRouges is still being quiet. They always do that, but it's still suspicious.
4. Book Wombat, because they're currently RNG voting. Subject to change if they post something more detailed.
5. The Outsider, for similar reasons. I'm not sure how to interpret a self-vote in this game.
6. Valmark, for similar reasons. While his posts have been decisive, detailed analysis is the most anti-Thing thing.
7. Elenna, for the same reason as Valmark, but with less decisiveness.
8. AvatarVecna, because my original reasons for voting AV only make sense if we had one vote.
9. CaoimhinTheCape, because their big analysis post makes me gives us lots of information to do detective work on.
10. Apogee1, they keep making quibbling posts about that they haven't been ruled out in ways that don't feel like a WIFOM take to me. Not a day one test, I feel.
11. Grek, because I'm still obviously not a Thing.


Regarding Caerula's list:
As AV points out, the weird numbering system is a little weird. But I don't personally find it Thing-leaning, because ranking everyone who hasn't posted equally and doing last minute edits to your post is also something I did, when I realized I accidentally listed AvatarVecna twice on my list. The list itself doesn't weem too weird, except for the fact that it pushes the non-random Thing assignment angle. Which I personally find iffy.

Regarding CaoimhinTheCape's list:
Obviously I found their arguments pretty convincing, given that I bumped JeenLeen up to my #1 spot based on them. But regarding the self-vote issue, I feel like the case for self-voting as display of conviction is being understated. Anyone can make an argument, but unless you've been confirmed as town (or have done like me and posted your take before finding out if you were a Thing or not), those arguments can't be fully trusted. I'd be interested to hear what other people think of self-votes.

Regarding AvatarVecna's math:
I hadn't realized that things were quite so dire as a 1/3 to 1/4 shot of Town Win, even with two votes. Dang.

[Elenna]

Spoiler: thoughts on posts, spoilered for length

Show

Not gonna bother quoting Caerulea's suggested plan, just commenting that it's less bad now that we know we have two votes, but still not a good plan. IMO any plan that involves blindly following a predetermined path without analysis is going to be bad (and also boring to play). Also their reordering the list is weird although I suppose moving from "tested D1" to "tested D2" is not that big of a difference unless they're the lone Thing.

A set order is definitely not the way to go. Because, in a worse case scenario:

10 town 1 thing d1 (1 test)
9 town 2 thing d2 (2 tests)
8 town 3 thing d3 (3 tests)
7 town 4 thing d4 (4 tests)
6 town 5 thing d5 (5 tests)

town loses at night

So we have, in the worst case with only 1 thing, 5 tests to run (more if we ever hit a thing). And to win, we have to double up on successful thing tests a number of times equal to the number of things at the time of the test that hits, so if they ever get to three things we probably just lose unless they make each other really obvious.
I think having a more "optimal" mechanical strategy than pure chance is impossible, because since no private communication the things can always plan against what the plan would be. Or you know, you and/or someone else proposing a plan is a thing

Of course we hopefully will be better than chance at deducing who things are.

Apogee gets townie points for mentioning that this was a bad plan - but, as he mentioned later, less townie points than they would have gotten, since they would have been the first to be tested under Caerulea's plan.

I would, however, point towards TheOutsider and JeenLeen

Wait, did you ever explain why you wanted to vote those two? At the time of this post there had been no suspicion on JeenLeen yet so it feels weird to me that you dropped this vote on them with no explanation.

Oh wait, I somehow missed the post where you explained. I basically agree with your reasons, but still don't like that you didn't explain when you were making the post, especially since nobody else had pointed at JeenLeen yet. If you thought they were suspicious, why didn't you give the reasoning so others would vote for them?

Because it's different from saying that you aren't Town and wouldn't lynch yourself with a subjective list. I'm voting Caerulea exactly because their original list (back before Grek's idea) made them not the day's lynch while they seemed to be ignoring the list we all had that made them the day's lynch target.

I kinda see your point, but to me Apogee's post read less like "don't test me" and more like "here's another way people could interpret my post and it's weird that nobody has pointed this out yet". Like, it could be suspicious, but it feels more to me like a townie who's honestly trying to provide observations/analysis?

I reckon there is some good testing scheme that will give us better odds on D1, but nothing seems inherently strong to me. Also, we don't want to ignore analysis in favor of a set program... and a set program necessarily has to react to who was tested prior... as was noted in recruitment.

I have an argument against testing AvatarVecna. Namely, she so often gets wagoned early in a game that I could see the Narrator moving the Thing from AV if she was randomly chosen for it.

I also have an argument against testing Caerulea. With ten players, we likely have 1 Thing. It seems very risky to draw heat to oneself by proposing a voting scheme if you are that Thing, especially this early in the Day (when you could be waiting for someone else to do such and draw heat.) I like Valmark's read that Caerulea put herself not in the top spot, but I also doubt Caerulea would have made those posts if she were the Thing.
I could be tricked by a strong WIFOM hope -- maybe Caerulea is the Thing and hopes we don't test her since the Thing obviously wouldn't risk making a plan -- but I kinda doubt it.

I'll start with Valmark. His questioning of Caerulea's motives is a good and legitimate one any Town could make, but it's also a good way for the Thing to try to last until D2.

I would be quite surprised and somewhat annoyed if the narrator changed who had been picked as the Thing(s) to make it non-random.

Hadn't noticed the 10/11 players issue until AV pointed it out but it does seem weird. Not a very strong argument against JeenLeen by itself but it definitely doesn't help when I'm already suspicious of them.

More importantly, I don't understand what you mean by "it's a good way for the Thing to last until D2. What is a good way? Arguing against Caerulea's post? Caerulea's idea would have had town test Valmark quite late in the game. Or are you saying that questioning/analyzing people is a good way for the Thing to last through the day? But that's exactly what townies should be doing. IDK, that comment is weird and I don't like it.

Note that the above only works for a double test game. If we only get one test, we have to do things way differently. That said, voting in a strict order isn't essential to the plan, so much as the act of announcing in advance who we're going to test next. As long as the Things know who we're eyeballing, they can't convert that person without throwing us a free Thing kill. We could get similar extra testing strength by having everyone submit a ranked preference list of who they think is suspicious, or by pairing people off and announcing that (absent any good leads) we'll decide who to test by flipping a coin and randomly testing someone from Group A, then from Group B, then from Group C, etc. until we find a Thing or get a good lead. Either way, it narrows the range of possible conversion targets for the Things by excluding some of the Scientists from the pool of reasonable conversion targets.

I've cut out a fair bit of this post but the point is I like the discussion of plans and the suggestion of ranked preference lists.

In this case, my preferance list is as follows:
1. CaoimhinTheCape (Haven't spoken yet. I feel like a thing would most likely not speak as much early, to avoid being the topic of discussion. They get a very large advantage by flying under the radar, and I don't think we generally vote people who are quiet early.)
1. PartyOfRouges (see above)
2. BookWombat (see above, less so because new player)
3. The Outsider (Made a very small contribution who's only effect was to wagon someone who isn't them. Quite suspicious.)
3. AvatarVecna (As above, but it's AV and reading her is hard)
4. Apogee1 (Pointing out that what I suggested is bad (and it is) under the one test per day system, but no investigation really. Easy way for a thing to appear without risk.)
4.5. Elenna (Pretty much same as Apogee1, but slightly less)
5. Grek (Suggestion of ranked preference lists seems good, even if I don't completely understand her proposal)
6. JeenLeen (I like their thoughts, and they are doing some investigative stuff. I don't agree about AV, she might get wagoned but I don't feel like she gets voted off day one that frequently. Also rogue_alchemist might want to give the starting thing to a skilled player)
10. Valmark (Excellent point regarding my list, and is being investagative. I don't think a thing would be that bold.)
11. Caerulea (I am not a thing, and so voting me wastes a test)

Voting AV because I would prefer testing her to me, and Cao in case anyone else wants to try and vote someone who's quiet.

EDIT: Switched vote to The Outsider because they seem quite thingy.

You switched from AV to Outsider, right? I agree with AV, I want to know why you're not voting the top two people on your list of suspicions. Also kinda confused why you went back and edited the Outsider vote into a previous post - it feels like you're trying to hide who you voted for which seems suspicious.

Also, want to stress that we probably shouldn't discuss how to alter our voting strategy based on who is tested yet... though, well, I can think of one potential idea where discussing it early could help, but stating that ahead of time kinda ruins the idea, and I'm not sure it's a good idea anyway. (If someone does think of a good reason to discuss it early, I could see it, but it seems unwise akin to how talking at Night is usually an unwise idea.)

This I generally agree with (although I might not have phrased it this way). I don't think we should be planning too far ahead for who we should test in the coming days. By all means think about it, but I don't really want to discuss options in public because it gives the Things a better idea for who they should avoid converting. Deciding who to vote tomorrow can wait until tomorrow.

Part of the reason talking at night is usually frowned upon is because there's private communication available to set things up so discussing things openly should be done as an attempt to throw off the enemy, or it shouldn't be done. I'm not sure how fond I am of the sentiment that we should have our usual "don't talk at night" rule for day-talking, especially since there's no private comms. Our options are to sit here quietly and just hope that all the other non-Thing players are all reaching the same conclusions you are for how to go about locating all Things before they overwhelm us, OR we can talk things out publicly. See who makes fishy arguments, or fishy counterarguments. See who tries to shift the vote in weird directions for unexplained reasons. Communicating might clue the Things in to our plans, but a lack of communication will doom us even more thoroughly.

I'm not saying we should avoid discussing strategy at all - I agree that's a terrible idea. All I'm saying is that we should save the majority of the "who do we lynch on day 2" discussion until day 2. Once day 2 comes we should definitely discuss it.

I'm going to switch my vote to JeenLeen. I see the point you're making about AV and agree in theory, but I can't deny that it would be an excellent way for a Thing to protect another Thing. This seems like it might be a stretch, though, so I might change it later. On another note, I'll gladly vote for The Outsider as my second vote.

Also, I like the idea of ranked lists. I'll make one after I've had a bit more time to observe, because I'm very slow when it comes to these games.

So you're saying because JeenLeen defended AV, that means that Thing!JeenLeen might be protecting Thing!AV? But wouldn't this make literally anyone who defends or townreads anyone else suspicious? Everyone should have people who they think are likely to be town. I mean, I also think JeenLeen is suspicious, but this is a weird argument to be making. Not a fan.

Regarding self-voting: I don't think it's useful very often. I guess I can see how someone who has suspicion on them might want to get voted so that their arguments will be more trustworthy, but all that would show is that the person's past arguments were coming from a townie, it doesn't prove anything about what they will be in the future. So if anything, self-voting in this game seems to me like it might be useful in later days, not D1.

My argument against her as a Thing has less weight if there's 2 Things, and with two tests I reckon there likely are 2 Things instead of just one. That is, I think her making a good Town-sounding case is less risky overall if there's 2 of them, since one dying D1 doesn't mean the entire game is a loss.

Ehhh I kinda see your point, but "my argument for her being town is less good than I thought it was" =/= "I think she's a wolf" so this still doesn't really explain why you decided to vote for her? Also I still don't get why you didn't say this when you first voted for her. Could still be a Thing who was caught out in a contradiction and made up some arguments to try and look less suspicious after the fact.

Hrrrmmm, since The Thing wants to infect those who are least likely to be (re)tested, from N1 onwards one test could to an untested (suspicious) person and randomly pick one of the previously negative tested? On D3 or D4 both votes could go to previously tested if 0-1 were found and if more votes would go as normal?
Just a random idea, probably not thought through properly.

Randomly picking people is generally a bad idea since it's basically guaranteed to get equal or worse results versus picking people based on analysis. Other than that... IDK, the general idea of splitting votes between tested and untested people is fine, but I don't really like the idea of only voting previously tested people D3 onwards. Not sure if this is a bad plan proposed by an inexperienced townie or a bad plan proposed by a Thing.

Okay, wall of text over, here's my list. Note that 7-10 are all fairly close together in terms of how much I suspect them.

1. JeenLeen - I don't like several of their posts, specifically the Caerulea vote after defending her, and their comments about AV and Valmark
2. Caerulea - Started off by proposing a bad plan, and nothing they've posted since has made me feel particularly better about them.
3. The Outsider - Has made a total of two posts. One was trying to start a wagon, which is a reasonable reason for a D1 vote I guess, but as Caoimhin mentioned there was stuff happening and Outsider avoided commenting on any of it, which feels like they were trying to fly under the wagon. I've explained my objections to their second post above. I will say that I think only one of Outsider and JeenLeen is currently a Thing, but of course that doesn't mean much for the future.
4. Book Wombat - Two posts, the first one of which is RNG (after a lot of stuff has been said that can be commented on, IMO random D1 votes are only okay up until the point where there's actually been discussion, and there had definitely been plenty of discussion before Wombat's RNG vote). The second one is not great either.
5. Apogee1 - Kinda going back and forth on them, their initial post about Caerulea's plan was nice but I don't like the unexplained-at-the-time vote on JeenLeen.
6. PartyOfRouges - Do they exist? Can they say something, please? If they're not going to say anything then there's no way to figure them out apart from testing. Although it's frustrating that even if we test them, they could be converted again the next night and we'll have the same issue, which is why they're not super high on my list right now.
7. AvatarVecna - A nice analysis post and some math. Looks pretty good (although nothing that wolf!AV couldn't fake).
8. Grek - Like Caoimhin, only one post so far but it's a nice, solid, helpful one. Putting them as slightly more suspicious than Caoimhin just because their post contained less analysis of players (as opposed to discussion of game mechanics and strategy, which IMO is generally easier for wolves to fake although I do really like the preference-list idea), but their post also came many hours earlier so they had less to work with.
9. CaoimhinTheCape - Nice analysis, I like their observation about Outsider ignoring the discussion, which I hadn't really noticed until then. Town lean.
10. Valmark - Talkative, good analysis, pointing out suspicious comments, I like it. Although I'm pretty sure I've town-read Valmark in several games where they were a wolf, for exactly those reasons.
11. Elenna - I'm not a Thing, but you knew I would say that.

[JeenLeen]

I did just miscount 11 as 10, but I totally get why that mistake is a good D1 reason to vote someone. I'd jump on that mistake if I were in your shoes.

Part of the reason talking at night is usually frowned upon is because there's private communication available to set things up so discussing things openly should be done as an attempt to throw off the enemy, or it shouldn't be done. I'm not sure how fond I am of the sentiment that we should have our usual "don't talk at night" rule for day-talking, especially since there's no private comms. Our options are to sit here quietly and just hope that all the other non-Thing players are all reaching the same conclusions you are for how to go about locating all Things before they overwhelm us, OR we can talk things out publicly. See who makes fishy arguments, or fishy counterarguments. See who tries to shift the vote in weird directions for unexplained reasons. Communicating might clue the Things in to our plans, but a lack of communication will doom us even more thoroughly.

I wasn't trying to discourage talking during the Day, but just specifically talking about us laying out how we are going to test people in the future lest we give away a strategy to the Things or help them figure out a good conversion strategy. There's some WIFOM-ish stuff I'd rather not discuss publicly yet, at least myself, but it's also pretty obvious stuff the Thing(s) are likely considering..

However, I'm persuaded that discussion is worthwhile for the "see who gives fishy arguments" to try to figure out who's the Thing. I know changing one's mind is a scummy thing, but I'm legitimately persuaded by good argument. I am fine if you test me today. I don't completely see how what I said was so scum-ready (except for the saying don't vote Caerulea then voting her, but I thought I explained that as I was for not voting her when I thought we just had 1 vote), but I'm also fine getting tested for folk to have analysis on who was willing to test me as opposed to others. (Again, it's nice that we don't die if tested and flip Town.)

I feel like I should probably go over my writing above and check it for phrasing and such, but don't really have the time or incentive. (And I realize that sentence sounds scummy, but throwing it out there to help later on reading of it.)

- - - Updated - - -

I still have a somewhat off feel of Valmark, and maybe that's just residue from reading him last game (or 2 games ago, forget which), but nobody else is voting him so I'll shift to Caerulea and AvatarVecna. I haven't checked the votecounts this morning, but those seem more likely to be meaningful than a sole vote on Valmark.
I was also persuaded against the Narrator not going with RNG if AV was picked randomly.

NOTE: I plan on doing both my votes in a pair for most if not all of this game, since it helps me keep track of it and maybe would help RA.

[AvatarVecna]

This isn't relevant fo my interests immediately, but...

@OP

Currently all votes are in one pool, with each player placing a vote on two people, and the two prople with tge most votes getting tested that day. If we feep strongly enough about a particular person getting tested today, can we vote for one person twice?

[CaoimhinTheCape]

Analysis of who you're willing to vote is only a small part of the analysis this game should include. Talking through plans and seeing who makes fishy arguments is a big part of that too, and if somebody wants their arguments to be taken seriously, self-voting and actually making the wagon happen is a good way to show people your arguments can be trusted and taken seriously - self-voting is a way to take you out of the middle-zone between "I trust this person's judgement" and "I think this person is so shifty that of course I'll vote them". Being untrusted and yet unvoted is a frustrating position that I'm unfortunately rather familiar with, so I'm trying to get ahead of the curve a bit. Granted, I still think I'm a bad person to test today, for multiple reasons.

But regarding the self-vote issue, I feel like the case for self-voting as display of conviction is being understated. Anyone can make an argument, but unless you've been confirmed as town (or have done like me and posted your take before finding out if you were a Thing or not), those arguments can't be fully trusted. I'd be interested to hear what other people think of self-votes.

OK, I can see what you guys are saying. There have definitely been games where suspicion just hung around on people but we didn't get a chance to test them (especially with having the person live if they're good). I don't think it's something I would do but there are reasons it could help.

Got a few other thoughts regarding the above but I think it's best to keep them to myself, at least until Day 2.

---

As for the other things that have happened since my post, Apogee answered (which is fine).

Book Wombat, I was more looking for why you chose people randomly when there are reasons to vote for someone.

AV's post was good, makes me feel better about her.




Not gonna do an updated list yet since it hasn't changed too much. I'm not a fan of voting AV right now though, so it would be the top half of my list that I'm willing to vote.

---

Vote Count:

Avatar Vecna (2): Valmark, JeenLeen
Caerulea (6): Elenna, Grek, JeenLeen, Valmark, AvatarVecna, Book Wombat
CaoimhinTheCape (2): Caerulea, Book Wombat
The Outsider (4): Caerulea, Apogee1, The Outsider, CaoimhinTheCape
JeenLeen (6): Apogee1, The Outsider, Elenna, CaoimhinTheCape, AvatarVecna, Grek

Not voting: PartyOfRouges, PartyOfRouges

[AvatarVecna]

Regarding AvatarVecna's math:
I hadn't realized that things were quite so dire as a 1/3 to 1/4 shot of Town Win, even with two votes. Dang.

I will say that my "wholy random testing" approach is a bit...unrealistic. The upside of the "100% random" approach is that it's impossible for the Things to predict or influence the results, both of which they could (and would want to) do for discussion-based tests. But 100% random approach also means we're not being very scientific about our testing - almost certainly we don't end up testing everybody, and we probably have more than a few repeats. Did a mockup 100% random game (ignore the first post, I forgot about some stuff in that one cuz I tried to do things quick). Initial Things are Caerulea and Grek:

D1 Tests: CaoimhinTheCape/The Outsider
N1 Conversion: JeenLeen (Things: Caerulea/Grek/JeenLeen)
D2 Tests: Elenna/The Outsider
N2 Conversion: Valmark (Things: Caerulea/Grek/JeenLeen/Valmark)
D3 Tests: Elenna/Grek/Book Wombat
N3 Conversion: CaoimhinTheCape (Things: Caerulea/CaoimhinTheCape/JeenLeen/Valmark)
D4 Tests: Elenna/Apogee1
N4 Conversion: AvatarVecna (Things: AvatarVecna/Caerulea/CaoimhinTheCape/JeenLeen/Valmark)

(EDIT: This is a B-->C-->D-->H-->S game, to use my earlier lettered days system. BCDHS occurs ~4% of the time, while ABCDHS - the same thing except we start 10v1 and failed the first tests - occurs ~3% of the time. It's a relatively common result among the almost 400 possible paths).

Conclusion: 5 town vs 5 things going into D5. Thing victory. Town's random testing is something the Things had no ability to influence or predict, but fully random like that is still mostly to town's detriment: as you can see, they only managed to test 6 people despite conducting 9 tests. Idk what Elenna did in this theoretical game to warrant getting tested every day after D1, but it must have been horrendous.

If we wanted the benefits of random testing (unpredictable/uninfluenceable), but didn't want to run such a high risk of repeated-testing screwing town over, we could make a slight alteration to the 100% random plan.

Spoiler: Better random voting

Show

For each vote, town rolls 2d11 (or whatever die is appropriate to the number of living players). If the die results are "two people we haven't tested yet", we pick which one to test. If the die results are "a person we've tested and a person we haven't", we test the one we haven't previously tested. If the die results are "two people we have tested already", we pick which one to test, even though we've already tested them both. We do that again with a d10 (or whatever) for the second test, and then again with a d9 (or whatever) for a possible third test if we found a Thing. This would massively cut down on the number of repeat testings, which should give town much better odds of hitting a Thing, but without giving the Things very much ability to influence who we vote for.

Of course, I've also yet to mention one of the major downsides to the "random rolling" approach, but you've probably pinged onto it if you read through that dice thread: if we just let the dice decide who gets tested and converted, couldn't we just play a whole game in like 5 minutes, the way I just did in that die thread? Well...yes. That's the downside. Letting the dice decide prevents the Things from having any say in how town plays, but it also prevents Town from having any say in how town plays. There's versions of WW/Mafia where rolling dice and going along with the results (only deviating to murder people who question the dice) is the play that results in town victory more frequently than any other playstyle, but it's also kinda missing the point, which is to actually play the game.

Finding a method of die-rolling that gives town the best odds could be one approach to this game, but even if one exists, I'm not sure I'd want to play it even if you could prove that it gave town a 99%+ chance of winning. I mostly wanted to figure out how good random testing could work for town because it gives a decent-ish idea of the default - if we guess worse than random, how likely are we to win? If we guess better than random, how likely are we to win? And the answer to both appears to be that the deck is stacked against us, which just goes to show how important analysis is going to be to this game in particular.

[Valmark]

OK, I can see what you guys are saying. There have definitely been games where suspicion just hung around on people but we didn't get a chance to test them (especially with having the person live if they're good). I don't think it's something I would do but there are reasons it could help.

Got a few other thoughts regarding the above but I think it's best to keep them to myself, at least until Day 2.

---

As for the other things that have happened since my post, Apogee answered (which is fine).

Book Wombat, I was more looking for why you chose people randomly when there are reasons to vote for someone.

AV's post was good, makes me feel better about her.




Not gonna do an updated list yet since it hasn't changed too much. I'm not a fan of voting AV right now though, so it would be the top half of my list that I'm willing to vote.

---

Vote Count:

Avatar Vecna (2): Valmark, JeenLeen
Caerulea (6): Elenna, Grek, JeenLeen, Valmark, AvatarVecna, Book Wombat
CaoimhinTheCape (2): Caerulea, Book Wombat
The Outsider (4): Caerulea, Apogee1, The Outsider, CaoimhinTheCape
JeenLeen (6): Apogee1, The Outsider, Elenna, CaoimhinTheCape, AvatarVecna, Grek

Not voting: PartyOfRouges, PartyOfRouges

My vote on AV was changed to Apogee1, although I want to move it on JeenLeen now. AV raised some pretty good points and it's better then anything Apogee1 did of potentially suspicious.

Now I'll go back to nursing an headache, sorry if I'm not writing. Apparently sleeping half the recommended time is harmful, who knew.

[JeenLeen]

The effort AV went to make me think she's Town, especially since she's advocating against many plans that could seem reasonable but end badly for Town. I wouldn't put it past a wolf!AV to try such for towncred, but it seems like a poor plan for the Things when Town could potentially be persuaded to a bad idea.

Leaving my votes as-is since I don't have a strong read on any other wagon, so I'll move my AV vote to The Outsider mostly since it's just the only competing wagon. Also The Outsider seems to do good at flying under the radar with any alignment. So The Outsider and Caerulea.

---

Also agree that it's more fun to actually play the game, even if we have a perfect (or nigh perfect) mathematical solution to the game.

[The Outsider]

Hoo boy. I am not going to do well with this game.
I don't have any good reason for my initial lack of commentary aside from "I'm bad at analysis and have a crippling fear of getting things wrong" (no pun intended). As already noted, my arguments so far weren't great arguments, and at this point I worry that any analysis I add is just going to end up parroting someone else's analysis. That said, I'm going to attempt to make a ranked list anyway, if for no other reason than to get my thoughts out of my head. From most to least sus:

Most sus-
1. The Outsider. Multiple people have pointed out that self-voting is a waste for either side, and I can see their point. At this point, however, I've done enough sus actions and been quiet enough to warrant a test. I've unwittingly made myself the best option.
2. JeenLeen. The thing I find most suspicious in any game is inconsistency, and Jeen has repeatedly ignored their own reasoning of why not to vote people in determining their votes. Keeping my vote on him.
3. PartyOfRouges, because they've been silent. I'm starting to realize that this is how they usually roll, but as a matter of principal I have to find it sus.
4. BookWombat: Random votes when there's reasoning to analyze is a suspicious maneuver. I did practically the same thing only with flimsy reasoning attached, and I'm now at the top of my own sus list.

Null sus-
5. Caerulea. People have made good arguments about their evil alignment, but I'm in agreement with JeenLeen's initial reasoning on this one: I don't think a Thing would have risked drawing attention to themselves like that, even if there were two of them. I can understand why people find their arguments fishy, and creating a new plan where they don't get tested does look a bit shady, but I can't help but feel like our attention is being diverted somehow.
6. Apogee1.
7. Valmark. Both of them having been making somewhat good points, but Apogee seems to have a rather limited presence in the thread and Valmark is impossible for me to read. So I'm putting them in the null category for the time being.
8. AvatarVecna. Their math and analysis goes over my head, but it always does that. What I can understand of it seems solid, and considering they're one of the best analyzers I've seen so far I almost have to trust them. Though going forward, that makes them a prime candidate for Thing conversion.

Least sus-
9. Elenna.
10. CaoiminhTheCape.
11. Grek. All three of these people are making solid analytical progress and calling people out for their more sus actions. Bonus points to Grek for having quiet people at near the top of their list, because quiet people are always sus.

[rogue_alchemist]

This isn't relevant fo my interests immediately, but...

@OP

Currently all votes are in one pool, with each player placing a vote on two people, and the two prople with tge most votes getting tested that day. If we feep strongly enough about a particular person getting tested today, can we vote for one person twice?

So technically you are voting once, testing that person, then voting again, but to save on time in the forum, I am having yall vote 2x at once. As such I will say that you can NOT vote 2x for the same person. I toyed with the idea of having yall do red for 1 vote and blue for the other to help me keep up with them, but I definitely appreciate people who are posting both votes every time, even if only 1 is changing. So if everyone could vote both votes every time, I would appreciate it.

[JeenLeen]

Hoo boy. I am not going to do well with this game.
I don't have any good reason for my initial lack of commentary aside from "I'm bad at analysis and have a crippling fear of getting things wrong" (no pun intended). As already noted, my arguments so far weren't great arguments, and at this point I worry that any analysis I add is just going to end up parroting someone else's analysis. That said, I'm going to attempt to make a ranked list anyway, if for no other reason than to get my thoughts out of my head. From most to least sus:

Most sus-
1. The Outsider. Multiple people have pointed out that self-voting is a waste for either side, and I can see their point. At this point, however, I've done enough sus actions and been quiet enough to warrant a test. I've unwittingly made myself the best option.
2. JeenLeen. The thing I find most suspicious in any game is inconsistency, and Jeen has repeatedly ignored their own reasoning of why not to vote people in determining their votes. Keeping my vote on him.

Something I realized from The Outsider's posts: I think I wasn't too concerned about seeming suspicious since there's not much to go on D1 usually and getting tested didn't have any serious negative ramification. Even though I know it's 1 non-Thing getting tested, it's likely no Things get tested D1 anyhow. However, my lack of care about how Towny I seem means I'm getting heat and attention that would better go towards the actual Things. I was inadvertently helping them.

As a result of this conclusion, I feel that some of the more silent players, or those who started a ball rolling then sat back to watch things continue, might be Things. Hopefully we'll get some good analysis out of looking back on who sat what about who after the tests. (Well, truly hopeful we hit a Thing, but doubtful.)

Still feel free to test me today. I don't expect the above to really change anyone's mind, as the above is probably the best argument a Thing could make to try to persuade tests going off them.

[Apogee1]

I'm pretty happy with Outsider and Jeen as my tests neither of the last posts has particularly inspired me to feel better relative to others.

I do think Valmark is being over-read as not a thing, but that might just be because I felt he came at the argument I was making rather perpendicularly to what I meant? Idk I haven't felt like, anything I really find town from him compared to a couple others.

[Grek]

Although it's not relevant for the first day, I am curious: Does the Town get +1 test per successful vote, or only if the second vote is successful? As in, if we grab two things in a single day with our original two votes, can we test two extra people or just one on the back of that? Obviously extra votes can repeat if we keep hitting things.

[Valmark]

Although it's not relevant for the first day, I am curious: Does the Town get +1 test per successful vote, or only if the second vote is successful? As in, if we grab two things in a single day with our original two votes, can we test two extra people or just one on the back of that? Obviously extra votes can repeat if we keep hitting things.

Should be that regardless of how many turn out 'positive' we only get +1 vote, since it says 'if either of them' (also because it looks unfair for Things to me if we could guess two Things and have a whole new Day).

[rogue_alchemist]

Although it's not relevant for the first day, I am curious: Does the Town get +1 test per successful vote, or only if the second vote is successful? As in, if we grab two things in a single day with our original two votes, can we test two extra people or just one on the back of that? Obviously extra votes can repeat if we keep hitting things.

Valmark is correct. You get one "free" test to start with, then you get an actual test. No matter how many of your 2 tests hit (so long as 1 does), you get 1 more test to vote on. So you cannot earn more than +1 test under any circumstances. It would be far too unfair to the things otherwise.

[JeenLeen]

From Google, it looks like it's about 9:50 PM GMT time now (about 4:50 PM Eastern time zone in USA).

Definitely understand if ending the day happens late, but wanted to check if I'm reading the time zones right. Day should have ended almost 2 hours ago, right?

[rogue_alchemist]

Yes, you are right. D1 has ended. Hold on while I double check my tally.

Edit:
Vote count:
Caerulea (6): Elenna, Grek, JeenLeen, Valmark, AvatarVecna, Book Wombat
CaoimhinTheCape (2): Caerulea, Book Wombat
The Outsider (5): Caerulea, Apogee1, The Outsider, CaoimhinTheCape, JeenLeen
JeenLeen (7): Apogee1, The Outsider, Elenna, CaoimhinTheCape, AvatarVecna, Grek, Valmark

You test JeenLeen first, but their blood work comes back normal. They have been telling the truth, they area scientist.

You then test Cerulea. They freak out and suddenly only an amorphous puddle is left. Turns out Cerulea was a Thing. You have 24 more hours to vote for a new test.

[AvatarVecna]

The Outsider
CaoimhinTheCape
PartyOfRouges
Apogee1

These are the people who didn't vote Caerulea despite her being at the top of most people's scumlist and despite her weird behavior.

Elenna
Grek
Valmark
AvatarVecna
Book Wombat

These are the people who voted Caerulea, but aren't 100% clear at this point.

I'm gonna read back through the thread about the four at the top first, see if there's something that feels fishier now that we know Caerulea is scummy.

[Elenna]

Well, not only was The Outsider third on my preference list, but Caerulea's alphabetical list moved them from third on the original player list to tenth on Caerulea's alphabetical list. I'll try to take a closer look tonight, but I'm almost certainly going to be voting them for this round.

EDIT: This also makes me feel better about Apogee and Book Wombat, since it seems unlikely that Caerulea would propose an alternate list which would move her fellow Thing from the bottom of the list to the top.

[The Outsider]

Well, I miscalculated. Badly. I don't have time for a full-on analysis or defense, but I'll do my best to try and come back to this as soon as I can.

[Apogee1]

Okay so Jeen is fully confirmed right now.

Let's look at votes on/off Caerula.

- - - Updated - - -

I don't think this is a good idea. For one thing, if any initial Things are within the second half of the list, we're just giving the game to them, because they can just infect the person we just tested every night, and by the time we reach them over half the players will be infected and they'll control the game. So this gives the Things a 50-50 chance of auto-winning, or a 75% chance of winning if there are two Things to start with. That doesn't sound like a very good plan.

Also, if we agree to test alphabetically every day, that will most likely kill a lot of the discussion, which will make it really hard to find suspicious people, so even if we don't lose by the end of the first set of tests, we'll have much less information to go on after that.

As such, Caerulea. This isn't a very strong suspicion, since of course townies can come up with bad plans, but I prefer this to voting randomly.

Caerulea, any specific reason for why you didn't just use rogue's list that would have led to you getting tested Day 1 according to the double voting theory but rather made another one?

I think Elenna and Valmark both are aggressive enough towards Caerulea early on and then stick out that wagon that they are like, never things (for today but we'll cross that bridge when we get to it). The other people on that wagon I don't think are things but am not fully confident clearing.

- - - Updated - - -

We do have two tests! Excellent.

Spoiler: Vote Counts

Show

AvatarVecna: Valmark, Caerulea
Apogee1: Caerulea
Caerulea: Elenna, Grek
Valmark: JeenLeen
CaoimhinTheCape: Caerulea

I am a little confused by this, but alright.

In this case, my preferance list is as follows:
1. CaoimhinTheCape (Haven't spoken yet. I feel like a thing would most likely not speak as much early, to avoid being the topic of discussion. They get a very large advantage by flying under the radar, and I don't think we generally vote people who are quiet early.)
1. PartyOfRouges (see above)
2. BookWombat (see above, less so because new player)
3. The Outsider (Made a very small contribution who's only effect was to wagon someone who isn't them. Quite suspicious.)
3. AvatarVecna (As above, but it's AV and reading her is hard)
4. Apogee1 (Pointing out that what I suggested is bad (and it is) under the one test per day system, but no investigation really. Easy way for a thing to appear without risk.)
4.5. Elenna (Pretty much same as Apogee1, but slightly less)
5. Grek (Suggestion of ranked preference lists seems good, even if I don't completely understand her proposal)
6. JeenLeen (I like their thoughts, and they are doing some investigative stuff. I don't agree about AV, she might get wagoned but I don't feel like she gets voted off day one that frequently. Also rogue_alchemist might want to give the starting thing to a skilled player)
10. Valmark (Excellent point regarding my list, and is being investagative. I don't think a thing would be that bold.)
11. Caerulea (I am not a thing, and so voting me wastes a test)

Voting AV because I would prefer testing her to me, and Cao in case anyone else wants to try and vote someone who's quiet.

EDIT: Switched vote to The Outsider because they seem quite thingy.

I think this is an interesting post in retrospect, considering how she votes her 1st and 4th thing preference despite the fact, you know, she rated two people as more likely things than Outsider?

I haven't decided if I think that's more of a distance move or a "PoR/Bunny" has the other thing in it so I want to distance or just weird.

- - - Updated - - -

Null sus-
5. Caerulea. People have made good arguments about their evil alignment, but I'm in agreement with JeenLeen's initial reasoning on this one: I don't think a Thing would have risked drawing attention to themselves like that, even if there were two of them. I can understand why people find their arguments fishy, and creating a new plan where they don't get tested does look a bit shady, but I can't help but feel like our attention is being diverted somehow.

This, however, reads like "let's save my buddy"

a) Boosts confirmed town's JeenLeen's reason for Caerulea not being a thing
b) vague assertions that "our attention is being diverted somehow"
c) the null just feels out of place relative to what the other people in the null tier and the thing tier have for reasoning

TheOutsider

[AvatarVecna]

Spoiler: Apogee1 ISO

Show

A set order is definitely not the way to go. Because, in a worse case scenario:

10 town 1 thing d1 (1 test)
9 town 2 thing d2 (2 tests)
8 town 3 thing d3 (3 tests)
7 town 4 thing d4 (4 tests)
6 town 5 thing d5 (5 tests)

town loses at night

So we have, in the worst case with only 1 thing, 5 tests to run (more if we ever hit a thing). And to win, we have to double up on successful thing tests a number of times equal to the number of things at the time of the test that hits, so if they ever get to three things we probably just lose unless they make each other really obvious.
I think having a more "optimal" mechanical strategy than pure chance is impossible, because since no private communication the things can always plan against what the plan would be. Or you know, you and/or someone else proposing a plan is a thing

Of course we hopefully will be better than chance at deducing who things are.

Immediately calls out Caerulea for her bad plan. Could be easy distancing, could be townie reacting honestly.

Oh we have two checks that does change things

If I weren't me I'd be hesitant to give me credit for explaining the (at the time what I thought was present flaw) in Caerula's plan considering, you know, it was me who was going to be tested first under it.

I think Elenna's post was very natural and flowed between ideas well enough so I'd say she's a scientist for now

- - - Updated - - -

With two tests a day the blast through the order plan is more compelling

Though maybe we do something like don't define a set order but have a pool of "tested" and "untested" where we probably but not necessarily try to test all the untested before we test some tested?

- - - Updated - - -

Wait I think I misremembered something re: Elenna. Still probably a scientist but not as much as I thought rip.


I would, however, point towards TheOutsider and JeenLeen

Still not sure what to make of this one.

Exactly.

Why did no one consider that until I brought it up?

- - - Updated - - -

As in, I was getting townread for going after a strategy that would have resulted in killing me if I was a thing.

...right, Apogee1 is clear. Caerulea's plan resulted in testing Apogee1 first. Unless this is some fantastic distancing attempting to set the two of them up as immediately antagonistic to each other, Apogee is town.

Okay I have no idea how this happened but I somehow got it in my head Elenna like, switched votes between three people in succession

Which apparently no one else did in this game either so I just fabricated an entire post in my head I thought existed? Idk

I still like that post

JeenLeen was because the townreads were maybe forced or TMI'd

Outsider for reasons you've described

yeah

Not sure what to make of this. Apogee is hard to read sometimes. Literally too.

Okay so Jeen is fully confirmed right now.

Let's look at votes on/off Caerula.

- - - Updated - - -





I think Elenna and Valmark both are aggressive enough towards Caerulea early on and then stick out that wagon that they are like, never things (for today but we'll cross that bridge when we get to it). The other people on that wagon I don't think are things but am not fully confident clearing.

- - - Updated - - -



I think this is an interesting post in retrospect, considering how she votes her 1st and 4th thing preference despite the fact, you know, she rated two people as more likely things than Outsider?

I haven't decided if I think that's more of a distance move or a "PoR/Bunny" has the other thing in it so I want to distance or just weird.

- - - Updated - - -



This, however, reads like "let's save my buddy"

a) Boosts confirmed town's JeenLeen's reason for Caerulea not being a thing
b) vague assertions that "our attention is being diverted somehow"
c) the null just feels out of place relative to what the other people in the null tier and the thing tier have for reasoning

TheOutsider

Some nice analysis.

Apogee1 is clear right now, AFAIC.