Back to Epicmafia

**Gamechangers**

over 9 years

Lie Detection: > [name] are you [mafia/town]? Less than a 1 second delay means anticipation of said question, often implying guilt. It takes around 500 milliseconds to process environmental input and another 500 to deliver a response, so if a person answers, "no" within a second or two (depending on typing speed), then they're most likely truthful. If they delay any longer or deflect the question, then they're likely lying. I made sure you were present before I asked if you were mafia, to rule out any false positives from you being AFK. More time can be given if someone types more than a simple, "no", but you have to look out for speech patterns that guilty people use. Saying, "nah", or longer versions than a simple, "no", is often an indicator of guilt. Repeating that they're blue usually yields the same results.

>Contractions These are shortening of multiple words like, "do not" into, "don't". Much like the rest of these techniques, you need to figure out a baseline of the person you're testing. A baseline is how people act/type when they're known to be answering truthfully. People usually stick with either using contractions or not using them when they're truthful; with the exception of emphasis on a statement. If you know which a person uses when they're being honest, then the presence of the opposite in a sentence can mean they're lying.

>Periods Like the contractions method above, establish a baseline of a person as to whether or not they use a period at the end of a sentence when they're honest. Any further sentences from that person with the opposite of what they usually use can mean they're lying.

Mafia Detection: >First to Vote/Talk While this may mean a person is eager to play, it can also mean that a person was in a mafia meeting at night, since partners usually talk and take longer to hammer than non night-meeting players.

>Grudge Kills When a person dies first, it can be random, or it may indicate the mafia played with that player in a previous game. By checking the last few games of the person who died first, you may find useful information on people to focus your attention on. I.E. if the player who died was cop in a previous game, who won, then the mafia from that game might kill him in the game they're currently playing.

>Adjacent Voting Interestingly enough, mafia partners take subconscious cues from each other. This can manifest as a less experienced partner voting directly after their partner votes.

>Manic Behavior Sometimes the Mafia based roles will talk excessively during the day, in an attempt to not seem "lurky".

>Tangents To deflect pressure and/or guilt, Mafia sided players are more prone to talking about things other than the game at hand. Also, the more details they give, the more they're under pressure.

PR Detection >First to vote PRs usually have night actions, so they know they're a PR. This will make them excited and anxious, so they sometimes will be impatient and quick to vote.

>Vote-Tailgating Since PRs generally do not want to out themselves at first, they will try to be subtle in all their actions, but that comes across in different ways. They may quick-vote directly after someone else votes, as to not draw attention to themselves.

Enjoy!

Was this helpful information to you?
16
No
7
Yes
over 9 years

Orly says

In short, even with them knowing you know their strategy, it is still their best move. With that fact in hand, we can better understand why certain tactics are pursued relentlessly in certain setups.


And if this becomes the case then every kind of read becomes arbitrary and the only rule of thumb is you will lose and win according to the expected ratio, from all possible actions.

A set-up is 53% town-sided? Looks like the majority of players will win the majority of the time!
over 9 years

VictoriaEldor says

Systematic approaches apply to every person in the game, because it is in Mafia's best interest to be pro-town in appearance as well. And if that should be an expected behaviour, people will just start lynching based on who is anti-town.

And then this process flips, making the entire thing completely redundant and based on dodgy assumptions, one of which being that I need to do anything more than have an opinion different from someone else in order to be labelled a troll


The process may flip occasionally, but usually, there are certain behavioral trends worth following and pursuing in-game, because the expected payoffs of a player pursuing them make that strategy their dominant play. In short, even with them knowing you know their strategy, it is still their best move. With that fact in hand, we can better understand why certain tactics are pursued relentlessly in certain setups.

A good example is the softing of a guilty on D1 by cop in Fancy Pants. Even though mafia knows it will happen, cop still usually does it, because even if cop dies, it's better than risking dying without outing the guilty.

Of course, because you know this is a dominant strategy, you see blues hypoing or softing as a potential roleswap, as well as mafia who prep for a cop cc by also softing. There are plenty of variations on the move, and sometimes cop will intentionally wifom and not even claim while a blue has softed (to stay alive), but you have to recognize the current meta in Fancy Pants is established by this specific development of expected payoffs.
over 9 years
zzz

Systematic approaches apply to every person in the game, because it is in Mafia's best interest to be pro-town in appearance as well. And if that should be an expected behaviour, people will just start lynching based on who is anti-town.

And then this process flips, making the entire thing completely redundant and based on dodgy assumptions, one of which being that I need to do anything more than have an opinion different from someone else in order to be labelled a troll.

And I will, consistently. Because I only care about solving puzzles and objective facts. Should people be offended by that they wont end up being my friends, and that's their problem, because I don't require their approval to exist, or make mistakes. And I certainly don't need other people's approval to take risks in the pursuit of bettering myself.
over 9 years
Damn that's some good s%@t. Every noob should be directed to that thread.

I'm just offering these tips so people won't use them anymore. Maybe they'll be a bit more conscious of these systematic approaches and make the game a bit more challenging.

EM has gone down hill for years now. It's not angular's fault. It's not Lucid's fault. It's the corruption of the moderators and the apathy of players.

S@%t when this site started the only troll I remember was ABC.
over 9 years
Victoria is a bad troll. pranay is jealous because this isn't a thread he made.

you offer some solid starting points for technical gathering of information as a player. i think you have a lot of promise, if you formulated these ideas on your own.

lots of players fail to recognize that these are key behavioral patterns emerging from standard gameplay.

yes, as arcbell says, it can't be proven, but that's because social dynamics alter the patterns endlessly.

still, this is a good place to begin forming reads in a routinized fashion. good job.

i recommend reading the following, if you're interested in more game theory:

https://epicmafia.com/topic/66277
over 9 years
It's so cute of how you're trying to understand a concept that is so far above your head. Now you're trying to save your ego by countering me at whatever I say; it's quite transparent. I'm going to stop talking to you for two reasons:

1. There are some tips from other players in the comments that I don't want people to miss.

2. Now I know you're not worth the time.
over 9 years

Nihilo says

You're right. It isn't funny; it's sad. Okay I'll spell it out for you.

Much like with Zeno's paradox of Achilles and the tortoise, there will always be infinite points and infinite data; however, the manifestations after a certain scale have similar effects. You don't need to know every single piece of data to make a judgement. All you need are the patterns that sets of data incorporate, to follow the axioms that accompany them.


Are you aware that the paradox you're talking about is complete bollocks as well and is a complete non sequitur? The only way that someone cannot overtake someone who is slower than them in a race is only due to the fact that there is a finite distance to travel.

If the slow runner is given a head start it becomes impossible to win against them in a race only assuming that from their current starting position they will reach the finish line in less than than the one who is simply faster.

And this has nothing to do with the variables presented, either. Because it is bollocks, and the premise of your argument became bollocks the moment that it was impossible for you to determine a baseline about someone from the first encounter.

And it was bollocks again because by the second reaction test it will have already been affected by the first, permanently.
over 9 years
You're right. It isn't funny; it's sad. Okay I'll spell it out for you.

Much like with Zeno's paradox of Achilles and the tortoise, there will always be infinite points and infinite data; however, the manifestations after a certain scale have similar effects. You don't need to know every single piece of data to make a judgement. All you need are the patterns that sets of data incorporate, to follow the axioms that accompany them.
over 9 years

Nihilo says

So you're saying that there are so many variables it's not even worth it to try and find patterns in the chaos.


I was supposed to laugh but this one isn't even funny considering my other threads.
over 9 years
So you're saying that there are so many variables it's not even worth it to try and find patterns in the chaos. LMAO. Sounds like a moronic cop-out for being inadequate at lie detecting.

As for the second part, I never said these work 100% of the time. These are tools that should be used in conjunction with each other, just like lie detection in real life. I have a feeling you don't know much about that. There's always an exception to a rule, but that doesn't mean you throw the rule out. Jeez.
over 9 years

Nihilo says


Everyone who plays is a human, and humans respond with surprised when they're presented the unexpected. Unless you're a psychopath and can falsify your reactions, everyone will act the same way.

As I already said in the thread, make sure the person is there BEFORE you rxn test them. That way you can eliminate false positives.


First part; everyone being human does not guarantee that all humans are the same, for this you'd need a level of tolerance. You're also going to have to factor in the tolerance given their internet connection, as well as personal influences in their life accumulating over a potential span of 16+ or more years in most cases to compare with others of that age group. The amount of variables to get a deterministic idea of that person would probably be a string so large it could not fit on any known memory storage device in the world, even people who remember every nanosecond of their lives are rare or outright non-existent.

As for the second point, what if they type with one hand after just sitting down, and you've just immediately jumped on them? This tell is bollocks.
over 9 years
Really accurate and helpful
over 9 years

pranay7744 says

i mean these do give some info, but ive learned not to rely just on this. I combine the above with meta on player, game situation, time of day, place where person lives and temperament carried from previous game.


First sentence, i is lowercase, the second I is capital. This deviation from usual capitalization signifies that pranay engaged in active deception, revealed by this freudian grammatical slip. Am I doing this right?
over 9 years
pukes
deletedover 9 years

Nihilo says

. I gave them to you for free without any strings attached.



just how i like it
over 9 years
WHY DA FAK ARE YOU COMPROMISING OUR SECRETS?
over 9 years

Nihilo says

It's not up to me to "prove" these things. I gave them to you for free without any strings attached.


What if I told you that if someone posts in the 4-8 second interval they're always town due to natural tempos. If you expect people to believe what you say unquestioningly, then you expect your observations to be obvious, meaning they're worthless. If your observations aren't supposed to be inherently obvious, you can't expect someone to believe them without giving any support.

So yes, it is up to you to "prove" these things if you'd like to be taken seriously.
over 9 years
@Victoria: At least you raised some good points without dismissing everything I said entirely, but you're wrong.

Everyone who plays is a human, and humans respond with surprised when they're presented the unexpected. Unless you're a psychopath and can falsify your reactions, everyone will act the same way.

As I already said in the thread, make sure the person is there BEFORE you rxn test them. That way you can eliminate false positives.

As for the people that see the question as pointless, I've met them. Those are the same people that troll and are a small minority.
over 9 years
Lol. I'm an original player giving you guys gold reads, in hopes for a new generation of non-dumbass players. I should have known that EM was broken beyond repair.

It's not up to me to "prove" these things. I gave them to you for free without any strings attached.

I'll leave this post up for anyone who isn't too arrogant for their own good, though; despite the naive criticism.
over 9 years
EM: gosh there are so many noobs on this site i wish they would learn how to play

*someone tries to help*

EM: Gosh this is useless advice who didn't already know this
over 9 years
Nihilo, dude, U have to prove it.
over 9 years
" so if a person answers, "no" within a second or two (depending on typing speed), then they're most likely truthful."

What about the times when the person replies quickly simply because they have been caught, and like a child who doesn't want to be in trouble, their kneejerk reaction is no.

This applies to people who were abused as kids more often than not, because they'd be especially averse to being 'wrong' in cases where being wrong meant punishment.

Then there are people who are not paying attention. Then there are outliers who just don't care and see the question as pointless.

Then there are people are are only fast simply because they desperately need to be confirmed as non-mafia. Etcetera.

This is why ranked / competitive should always be anonymous.
deletedover 9 years
ya i'm not being 100% serious but it is a relatively decent read i've used a lot of times just as a starting point for fosing someone in jan.


another is when a noob comments on the naked (but not naked mafia only naked town. If they say "devante you're naked" at the start of the game usually a noob is mafia, there's a lot of little things that i use but those 2 are good ones then i poe partners based off those little tells
over 9 years
Someone told me about it, however the following 2 games the person was town so I don't use it unless their are other gaps in their play.
deletedover 9 years
if u hammer the person who says inno then the name in cop cc's you'll always hammer mafia, that's how i scum hunt bros