On Compromises

[hitogoroshi]

Your job is not to always be voting your top scumread. Your job is to secure the best lynch out of all the lynches that are possible.

I think this is something most newbies don't get (ever seen a newbie player, in their first non-newbie game, call out everyone for "bandwagoning"?) and most good players do. More than that, I think learning this is one of the big thing that turns you from a newbie to a good player. I know it made a big difference for me. I saw it in Mafia 105 (Caught in the Crossfire), I took some time to digest it, and I used it in Open 189, where multiple players noted the sharp improvement.

I'm not making this as a "look at this new thing" thread. I'm sure most of you know it already. But I think newbies DON'T, and it's something that should be written down to perhaps ease the transition.

Here's a way to think about it: every player on the game is in one of two lists: "Can be lynched today" or "Cannot be lynched today." Your goal is to get your top scumread in "Can be lynched today", and then lynch them. But as the day charges on, people are less responsive to 'new' cases. What if you're still alone on your wagon and you hate the leading wagon? What if, despite your sincerest efforts, your top scumread isn't getting lynched today?

Bad players sit on their wagon, loudly criticize the leading wagon and keep their vote on their top suspect. This is called "being worthless." It would literally be better if you weren't in the game - because then the threshold of how many townies it takes to lynch a scum might go down.

Good players look around, say "which of the potential wagons do I dislike the least?" and vote for that wagon.

Your goal is to lynch scum. Ideally you're lynching the scummiest person, but it's better to compromise on the "scum" part than the "lynch" part. Lynching someone you're not sure is scum is better than lynching someone you're REALLY not sure is scum. Not voting to lynch because you don't love any wagon makes it all the more likely scum are involved with the lynch that ACTUALLY went through.

Or, to to say it another way:

Always be influencing who is getting lynched.

[Amrun]

Agree.

But prepared to be called scummy for it by SOMEONE.

[AlmasterGM]

Bad players start their own wagons and then sit alone on them.
Good players change wagons in order to secure lynches.
Great players start their own wagons and then convince the good players to join them.

[hitogoroshi]

Yep. If you can always get your top scumread lynched, there's no need for this thread.

Usually even the hyper-rhetoric players have SOME lynches they can't get through, though.

[hitogoroshi]

Chatting with Mina. Found another way to say this.

These are the town skills:

1. Being correctly read as town. ("Transparency")
2. Having accurate reads. ("Accuracy")
3. Actively working to secure the best lynch out of all the lynches that are possible. ("Competency")
4. Accurately judging which players can be lynched. ("Foresight")

My point is, in essence: too many new players think mafia is just Transparency, Accuracy, and "lynch your top scumread." (Or, the worst sort of all, who take Transparency off that list. That's pretty much the definition of a VI.)

There are two facets to Competency. The first is case building/rhetoric - the ability to get your top scumread lynched. If you can always get your top scumread lynched, that's all you need for complete Competency and Foresight - you KNOW (Foresight) that you can always get your top scumread lynched (Competency).

But that's virtually impossible to have 100% uptime on. And here we find the second facet of Competency - Pragmatism. This is, in a nutshell, looking at the available lynch options and voting for the least bad.

So, let's split up Competency. I'll shorten the "Rhetoric/Casebuilding" thing to just "Rhetoric" - rest assured that title applies equally to RIGHTEOUS CAPS RAGE and giant, imposing wall cases. As long as it works to get someone lynched.

1.Being correctly read as town. ("Transparency")
2. Having accurate reads. ("Accuracy")
3. Actively working to get your top scumread lynched. ("Rhetoric")
4. Accurately judging which players can be lynched. ("Foresight")
5. Securing the best possible lynch when your top scumread can no longer be lynched ("Pragmatism")

Notice something about Pragmatism? The "best" part falls under Accuracy, the "secure" part falls under Rhetoric, and the "when" part falls under Foresight. It's not its own skill.

I contend that

Pragmatism is something you just answer yes or no to, and if you answer no, you ARE a bad player.

Fortunately, it's easy to fix. Just say, right now, that you'll agree to lynch someone ELSE if your top scumread can't get lynched. You won't always succeed .Maybe you'll judge it wrong and be too stubborn, or flow too freely when you could have pushed it all the way. It happens. Sometimes your Transparency, Accuracy, or Rhetoric are bad too. The important thing is that you LEARN from it, and realize that Foresight is, like anything else, a skill to be worked on, and that Pragmatism is not a skill at ALL but a yes/no that serves as a stepping stone into good play.

---

As an aside, you can classify peoples town games by the first four categories. (I assume that if you were someone who answers "No" to Pragmatism, I just converted you.)

For example, a self-assessment:

Transparency: Excellent
Accuracy: Below average
Rhetoric: Above average
Foresight: Below average

[SpyreX]

You do run into some sticky ground with this. Specifically,

when

does pragmatism come into play? I've seen more than enough last minute pushes flip scum in calm circumstances (and, of course, if someone has enough 1 and 3 they can make almost any push) that pushing for "give in" in essence could cause suboptimal lynches.

Should you be that guy at deadline-2 hours sitting on your prize pony? Probably not, but don't make the mistake of caving in days early.

[hitogoroshi]

Random point: I dislike self-avowed Cassandras, and I finally have the words to say it. If you have excellent Accuracy, but awful Transparency, Rhetoric, and Foresight, what that means is that you need to work on looking town, getting your suspects lynched, and judging when it's time to switch. I find too many townies will judge their effectiveness SOLELY on their Accuracy, and bitch in the postgame about how great they are when they were mislynched. Don't they make you angry? Now you have words to say it.

You do run into some sticky ground with this. Specifically, when does pragmatism come into play?

The answer to this question is game-dependent, and how right your answer is what Foresight is.

I was arguing with Mina, continually hitting "I agree with your main point, but..." when I KNEW there was a black-and-white element of good player/bad player in there. Eventually, I figured out the answer. Pragmatism is ALWAYS the answer, and a non-Pragmatic player is bad. WHEN you should be Pragmatic is the entire skill of Foresight, and some players are better at it than others.

In particular, I think both Mina and I have rather poor Foresight, but mine is too sensitive (I'm too much of a push-over) and hers isn't sensitive enough (Fast-game syndrome of always assuming a last-minute push can be rammed through.) That was more or less what we were arguing about.

[hitogoroshi]

Another thought: While all four skills are important, I think they're in roughly descending order of importance.

The most important thing is to not be mislynched. The next important thing is to ensure that when you direct the game, that you're doing it intelligently. Then it's important to GET LYNCHES on those reads. And, finally, it's important to admit when your rhetoric wasn't enough and compromise.

[Regfan]

Lack of compromise is something I also noticed not too long ago, I've spectated and participated in some games which were lost soley because players were unable to compromise. Unfortunatly it's also one of my pitfalls, too often do I push my own case rather than voting along with a lynch I'm comfortable with and it's something I'm attempting to work on. Although not everyone keeps lists of reads I believe every player should know flat out which players they will put their foot down and refuse to lynch regardless of the situation, once they've managed to defend and prevent their strong town-reads being lynched they need to prioritize their lynching order and to be perfectly honest priority isn't based soley on how scummy you read them to be.

A skill I believe you have missed mentioning or bringing up is actually tactical analysis, evaluation and preparation, although the towns biggest opportunity to influence the game is during the day it's also entirely possible to do so at night. I'm not suggesting mass-setup speculation in every game but I believe a good player needs to know when to take a brief break from scumhunting and plan the optimal way to progress taking into account the power-role claims and setup mechanics.

[Hoopla]

Excellent transparency indicates a poor/undeveloped scum game. Transparency is a necessary sacrifice if you want to protect your scum game. Unless you're playing with different people each time, or constantly one-upping your skillset to be seen as town in different/deeper ways than before (both are unrealistic beyond the short-term), you won't always be seen as town.

Accuracy and Pragmatism become more important town qualities once you go beyond that level.

[RedCoyote]

Always be influencing who is getting lynched.

I like this quote. I like this quote

a lot

.

[Vi]

This thread is awesome and when I get back I will take a more productive look at it.

[Faraday]

Yes compromise is important butttttttt.

How many players actually 'refuse' to move their vote from their top scum read though? This seems like something that is incredibly rare. It's also not even really something I've seen play a part on newbie games either, most people are willing to switch to ensure a lynch, I mean we don't have a slew of no lynches or anything. I don't think lack of compromise is a significant problem in games I've played. Newbies or otherwise. I mean there are some rather hilarious exceptions but still.

On the flip side of the coin if you're 'too pragmatic' and then complain about how you were voting that player but had to switch to secure a lynch then you clearly weren't working hard enough either.

Bad players sit on their wagon, loudly criticize the leading wagon and keep their vote on their top suspect. This is called "being worthless." It would literally be better if you weren't in the game - because then the threshold of how many townies it takes to lynch a scum might go down.

Well not always, no. There are times when doing this is useful, especially if there's plenty of time till deadline and you think criticising the leading wagon can help stopping it. Of course I don't think that's actually the times you're talking about here, doing this right up untill deadline is bad play, but not all of the time, that's way too black and white.

Anyone who sees a deadline in X days and switches(too early) to 'ensure' a lynch isn't working hard enough. 'Oh golly I got no chance your wagon has 2 more votes I'll never get the lynch on the guy I think is obvscum'. If you can't do things to switch momentum in X days you're not trying hard enough. Don't be afraid to push your wagon, that's being a scrub. Getting a 'last minute' vote switch isn't that hard, people get cold feet on wagons all the time.

Ftr I'm probably one of those that believes the lynch I'm pushing is always significantly better, and I think more-often than not I can manage to get it pushed through, there are exceptions but I think if you put in an effort people will look at the player you suspect, and it's much much more effective if your vote is there for that, too. Makes the chances of people switching far higher.

edit: way too early to be arguing mafia theory

[Faraday]

Excellent transparency indicates a poor/undeveloped scum game. Transparency is a necessary sacrifice if you want to protect your scum game. Unless you're playing with different people each time, or constantly one-upping your skillset to be seen as town in different/deeper ways than before (both are unrealistic beyond the short-term), you won't always be seen as town.

I also agree with this.

In fact I completely disagree with Hito that they're in descending order of importance. Being mislynched and being transparent are not 2 sides of the same coin. There's a fucking massive wasteland of middle ground. I'd take being right + convincing while people find me scummyish over being 'obvtown + convincing' but wrong any day. Of course it's not near as black and white as this, and it's really no surprise hito values transparency more than me. Ideally I suppose you want all aspects of your game to be very good and to be able to replicate them identically as scum, but that's a pipe dream.

[hitogoroshi]

A skill I believe you have missed mentioning or bringing up is actually tactical analysis, evaluation and preparation, although the towns biggest opportunity to influence the game is during the day it's also entirely possible to do so at night. I'm not suggesting mass-setup speculation in every game but I believe a good player needs to know when to take a brief break from scumhunting and plan the optimal way to progress taking into account the power-role claims and setup mechanics.

Yeah, this guide is for vanillas in simple games. Didn't want to bother talking about getting Theme's in line/coordinating PR's and such.

Excellent transparency indicates a poor/undeveloped scum game. Transparency is a necessary sacrifice if you want to protect your scum game.

To a degree, yes. But remember, transparency is just the word I took to mean "being correctly read as town".

Having all of your votes be for scum would also get you correctly read as town, for example.

It is true that a deliberate transparent playstyle (like mine) will hurt your scum game while helping your town game. But Transparency, capital T, isn't NECESSARILY having a transparent style. It's getting read as town correctly by however it is you do that.

How many players actually 'refuse' to move their vote from their top scum read though? This seems like something that is incredibly rare. It's also not even really something I've seen play a part on newbie games either, most people are willing to switch to ensure a lynch, I mean we don't have a slew of no lynches or anything. I don't think lack of compromise is a significant problem in games I've played. Newbies or otherwise. I mean there are some rather hilarious exceptions but still.

On the flip side of the coin if you're 'too pragmatic' and then complain about how you were voting that player but had to switch to secure a lynch then you clearly weren't working hard enough either.

Oh, of course someone will always GET LYNCHED, or at least almost always. But you shouldn't compromise on whatever lynch is around five minutes to deadline - you should "compromise" only on the BEST POSSIBLE lynch. This is either Rhetoric making your scummiest read a possible lynch, or Foresight leading you to support a counterwagon in time to MAKE it a possible lynch.

The point is, it's not park-park-park-park-oh shit five minutes to deadline vote: townread at l-1. It's park-park-park-ah fuck me he's not getting lynched today, let's get nullread at l-2 cause that MIGHT happen, and then maybe nullread is lynched instead of townread. You should switch the instant your lynch isn't happening. Foresight is the skill that tells you when this is, and Rhetoric is the skill that enlarges the timeframe for how long it's possible to get your top scumread lynched.

Well not always, no. There are times when doing this is useful, especially if there's plenty of time till deadline and you think criticising the leading wagon can help stopping it. Of course I don't think that's actually the times you're talking about here, doing this right up untill deadline is bad play, but not all of the time, that's way too black and white.

Anyone who sees a deadline in X days and switches(too early) to 'ensure' a lynch isn't working hard enough. 'Oh golly I got no chance your wagon has 2 more votes I'll never get the lynch on the guy I think is obvscum'. If you can't do things to switch momentum in X days you're not trying hard enough. Don't be afraid to push your wagon, that's being a scrub. Getting a 'last minute' vote switch isn't that hard, people get cold feet on wagons all the time.

Ftr I'm probably one of those that believes the lynch I'm pushing is always significantly better, and I think more-often than not I can manage to get it pushed through, there are exceptions but I think if you put in an effort people will look at the player you suspect, and it's much much more effective if your vote is there for that, too. Makes the chances of people switching far higher.

Right, what you're saying here is that you shouldn't get off your preferred lynch if you can still get them lynched. That's true. Foresight is the skill that tells you whether you can still get someone lynched. All you're saying here is "some people have bad Foresight", which is true.

In fact I completely disagree with Hito that they're in descending order of importance. Being mislynched and being transparent are not 2 sides of the same coin. There's a fucking massive wasteland of middle ground. I'd take being right + convincing while people find me scummyish over being 'obvtown + convincing' but wrong any day. Of course it's not near as black and white as this, and it's really no surprise hito values transparency more than me.

I think you're missing my definition of Transparency here (being correctly read as town.) If you convincingly lead lynches on scum, you'll probably be read as town.

And I can't really think of a situation where people don't correctly think you're town but are still being convinced to lynch the people you think are scum. If someone is voting for your scumspect because of your case, implicitly, they think you're town. That's why being correctly read as town is the first step to anything. The most important thing is not being mislynched, which requires you being read as town, and every subsequent step is getting your way as much as possible, which requires you being read as town.

[Faraday]

I guess I just don't think:

I contend that Pragmatism is something you just answer yes or no to, and if you answer no, you ARE a bad player.

This is something I see people answering no to. I can't think of ANYONE who refuses to vote outside of their top suspect in my last I don't know how many games, apart from one guy.

Which brings it back to foresight but I think you're underestimating the way momentum can change really fucking dramatically in a game (or overestimating the amount of foresight any one player can have), lateish good compromise lynches aren't hard to secure, they happen quite often, I guess my point is compromising too early is really really bad. Wagons can be put together really fucking fast, you don't need to be pragmatic early and should be trying to get your top suspect lynched for most of the day, really.


And yeah, your definition of transparency is not what I thought. Sure, that's fine. Getting scum lynches leads to that anyway, I don't think you need to actively try to 'look town' personally but yeah, I understand what you mean now.

[Zachrulez]

Pragmatism comes into play only after the moment passes where it becomes impossible for your top suspect to be lynched that day, not before. (Which is a lot closer to the deadline than a lot of people tend to consider)

[hitogoroshi]

Pragmatism comes into play only after the moment passes where it becomes impossible for your top suspect to be lynched that day, not before. (Which is a lot closer to the deadline than a lot of people tend to consider)

First sentence is objectively true. I would disagree with the parens - I think that most newer players wait too long to compromise and the lynch they secure isn't as good as the one they could have secured by pushing a counterwagon sooner (even if neither is their top scumread.) Haven't played a newbie in a while, though, so maybe I'm wrong.

[Faraday]

Honestly a lot depends on the gamestate obviously, you should know that in a slow lethargic game the chance of a bigger wagon creeping to deadline is high. Luckily I don't play many like that, and the faster the game pretty much the longer you can wait.

I think it's something that comes with experience.

[The Fonz]

Game days can basically be broken down into three parts.

1) Early-day: up to halfway to deadline, or before the first player gets put at L-1.
2) Wagon consolidation time: second half of allotted time, or when there's one strong lead wagon.
3) Deadline time.

The level of viability needed for a vote to be worthwhile goes up as you progress through the stages. At the start of the day, you want to be voting your top suspect, regardless.

By about a week in, you should have an idea of whether there's going to be enough support for there to be a serious chance of getting suspect number one lynched, and if not there will hopefully be more than one wagon available.

In part three, you need basically be willing to vote for almost any wagon if it's no lynch without majority. If it's plurality lynch, it plays out remarkably similarly to phase two.

I term these parts the 'Vote top suspect' 'Vote best viable wagon' and 'Vote any viable wagon' stages. The key thing to remember, is even if it's early day, there's no really good reason not to comment on the top wagon, or if there are two big wagons, express a preference between them. This doesn't mean you've given up on your top suspect, but it shields you from accusations that you are deliberately tunnelling to avoid commenting on the most important game issues.

By about halfway through the day, you should know which wagons are viable. If the wagon you're on is within a vote or two of the leader, or is the second biggest wagon regardless of its strength relative to #1, then keep going. If not, there's a few things that are decent town options. If there's a rainbow votecount, you want to start engaging with the other people on one-man wagons, to try to see if they can be persuaded to join you, possibly by offering them a good look at their top suspect in future days. If, however, there are multiple viable wagons and the one you're on isn't one of them, it's time to decide between them. There's absolutely nothing wrong with saying 'X is my top suspect, but isn't viable, therefore Y is a better lynch than Z.'

[Hoopla]

Excellent transparency indicates a poor/undeveloped scum game. Transparency is a necessary sacrifice if you want to protect your scum game.

To a degree, yes. But remember, transparency is just the word I took to mean "being correctly read as town".

Having all of your votes be for scum would also get you correctly read as town, for example.

It is true that a deliberate transparent playstyle (like mine) will hurt your scum game while helping your town game. But Transparency, capital T, isn't NECESSARILY having a transparent style. It's getting read as town correctly by however it is you do that.

If you're often being correctly read as town, then you must be doing certain things to warrant people reading you that way. It logically follows, that if you don't want your scum game to suffer, you must adopt a similar play style to how you do as town. If you can't perform to that level as scum, then yes, your scum game is being suppressed by your town game. If you can perform to that level, you won't be read as town for the things you do any more. Thus, you need to be constantly improving at an increasing rate, with your town-game one-upping your previous efforts as scum and so on. There will be a point where the amount needed to grow to be at the next level will be beyond your ability.

[Glass]

If you're often being correctly read as town, then you must be doing certain things to warrant people reading you that way. It logically follows, that if you don't want your scum game to suffer, you must adopt a similar play style to how you do as town. If you can't perform to that level as scum, then yes, your scum game is being suppressed by your town game. If you can perform to that level, you won't be read as town for the things you do any more. Thus, you need to be constantly improving at an increasing rate, with your town-game one-upping your previous efforts as scum and so on. There will be a point where the amount needed to grow to be at the next level will be beyond your ability.

What I am hearing is that you want our town game to suffer because otherwise our scum game will suffer. Or am I missing something?

[Regfan]

I believe Hoopla is saying that you need to find a middle ground, you don't want to be so easily read as town that you're defenceless as scum.

[Vi]

If you're often being correctly read as town, then you must be doing certain things to warrant people reading you that way. It logically follows, that if you don't want your scum game to suffer, you must adopt a similar play style to how you do as town. If you can't perform to that level as scum, then yes, your scum game is being suppressed by your town game. If you can perform to that level, you won't be read as town for the things you do any more. Thus, you need to be constantly improving at an increasing rate, with your town-game one-upping your previous efforts as scum and so on. There will be a point where the amount needed to grow to be at the next level will be beyond your ability.

What I am hearing is that you want our town game to suffer because otherwise our scum game will suffer. Or am I missing something?

You heard Hoopla correctly.

If you're like me and Always Town, being transparent is a great thing. People treat you like royalty, you pick the Day 1 lynch, you set up the foundation for opinions that people will use in later Days, you die within the first two Nights and don't have to worry about pesky things like LyLo decisions or spending too much time with people who annoy you. But if you draw scum, you will find out the hard way what all the REAL scumtells are because you'll blunder into making them. And looking as Town as normal when you're scum is as hard as the bar you set with your Town transparency. Your scum games will be miserable experiences as long as you're playing with people who know your Town meta, because if they're at all good at this game they'll pick up on the difference. They may not be able to put their finger on it, but they'll just say that you "feel off" and there's nothing you can do about it.

I suspect this is why some of the most renowned Mafia players onsite have a tendency to be somewhat insufferable ingame (the converse does not hold, of course).

[Llamarble]

I feel like playing fully to win condition as town means holding nothing back to protect my scumgame, so I do that.
This means my scumgame will suffer against people I've played with previously,
but I can only be so transparent in a given amount of gametime so it's not like I'll be automaticobvscum in all my scumgames.

As for the compromise thing, if I think one player is most likely to be scum I try very hard to lynch them.
I'll lynch someone who isn't my top scumread only if deadline or other circumstances makes lynching them absolutely hopeless.
Ideally that backup lynch won't be too bad anyway since I defend people from lynches I think are bad.
I'm not sure what uncompromising play would look like. No-lynches at deadline or something?

And I would add the following skills to your list:
1. Helping the town get better reads (primarily providing + clearly explaining insights / getting others to make content off which they can be read)
2. Taking full advantage of mechanics (playing and analyzing nightgame stuff well & manipulating scum into playing it badly / town into playing it well)