This is a big part of why I'm generally highly skeptical of "tells" - even if they happen to be more than coincidence (which I think is true less than people think) if the person giving off the "tell" ever figures it out it is almost always fakeable with zero ramifications to them.

I'm just going to go ahead and post some of these in chunks as I finish them - they may or may not be useful, but I'll feel better posting them whole instead of trying to choose pieces to post

Midgame Villaging

For the purposes of this post I'm defining midgame villaging as beginning on day 2 when the village first has information from lynch + nightkill and ending the day after the last seer dies (sometimes there is no midgame, but we'll ignore that for now). This is the time of the game where as villagers you are primarily concerned with three things - clearing villagers, finding wolves, and keeping the seer alive as long as possible. Different people disagree about the relative importance of these things, but in my opinion too much thought and emphasis generally goes into keeping the seer alive and not enough goes into clearing villagers and finding wolves. So in my relative order of importance

1. Clearing Villagers - This is important for many reasons - if done properly it gives you a base of people that you can trust to work with and discuss/debate the relative wolfyness/villageryness of others. It also allows you to focus more on a smaller pool of potential wolves, and it makes you more likely to be nightkilled in a seer hunt. The most important thing to remember here is that if you're really clearing someone you need to be SURE that you are correct - in my opinion hard clearing a wolf is the worst midgame mistake you can make.

2. Finding wolves - Still the basis of the game and the way to win, but marginally less important than properly clearing villagers because while villagers can be cleared with close to 100% accuracy if done properly you are going to have more error here mostly because there are going to be times when villagers inadvertantly look a lot like a wolf due to some combination of their own lack of knowledge and wolf team manipulation of the evidence. Generally if you can stay at or above 50% lynching non-seer peeked wolves during this time you're keeping your team on track to win the game.

3. Keeping the seer alive - My theory on this has always been that if you are playing a strong solid villager game you do not want/need to make the wolves think that you ARE the seer but instead you need to keep them from ruling you out as the seer. This means primarily that you want to avoid hard clearing any wolves at all costs. As long as your top villagers are actually villagers it becomes very difficult for wolves to rule you out as the seer even if your wolf list is entirely wrong. The problem in my opinion with hero fakes and leaving very obvious fake peeks is that it makes it very easy for the wolves to rule you out as the actual seer if even one of your fake peeks is wrong (and you don't even get that much information out of being left alive because you don't KNOW that one of your peeks is wrong, plus even if you did you wouldn't know which one).

I think your (3) makes a very good point as far as thinking about seer cover. When I think about games where i've been successful seer hunting versus games where I've done poorly, excluding games where I had some metagame soul read on the seer, the biggest difference has been villagers excluding themselves from being the seer early on

herbie how does your "midgame villagery" difference from early? Midgame is probably where seer has to consider claiming anyway, so getting him to midgame is the hardpart imo

it's not a waste of time at all, i would probably be a way worse villager without it.

anyway, i'll write an article strictly on this matter. there's a lot of noise and it's easy to level yourself into thinking you have identified a tell when it really might just be a coincidence. i think i'm really good at identifying what truly separates somebody's villa vs wolf game with a re-read.

Article #5: How to Approach Must Lynch
- written by Toganim

This article will debate two main subjects: When to vote and who to vote.

Unless specifically mentioned otherwise, both subjects will be discussed in a situation where:

• It is ML.
• There are at least 2 wolves left.
• No one has voted yet.

For future reference, I’ll use these terms:

Hero = us (villager)
Villain = everyone else (can be v/w)

When to vote

POG consensus seems to be that “if a villager votes for another villager in ML, game is over”. This is not entirely true. Manipulating the timing of votes can prove to be a game-saver.

First of all, we want to initiate the voting process by making wolfy people vote other wolfy people.
Though it might seem obvious that we don’t want clear people to vote other clear people, people rarely actually force wolfy people to vote other wolfy people.

Secondly, we want wolfy players to vote sooner rather than later. Weighted village consensus should decide the timing of votes (clear people weigh more than wolfy people). Forcing wolfy people to vote before villagery people has a lot of benefits, but mainly:

• The village increases the chance of putting wolves in distress. Villain will be forced to take a stand and justify the vote. Villain will no longer be able to express an opinion and then change it to snipe later.
• It minimizes the risk of wolves piling on a villager if a villager votes wrong.
• The villagery people will have more information to base their vote on.

Example:
ML in F7. Village consensus seems to be (rank of villageriness): A > B > C > D > E > F > G.
Truth is that D/E/F are the wolves.

Scenario 1)
Villager A and B are the most clear, so people trust them the most and let them decide. They each have trouble deciding if they want to vote F or G. If one of them votes G it’s game over, since D/E/F will pile on G, making it majority.

Scenario 2)
Village forces G to vote first. He’ll vote F, making F vote him. Village inverts the order of votes, making E vote first. Wolves E and D are voting before villagers C, B and A. E and D will need their vote to reflect whatever reads they had previously (or at least have a great reason to not vote like that) or they will most likely out themselves.
Let’s say that D+E are voting for F/G. This makes it 2-2 with 3 villagers to go. A ML situation with 2-2 votes and 3 villagers remaining is better than 1-1 with 3 villagers and 2 wolves remaining.

If D+E decide to go for the win and they both vote for the villager, A+B+C will get more info and might be able to solve the game better. Important information would be something like:

• FG are not v/v (or game would be over) [This might not seem too helpful, but it’s more important than knowing that AF are not v/v since A is the clearest person in village consensus anyway].
• GED includes at least 1 wolf (or game would be over).
• If G is villager, then ED are wolves (or game would be over).

This info was not obtained if A or B decided to vote for whoever they found the wolfiest. ABC now needs to vote on either F or G. And village can afford one of them to be wrong and vote for G if the wolves decided to spread out their votes.

Why would G vote first when he’s not a wolf? G wants to vote first because:

• He realises that he’s in high danger of getting lynched.
• He needs to pick a battle; if a super clear villager votes for him, he will be lynched. He needs to make sure that he can get someone else lynched. Bringing us to the next subject…

Who to vote

It’s your time to vote. No one made it easy for you by voting you first, so you have to vote first. This is what you do:

Voting for whomever you think is most likely to be a wolf will look good in a vacuum, but it’s not always the best vote to optimise the chances of winning for the village. You have to weigh in the factor of village consensus.

Voting someone in ML will always result in them auto-voting you, making ML about you or them. One of you has to go.

If village consensus is that hero is more likely a wolf than villain, then hero gets lynched (and vice versa).

From hero’s perspective, voting the second most likely wolf and getting him lynched is always better than voting for the most likely wolf and ending up getting lynched yourself.

Whoever you vote will auto-vote you back, which will result in 3 possible outcomes:
1) You voted a villager; wolves win
2) You voted a wolf, wolf gets lynched; resulting in another day
3) You voted a wolf, you get lynched; wolves win

Voting for your second most-likely wolf will increase the possibility of outcome 1, but it will also potentially remove outcome 3. Overall EV massively improved, since everyone is a better lynch than you.

Don’t despair if you can’t get your top wolf suspect lynched today. Village consensus might change overnight.

Cliffs:

• Force wolfy players to initiate the voting by voting for other wolfy players.
• After these first votes, let the remaining wolfy players go next.
• If you are first to vote, don’t necessarily vote for whoever you think is most likely to be a wolf, but instead consider who might actually be lynched over you.

pretty sure the optimal time to vote is at :00

I think Toga's article is good

Problem is people are rebellious at following correct ML procedure and ranking of villageriness to wolfiness can sometimes vary widely.

Waste of time was probably pretty hard. I'd still prefer a villager who spent his time doing other stuff though.
Inexperienced wolves will be caught yes, but;
- inexperienced villagers will be caught as well because they havent found "their style" yet
- experienced wolves will be able to fake it / argue against it
- Seers will be busted as well

Well I define midgame as starting on day 2, because I generally think of day 1 as being a sort of mini-game within the game. My general feeling is that the only thing you really need to think about when it comes to day 1 seer cover is avoid doing things that would cause the wolves to completely rule you out as seer (things like calling someone a lock villager then wavering on it later in the day). You can generally avoid this by simply avoiding making any definitive proclamations about people's roles on day 1 unless you're really really really sure.

These are both cool imo.

FWIW I was planning on being social this weekend so I can't promise I'll auto-in to an all saturday long affair.

Also: I wouldn't mind slightly longer nights, maybe a post restriction (i.e. 25-30 posts per day just in case people decide to spam as if it's a normal turbo and the thread gets out of hand). like 3 hour days 1 hour nights sound good to me so as i won't feel rushed seer hunting. or maybe 3:15 vs 45

For the other one, 40/20 sounds better to me I think. Or maybe something like 45/15 and then we have an intermission which is like an hour long n3 or whatever.

Another benefit by the method that Toga is advocating is that it removes some power from clear wolves. I have two times been in a must lynch situation where I sensed that one of my fellow wolves were about to get voted and so I stepped in, as the clearest wolf, to quickly vote an unclear villager, making it a battle between me and that person which I would easily win.

It is actually 100% true that a villager voting for another village with normal ML rules (no unvoting, no vote changing) ends the game, provided only that

a) No wolf has already voted some other wagon
b) There is no possibility of wolves being prevented from voting based on time constraints

fair enough- i thought midgame was later

seers should be able to say and do what they want as long as their peeks are clear, especially if this can rule them out as seer

you're just going to have to wait for my article to come out because none of these have proven to be issues for me.

and yes, if i correctly read shortline as wolf he will be able to argue against it fairly well. but even if he dodges lynch anyway, i'm definitely not going to get him lynched if i (mis)read him as villager.