Identifying what works for you
-Scale usefullness of hunting techniques (no bullshit quantify?)
Forcing different playstyle limitations
Fewest posts possible
Fewest words possible
Eliminating the scum hunting method you depend on
Using only a scum hunting method you want to improve on
Power wolf, Lead town, Lurk, Roleplay ect
Understanding your perspective bias
Too scummy to be scum fallacy in line with projecting your level of skill onto others / overestimate others skill as fos
Counter feeling of 'caught but caught for bullshit reason' -If they caught you you triggerd something in their mind
Analize what has worked and has not worked for you (Stop ate if it gets you lynched) (Stop using meta if it fucks you ect.)
This is something I was working on to provide a path for player improvement and a bit of a work in progress but this is how I see it:
Basic understanding of voting, game mechanics, and roles
Basic understanding of WIFOM, seeding, tells, traps, and simple mechanical hunting
Can use and understands the commonly used mafia slang/abbreviations
Grasps basic strategy's like RvS, and follow the cop
Some ability to scum hunt past depending on Meta, Gut reads, and simple mechanical hunting
Ability to avoid being scum read by default as town
Understands the majority of mafia slang/abbreviations
Comfortable enough to play as scum and work within a wolf team without going for a lurk win or screwing your team mates
Some ability to lead a town and guide their focus
Grasps general strategies
Understands how to set up traps and gambits as well as how to pull information from their use without getting themselves lynched
Can comfortably play many game types (Vanilla, PR heavy, Multiball, Mashups, Turbo, Real life, Skype, ect..)
Can evaluate setups to understand what tells to focus on and what strategies to use
Can sometimes spot other players seeding and picks up on small details
Ability to form reads through multiple methods and communicate those reads to other players effectively and persuasively
Ability to manipulate how they are read by others as needed
Effective at leading a town and guiding their focus
Some ability to Power wolf, push misslynches, track town bias, and lead a scum team
Ability to gambit effectively and set traps, has some understanding of more subtle maneuvers
Can comfortably play on multiple sites
Mastery of using Mechanical hunting, Wagonomics, and Theory crafting in tandem
Strong understanding most scum hunting methods and when one should be favored over another
Understands the majority of strategies and when they should be used
Ability to peg scum players on D1 with some confidence
Comfortable power wolfing, always tracks town bias and can push misslynches effectively
Understanding of game tempo / timing to reveal information, use gambits, and push for days end lynch
Ability to shade a player for lynch days in advance
Strong understanding of how to track town Bias to manipulate a mislynch
Ability to communicate subtly in day chat
Can use Game Theory to identify optimal play or leverage sub-optimal play to submarine the town
Has some understanding of the overall Mafia/werewolf community to include the sites and players of note
Can push a misslynch so effectively they are not even scum read for doing so post flip, sometimes gets town read by players for scum reading them because their reasoning is so good
Innovates and creates effective strategies
Ability to manipulate how others are read by others through various methods
Ability to manipulate overall game moral as necessary
Ability to recognize how days will play out before the day starts through experience
Mastery of evading lynches
Ability to recognize and read players in context to how they understand the game
Comfortable pulling FPS gambits
Regularly picks up on subtle communication between other players and seeding
Wins fucking championships
I don't think skill can be strictly quantified but I do feel that a list like this could be made and players could be roughly categorized in a general way. My goal in building this is to allow players to go through the list themselves and help them cut through the 'cherry picking' nature we all have to point to our good plays and say we are good.
Technique to improve- Break your posts into deductive logic outlining what another poster said as well as your argument
-Play to weaknesses
-Read before posting
-Slow down and post with intent
-Break down arguments identifying their appeals, claim and warrant
Add tips section?
Town Power Role
Evaluating Gamestate for hunting
As a Town Power Role it can be very useful to 'seed' or hint at a night action you took. Doing so gives you a strong defense if you end up counter claimed by a scum in a 1v1 as well as allows the town to recognize what action you took after you died. This is particularly critical for a Sheriff or Alignment Cop. Its common practice for a sheriff to reveal their peek as their first read in a read wall or sneak in a hint that will be understood in hindsight (like "I town read this guy so hard I would town read his slot even if it got subbed out") It is a fine line to walk of seeding in such a way that you are not identified by the scum but its still obvious enough for people to notice after your dead. As a sheriff you can equally seed a scum read in a read wall easy enough.
N0 Sheriff Peek Shelter
In some setups a Sheriff or Alignment Cop will start D1 knowing 1 player is Town. If this is the case all town players should (preferably in their first post) claim an N0 town peek on a random player. This way when the sheriff dies everyone can look back at his first post and know who is a confirmed town without requiring the Sheriff to take his peek to his grave or risk revealing himself to voice who he checked. This is considered basic play and a general expected norm in pretty much every site I have played Mafia/Werewolf.
Random Vote Stage (RVS)
To play day chat there has to be reasoning to analyze. The objective of RVS is to create that reasoning from nothing (given a game starts with 0 posts to talk about). This is done by casting a vote on another player with as strong of reasoning as possible. As the pattern continues and votes are cast the reasoning gets stronger and less 'random' allowing to 'break RVS' which is where the game really starts. It is critical as town to push to break RVS as rapidly as possible because the day is limited and its where you get to hunt. Scum will want RVS to last as long as possible as their goal is to not be caught and get back to the night phase so they can kill. Just remember to do this you should always give a reason. A random vote does absolutely nothing to break RVS.
Spoiler : Example :Sally votes John giving the reasoning that 'he smells funny'
Camila votes John agreeing with Sally
Santiago votes Camila pointing out he vote is opportunistic and she is just jumping on a train (At this point a small reason for a vote has been created out of nothing)
It can also be of value to Gambit yourself as Lynch bait to break RVS (See Gambits)
Alternative to RVS
Instead of building information off vote reasoning you can generate reasoning by analyzing the setup and strategizing about how town should react to potential situations. This is particularly viable in setups that are Power role or Game mechanic heavy. One advantage of this is that it can help town identify situations that would help or hurt them while allowing players to get familiar with the setup. I have often seen debates over how a situation should be delt with break RVS faster than the random votes would have and the setup speculation can be a nice excuse to fake a 'scum slip' to gambit yourself as lynch bait in a way thats not damning.
As town I believe its generally better to reveal pertinent information you notice unless you have a reason not to. This is because it only takes 1 member of the Mafia team to notice something in order for them all to be aware while each individual town member has to notice that information to be aware. By doing so you reduce the information advantage a well oiled scum team shares.
Reasons to hide information could include potential Town Power Role indicators, not alerting a potential scum you are onto them so they can keep slipping, or noticing a potential gambit in play that would be ruined if you revealed the it.
Never express you will not lynch while you vote
I can not stress this enough. The power of your vote is a threat to lynch, the moment you vote but express you are not willing to lynch your vote means nothing. Your vote is one of the few tools at your disposal to add pressure and weight to your voice and you should never give that up as Town. I generally always voice my votes intention Change to express reasoning with vote/add more scenarios for voting
Mutt and Jeff Technique
(More commonly known as good cop bad cop interrogation technique.) It builds trust with a suspect leveraging their primal desire for protection from a threat making them more cooperative and fostering a desire for the suspect to please the 'Good cop.' I have found this to be extremely useful and can be used when a target player is under heavy pressure from others by you acting as the 'good cop'. A particular way I commonly use it is on a player that is at substantial risk of being lynched. By pointing out they will likely be lynched and after the fact with their alignment confirmed via flip, anything they say now holds great value you leverage a towns motivation to secure their win even after death while a scum player would actively want to avoid giving information and instead just focus on survival. In this way you get a win/win if they participate, a town indicator if they work hard to dump their minds for the town after they flip, and a scum indicator if they instead choose to go quiet or hardline focus on their survival.
Triangulating (Communication to use different part of brain) Find reference again
Gambits Subsection? Add Notes/Tips?
This is essentially a trap play by allowing yourself to appear as an easy lynch target and see who jumps on you or to break RVS. The belief is that because scum have to make up reads they often go for low hanging fruit that will make an easy pressure target and that by making yourself appear as an easy lynch / obvious scum for a moment you can draw them out. This is a decent strategy but it has potential to backfire if the player attempting it is unable to dig himself out of the hole he created. Some plays will be viewed as ‘scum slips’ with the claim that it was a lynchbait play being viewed as scum trying to distance themselves from this slip. I have seen scum capitalize on a players lynchbait play to push through a mislynch and walk out blameless on the other side. When using this play its good to seed or leave some kind of crumb you can reference back to in order to show that it was a planned ‘slip’ to prevent this from backfiring and its generally a better play for players who are comfortable taking pressure and using their verbal kung-fu.
This is among the most common techniques. A player claims to be a Cop with a damning night peek on another player to gauge their reaction as well as the reactions of other players in the game. This can be an effective although somewhat disruptive play that usually derails any other town train in progress. Players doing this should consider seeding something to reveal it was a gambit after the fact and acknowledge the potential of the real alignment cop outing themselves by counter claim or their reaction to your gambit. I also highly recommend players consider what information they are trying to generate and what they can say to generate that information prior to making a play of this sort. I have honestly seen this sort of play hurt the town as often as I have seen it help them and I personally do not favor this sort of gambit.
When trying to figure out the game take the time to question yourself when people post opinions that disagree from a position you support. There is something in Engineering control theory has something called a feedback loop. Basically if you remove negative feedback the loop always goes divergent and chaotic. A balance is necessary because it keeps you (or the town as a whole) from drifting to an extreme. Many great players struggle to move past their hubris resulting in occasional blunders they look back on and say “Man I should have known better”
Weighing the value of a lynch
This is strictly a town play but can be used by scum to manipulate a lynch. The basic idea is that you can break down potential lynches into a raw evaluation of cost / benefit in order to identify the best lynch. At face value this sounds as simple as ‘Lynch the most likely scum’ but it is actually much more complicated. What if there are multiple scum teams? What if there are multiple trains on players you read as scum? How do you objectively weigh which player is most likely scum? These are some very basic concepts to keep in mind that may assist you in making these tough calls.
-Identifying a good D1 lynch target
For your D1 lynch you should evaluate 3 areas on the target-
‘How certain are you that this slot is scum,’
This is obvious and unfortunately the only thing most people think about when picking a lynch target.
‘What information will you gain from the players flip,’
Think about association reads, how players pushed on the player, what
‘How valuable will this player be to helping you solve the game if he is town’
By breaking this down you will be able to peg a good D1 lynch target. You need to force players to give reasoning on the train in order to create material you can work with the following day; this way if he flips town you can look at players reasoning behind their votes. A ‘policy lynch’ on a lurker or non-participating player is always a terrible idea D1 as that if it flips town you just killed a townie and gained absolutely nothing from that train.
-Sorting a damning PR result on a player you Town Read
In general use Occam's to sort a scum peek on your town read. If your evaluating night actions and in the players defense he points out some scenario where he could be town but it requires 2 other PR’s doing some exact action in a 20 player game just evaluate that probability and compare it to your certainty that the player is town. If you are not die hard certain that he is town you should grudgingly lynch the slot. (This is assuming it is early game and there is no reason for scum to gambit a peek, In a Lynch or Loose situation treat it as a 1v1 and cut back to analysis. Never go IIoA in these situations)
-Choosing between 2 scum
Something to consider while weighing a lynch in this situation is the value of each target to the town. For example eliminating the Serial Killer early in a game will usually drop the KPN and provide a massive advantage to the town. If for whatever reason you think one player is mafia (With multiple mafia alive) and the other is an SK the SK obviously has more value to eliminate.
Clear yourself as town
Pressure others for reasoning behind their reads (to build reads)
Have accurate reads
Explain the reasoning behind your reads / Persuade others to follow your reads
Only a few town players in any given game will have accurate scum reads. The only way the town can win is if those players speak up and persuade other town players to believe their read enough to lynch the scum. Identifying scum and eliminating scum are 2 separate skills and one without other other is totally useless. If you play every game with perfect reads but nobody ever listens to you then you did not play a good game. You completely failed in a very important way.
often people do have correct reads but end up getting lynched, and then complain about how no one listened to them and feel like they played a great game. but they are wrong. because they aren't looking at the big picture. if you fail to convey your ideas in a way that sways villagers minds, then your ideas are useless. and if you get mislynched, that is a failure on your part.
Night Feedback Analysis
This is using raw feedback information to break down the game into logical reasoning that gives pieces to solve the puzzle.
So in this hypothetical game we can make a few solid assumptions using just these 2 charts. From the starting point we know that there is a role blocking PR in the game but we do not know if it is a Mafia or Town PR.
1- The night kill on D3 was targeting Bob but redirected to Sally. This means that Bob cannot be the mafia.
2- On Day 2 there was no factional kill and Joe was role blocked. This suggests 3 things.
A- Joe could have been attempting to perform the factional kill and was role blocked preventing it
B- A lynch on Joe ‘could’ verify the alignment of the escort/consort. If Joe flips mafia the escort is potentially town.
C- The doctor healed the target that was attacked
3- On Day 1 there was only 1 bussed feedback claim. This suggests one of 3 situations
A- The Bus Driver swapped himself with John
B- The bus driver swapped John with Travis (Who was factional killed, This would suggest John was targeted for the kill N1 and could not be mafia.
C- The other person that was bussed withheld the feedback
By breaking down these night actions and their implications conclusions can be drawn that provide greater insight into the game state.
Correlation does not equal causation but correlation can still be used to predict. If more people have colds and ice cream sales are low they may not cause etch other but the lurking variable of season temperature may alter both factors allowing them to still predict etch other to some degree through correlation
Follow the Cop
This is pretty much cutting back into IIoA. Its when the players in a game shift into reliance on investigative results instead of analyzing the day chat to hunt for scum. This is a highly sub-optimal strategy that is often toxic to games and usually when adopted by a town that town ends up failing if the cop dies because the town has not generated information to work off of for an analysis based lynch. This is also something hosts strive to prevent from happening.
Track town bias-
If you do nothing else the most critical thing you can do as scum is to track towns bias. Team Scum’s biggest weapon in day chat is that they are informed and at the end of the day that allows a coordinated scum team to manipulate the direction of a lynch. By understand what players a town has expressed willingness to lynch you can manipulate the direction of the overall town as well as the individuals. You will know you can poke them on a town they have expressed a scum read on and it will draw push towards a TvT. Understanding a large portion of town players have expressed willingness to lynch or scum reading a player will allow you to set that player up as the days lynch and through manipulation of individual bias you can line up potential lynches in advance. Its key to keep in mind that the vote count does not matter- As a wolf the only thing that matters is where the town votes.
The most basic (and arguably the easiest and most effective) gambit you can pull as a wolf is to claim to be a power role on the stand. In my opinion this is not done nearly enough and is especially effective if you have planned ahead and seeded your claim earlier in the game. It puts the real TPR’s in the position of choosing between:
1- Express disbelief in your claim pointing that they ‘may’ be the TPR
2- Counter claiming which outs themselves (which is good for your team if you get lynched or if you can shift the lynch on them it prolongs your life for a day and you can LOL cat)
3- Staying silent and giving you a pass for the day
Every one of these options is great for you as scum and it puts the town in a much worse position. The ‘meta culture’ of a site should be considered as it greatly influences how players respond to such plays.
-Cute Fuzzy Kitten Defense
Basically it’s a bid for sympathy with a scum claiming some insane role and (hopefully) with town assuming that there is no way any scum in their right mind would claim something so absurd. This is pretty entertaining to do but depends on town to read and defend you as 'too scummy to be scum' which can be hit or miss while also depending on specific setups to be viable.
Some people will hunt for town slips and you can manipulate this by presenting a plan or talking about something that you would obviously know if you were scum. Things like how scum chat functions or the roles scum could receive work although I have had more success arguing some idea or strategy that hinges on a misunderstanding that will help clear you.
-Deep Wolf and Hard Bussing
This is a particularly high risk high reward scum play that helps counter many scum hunting techniques. The idea is to work out in your night chat to lynch one of your own team mates and have another one champion the lynch. Without a TPR stopping you if you hardcore championed the lynch of another mafia you will have an extreme unaligned tell between yourself and your team mate. I have rarely seen town reason around such a play when its done properly but doing it creates a WvW and can place you in a 1v1 which hurts your chances of establishing a TvT (which is generally your objective as team scum.) It absolutely helps to establish signals before hand in your team so you can call for more/less pressure and is also a good idea to do it early in a day cycle allowing time for it to resolve before a lynch is decided if you want to hard bus instead of going for the full deep wolf play.
Split positions as scum while bussing
In general I recommend splitting your stances on reads while you bus. The objective is to disassociate in case of a flip and in general players jump to a TvS assumption post flip so by splitting your positions you hurt any aligned hunting mechanical or inductive/subjective methods while also broadening the pushes your team can make organically against the town.
Call out your own slips and weaknesses
This is a pretty basic strategy that both allows you to control ‘the way’ your slip is revealed as well as pushes the WIFOM of ‘would a wolf call themselves out like that?’ I particularly use this in a response to a FoS on a totally unrelated issue as ‘If you had scum read me for this I would understand but what your saying is stupid’ which simultaneously invalidates a push on me while also insulating myself from a mistake I made.
In general I place town in 1 of 3 categories: Pocket (which expands my ability to manipulate town), Lynch (Keeping the lynch off your team and saving your NK), NK (Eleminating potential TPR’s and strong players there is not enough town bias against) I always think in these terms and usually run a section in my notes tracking this
I feel this is not done nearly enough but is a very good play in a ‘Mislynch and loose’ situation where town lacks PR’s that can take advantage of additional night phases. By killing a townie you just stack the odds further in favor of the town to make a successful lynch. If your team has the numbers to fight it out or the player ability to do so I would highly recommend this as a play.
This functions well to break any ability for town to identify some motive or reasoning to your kills but your sacrificing strategic advantage to eliminate the towns ability to use strategy. I especially favor this as a method when the town has multiple players who are overthinking kills and drawing bad conclusions (although you should recognize that the same ratio of RNG could land a bad conclusion on one of your team)
Target empty slot
I rarely see this done but I feel it’s a strong play. It kills towns ability to apply any reason to the kill while also denying them a slot that would step in with a fresh perspective. I especially favor this as a kill target when as a scum team things are going the way we want.
Notes as scum-
Pocket, Lynch, Kill
Agenda Power Wolf plays
Divide and Conquer
Seeding distrust/ setting up lynches in advance
Night action selection - RNG, Killing empty slot
-Catering bussing to how most influential town members play (or one you wont kill), measure town pockets
Optimal use of night chats
Split positions as scum when bussing
Clear yourself as town
Identify and target who will be your enemies
Pressure others to discredit them and hinder their ability to convince others to follow their reads
No mens-reya (too dumb to crime) defense legal strategy
Estimate the strength of your reads
Make a good first impression
Same perspective = same alignment?
Too scummy to be scum
Lynch All liers
Scum avoid giving hard reads
Ask for towns perspectives to identify what they see
MM Doublethink- I've had this thought and decided to share it here, both for the sake of sharing something interesting and to know if it made sense.
Given that scum can be caught by perspective "slips", or more exactly perspective dissonances, and given that those dissonances are quite common, the concept of doublethink is interesting for successful scumplay. The possibility of convincing yourself that you are town and do not know the scum, even though you have read your rolecard, could prevent perspective dissonances. Therefore, "fooling yourself" while remaining in control (to push mislynches instead of teammate lynches, for instance) could be the key to the most natural and effective form of scumplay ever. You'd basically have to be O'Brien for the duration of the game. That seems... interesting.
Divide and conquer
This is exclusively a scum play and arguably something scum should always try for in day chat. The idea is to identify players bias and try to divide the town into two or more groups playing their bias and grudges against etch other while ‘pocketing’ players to increase your influence over day chat.
-Identifying Confirmation Bias and splitting the town
So what you want to look out for here is players grudges and their reads on other players. Keep track of who thinks what about who and play that to your advantage. Also watch for how players are reading other slots. If you notice Player A scum reads player X and puts a lot of value in connecting contradictions to scummy behavior then include some contradiction in your read of player X. This will help pull player A onto this train or if you are not starting a train it could push him over the edge to really gunning for that player. Generally speaking you just need to split the town into two or more groups and then you can start to kill the town off one by one maintaining a balance of power between them and keeping you in control of the situation.
from a wolf perspective, you realize that you have to pick your targets carefully. you can't be on good terms with everyone; as you said, you have to lynch villagers. and often the people you lynch will be among the first to figure you out and suspect you. so you have to balance your targets carefully, because its dangerous to make enemies as a wolf. you pick someone to lynch so that when they turn against you, you know you will be able to overpower them and get them lynched over them getting you lynched
This is the other half of the play. You need to get a few players to pretty much read you as cleared town and follow your reasoning. For every player you pocket you greatly increase your ability to control the day lynch. Even if you are not able to fracture the town if you can simply pocket 2 players you will immediately have massive influence over the town and put yourself in an excellent position. The simplest way of doing this is to watch for when a player gives away soft tells that they scum read a slot and then follow up behind them with much more solid reasoning along the same lines. This will get them to think that you are thinking from the same perspective and objectively make them trust you more. It also helps if you see them in 1v1 situation to pressure the other player (as opposed to defending them.) Players get cautious when someone else is defending them but they get along well when there is a ‘common enemy’ between you two.
Another large advantage to playing in this way is that it gives you purpose in day chat so instead of just hanging around trying to not get lynched you have focused goals that provide a reason to create content as well as have passion in your posts. Its possible to get caught for having this kind of intention in your plays but in my experience it is pretty rare.
Useful Bits Shit name?
Game tempo- Setting a days lynch
-Writing to use different part of brain
LYLO and MYLO
Kingmaker play- setting yourself up to control end result. Always go for or go meta
Getting Read as (gut read indicators)
The Explanatory Gap
Proper use of Codes
This is the all too familiar situation of posting a code to solve on the next day in order to confirm that you are still you. In multiple games I have seen someone post a weak code that the disguiser then solves falsely confirming themselves as ‘not disguised’. In a word a bad code can do more harm to the town than not making a code at all. Keep in mind that the scum player does not have to actually solve your code at all, he only needs to convince the town that he has solved it. Here are some examples of Bad Codes:
_ __ _ _____ = I am a robot; This could very easily be faked with any phrase that matches this letter count.
26535 89793 23846 26433 83279 = a section of pi in math; Kinda clever but simply google searching breaks it along with any other string of numbers. Even if its not what you meant someone could just lie and say it was.
Pretty much anything math related
Anything using a cypher program; These are very easily broken with some programs made specifically to do so
Data can be taken from a players games and analyzed to identify meta trends.
These trends can then be compared to data taken from a current game to identify specific factors of a player such as alignment
No clue where to put the shit
Focus on having fun- Avoid pushing to draw emotional reactions
(Liers Delight) factor of enjoying Mafia
Potential shit for theory giggles
-Sociometrics- Weighted social networks system
Cam mafia/ Real life shit
Unethical- Macro meta crunching, Sociometrics, Not looking at your role, Not revealing tells on players with meta,
Reason to see both sides:
Engeneering control theory has something called a feedback loop
If you remove negitive feedback the loop almost always goes divergant and chaotic
Balance is imoportant because it keeps you from drifting to an extreme. If you only listen a postion you support that position will only be amplified until its screaming its position as opposed to presenting logic
Detective interview (This is a non listed interview and I would appreciate not poking the interview in youtube comments or distributing the address)