It seems that you're using an outdated browser. Some things may not work as they should (or don't work at all).
We suggest you upgrade newer and better browser like: Chrome, Firefox, Internet Explorer or Opera

×
avatar
bler144: Perhaps the best argument for not lynching me, unless you're wholly convinced, is that, if I were scum, I'd probably mess it up.
As you can see, I'm still not trying to lynch you, but unfortunately you no longer stand at the good part of my list.

Krypsyn has provided what seems to be the academic work you asked for, but I somewhat doubt it'll make enough sense (I apologize if I'm underestimating you and you can actually extract the value of out of it), so here are some reasons, once more:

1. We forfeit our only way of dealing with the scum. They practically gain a free NK.
2. Lynching has a chance of hitting scum. Not lynching has 0% chance of hitting scum.
3. Even mislynching has the added benefit of a vote wagon to analyse on the next day. Not lynching gives us yet another Day 1 with one less Townie (possibly more if there are multiple NKs, which is a a very fair possibility given the hints in the OP). We can just not lynch once more since we are still in the dark.
4. Some numbers (they may not be accurate and are simplified down to a single town vs mafia factions without investigations just for the sake of the example) -

Let's say from the 14 of us 3 are scum.

Option 1 - No lynch

D1 - 11 town vs 3 scum = 21.42% chance of lynching scum
N1 - 1 NK
D2 - 10 town vs 3 scum = 23.08% chance of lynching scum (remember we still don't have much more info from D1, so it's essentially D1 with a slightly increased chance)

Option 2 - Mislynch

D1 - 11 town vs 3 scum = 21.42% chance of lynching scum
N1 - 1 NK
D2 - 9 town vs 3 scum = 25% chance of lynching scum (and we have the added bonus of the vote wagon)

Option 3 - Spot on lynch

D1 - PROFIT!!


If someone else has anything to add or correct, please do so.
avatar
bler144: Perhaps the best argument for not lynching me, unless you're wholly convinced, is that, if I were scum, I'd probably mess it up.
As you can see, I'm still not trying to lynch you, but unfortunately you no longer stand at the good part of my list.

Krypsyn has provided what seems to be the academic work you asked for, but I somewhat doubt it'll make enough sense (I apologize if I'm underestimating you and you can actually extract the value of out of it), so here are some reasons, once more:

1. We forfeit our only way of dealing with the scum. They practically gain a free NK.
2. Lynching has a chance of hitting scum. Not lynching has 0% chance of hitting scum.
3. Even mislynching has the added benefit of a vote wagon to analyse on the next day. Not lynching gives us yet another Day 1 with one less Townie (possibly more if there are multiple NKs, which is a a very fair possibility given the hints in the OP). We can just not lynch once more since we are still in the dark.
4. Some numbers (they may not be accurate and are simplified down to a single town vs mafia factions without investigations just for the sake of the example) -

Let's say from the 14 of us 3 are scum.

Option 1 - No lynch

D1 - 11 town vs 3 scum = 21.42% chance of lynching scum
N1 - 1 NK
D2 - 10 town vs 3 scum = 23.08% chance of lynching scum (remember we still don't have much more info from D1, so it's essentially D1 with a slightly increased chance)

Option 2 - Mislynch

D1 - 11 town vs 3 scum = 21.42% chance of lynching scum
N1 - 1 NK
D2 - 9 town vs 3 scum = 25% chance of lynching scum (and we have the added bonus of the vote wagon)

Option 3 - Spot on lynch

D1 - PROFIT!!


If someone else has anything to add or correct, please do so.
Post edited July 08, 2015 by dedoporno
EBWOP

Oh... hmm... both of those quotes in post 465 should be for bler144 from post 464. Post 262, post 464, same thing, right?! Bah!
dedoporno's post seems to have been a double post and not an edit, I saw the original and nothing has changed.
Ninja'd waiting for my own timer to tick down, but I concur with trentonlf.

The post time and the edit time seem to be almost identical (around a second or two apart, at most, possibly concurrent), so I suspect it was just some strange forum glitch.

Anything worth say is worth saying twice, eh, dedoporno? ;)
avatar
dedoporno: *evil*
Tsk, tsk.

Be more careful and don't do it again.

Be glad you had trenton and Krypsyn here to vouch for you.




Wow I was able to put that whole into bold marks? Don't tell me GOG improved this sometime in the past?! That's so unlike them.
Post edited July 08, 2015 by Vitek
variables
avatar
dedoporno: Krypsyn has provided what seems to be the academic work you asked for, but I somewhat doubt it'll make enough sense (I apologize if I'm underestimating you and you can actually extract the value of out of it), so here are some reasons, once more:
I read the paper. It doesn't directly speak to the issue, since the only numbers it runs assumes a single DK + NK each day, and that the lynch is fully random. It also doesn't consider that either faction would have roles that might change the calculus from being entirely random to informed.

It does not directly compare no-lynch vs. yes-lynch scenarios, or third party players. It simply assumes each and every day has a random result deriving from n-2 and that all players' strength are equal and what the odds of victory are for m as m and n vary.

The thing I thought most interesting in the paper was the analysis that the game has anticipated balanced outcome when m = n^0.5. In a game of 14 that would be 3.7, give or take, which I'd wager you could approximate by having 2 factions of 2.
avatar
bler144: I read the paper. It doesn't directly speak to the issue...
No, it doesn't. It is far more generalized and simplified. If all possible variables and game variations were taken into account, I think the paper would span thousands of pages.

avatar
bler144: The thing I thought most interesting in the paper was the analysis that the game has anticipated balanced outcome when m = n^0.5.
I thought the discussion on voting probabilities depending on whether players were odd or even in number was fascinating.

avatar
bler144: In a game of 14 that would be 3.7, give or take, which I'd wager you could approximate by having 2 factions of 2.
That is potentially one solution, yes.

I thought the discussion on voting probabilities depending on whether players were odd or even in number was fascinating.
As did I. I would like to go back and look at that more closely, since an intuition on that was part of my own assumption that no-lynch was viable, or at least not harmful, since it factors into both game length and majority coalition building.

Also should have also said a 3 mafia + 1 independent model fits that guideline as well, but of course could not edit.

I see what dedo is trying to do with the math there, but personally I see a number of flaws with trying to model it that way. But sure, for the sake of argument in that very basic model, by the dawn of Day 3:
No-lynch on day 1, lynch on day 2:
3 dead bodies, 23% chance of 1 dead scum (9 v 2 in your model), 77% chance of 0 (8 v 3)
Lynch both days
4 dead bodies, 5% chance of 2 dead scum, 36% chance of 1 dead scum, and 59% chance of 0.

Which sounds great if you stop right there, but you've also very likely shortened the game a day (which is one place where the odd vs. even issue), and potentially cost your faction one day at the end where the odds of guessing right randomly, or having credible info, are even higher. You also have to weigh in some assumption that you might have offed not just a regular townie, but one with a useful role.

Likewise, the 'vote wagon' theory assumes we garnered useful info from day 1 patterns, which is, at least, not guaranteed. The "added bonus of the vote wagon" theory assumes that on day 2 there are 9 townies, rather than say, 8 townies + 1 independent, and probably more crucially, that you can in practical reality get 7 of those 8/9 to cohere around a single suspect, rather than, say, scum tipping over the wagon on a townie.

Anyway, I really need to drop this and go finish cooking.
avatar
trentonlf: <snip>
Trent: The quoting is so hosed up in this post that I'm not sure what you are trying to say. I'd rather let you try again before I attempt to respond.

avatar
bler144: I believe 'no lynch' is a viable strategy, y'all don't.
Beware generalizations: Most players here think No Lynch IS a viable strategy in certain situations. Day 1 will virtually never be such a situation.

There are a couple of standard guidelines that you may or may not be aware of. I'm going to run through them in case you aren't. A particular game may break one or more of them, but usually these hold true (at least here on GOG):
-All scum know who the other scum are. By extension, all scum know who is town.
-Each townie only knows their own personal alignment.
-Townies should not lie. It is in town's best interest NOT to lie, so that when you flip town other players can trust you meant what you were saying. There is a mantra, Lynch All Liars, because scum have to lie about being town aligned, and scum will lie about other things to make themselves look better or townies look worse. I'm sure you've noticed yogsloth talking about catching people in a lie.
-Town's primary weapon (and quite possibly only weapon) for killing scum is the lynch. If town never lynches, then scum will eventually win.
-A player's alignment and role are revealed by the moderator when the player dies (so the alignment and role become known facts).

I'll take a stab at explaining why No Lynch is usually not a good plan.
Most players here believe that when one lacks any information at all, then No Lynch is not a viable town strategy without some kind of plan to recommend it. On Day 1, there is a general lack of information about the setup, player's alignments, assigned roles, and so on, and unless you know that the information can be gathered in a different way (example, informational roles are guaranteed to be present and town aligned), then you need to lynch people to get reliable information to identify scum. As the bodies start hitting the floor, you can look at who is town and who isn't, and evaluate player's statements with this knowledge in mind. If town doesn't lynch, then only scum will be selecting which people die, and that lets scum rule the game. If players won't lynch without being certain about who is scum, then scum can more easily direct the game or possibly just sit quietly and let town strangle in indecision. Notice that one scum victory condition is when they equal the number of town players (thereby letting scum dictate the lynch).

When townies are active in lynching people though, scum don't have nearly the same level of control in the game. Townies outnumber scum so that a few mislynches can occur without town losing. If townies are lynching then scum have to participate (instead of sitting on the sidelines), which means scum risk making a mistake. Scum will also have to sometimes help lynch other scum, or else risk being outed.

As for whether No Lynch is a good idea in our current situation: Lets say we No Lynch. Scum will select a townie to execute tonight, then we start discussion again tomorrow. We can speculate on why scum selected their victim, but we can't prove anything. If there are no investigative roles, then No Lynch puts us in a worse situation tomorrow than we were in today (there is one less townie, and we aren't any closer to finding scum).

If you can satisfactorily demonstrate to other players that there are town investigative roles that can determine who is scum in time for the scum to be eliminated, you might get more traction for a No Lynch. If the only such role is YOURS, however, then claiming you have that role will pretty much paint a giant target on your back for scum to shoot at tonight. Scum would love to know who has a power so that the power can be neutralized before it becomes a threat to them.

avatar
bler144: sometimes you have to put an idea on the table and let it breathe to realize that it was a dumb idea.

As for Wyrm, I don't think he was scolding me.
I can empathize with "Well, it seemed like a good idea in my head". Also, I wasn't scolding, just informing.

avatar
agentcarr16: Bookwyrm acted strangely, with his little trap. And his general 'wounded' attitude, when many other players (at least some of whom are definitely town) have told him that his traps are unhelpful, is really not what town needs.
Until I return then, on Friday morning, I am going to

vote Bookwyrm

I hope that the deadline hasn't passed by then, as I really want to get a longer read on everyone before I vote for the deadline.
I'd like to ask about this:
1) How do you know that town players have told me my "traps" are unhelpful?
2) If my attitude upon being questioned is unhelpful, then what attitude SHOULD I have displayed?
3) It seem that you are unwilling to lynch most players because they've been quiet or non-helpful. I have a hypothetical for you: If you removed yog, trent, and I from the list of players, who would you select for a vote instead of me?

avatar
bler144: I think yog's post only helped persuade me further "if you can survive to late game, you’re a rock star. I’ve seen it." Assuming that's accurate, it seems a bit like sacrificing one's queen just because you don't like the way her crown was cut.
That assumes that JMich is actually town. If he's scum, then he'd be working to lead town astray at that point. While you don't want to uselessly throw away your queen, you generally would like to eliminate the opponent's queen early.

avatar
bler144: ...that you can in practical reality get 7 of those 8/9 to cohere around a single suspect, rather than, say, scum tipping over the wagon on a townie.
Scum can't just all vote as a block. If they do, then as soon as it is detected, the scum are hosed, and town should be able to nail at least one scum even if only by accident. By the same logic, scum may vote for other scum.

This is why distancing and bussing are things. Also, the hammer vote tends to be scrutinized by townies precisely to check for a scum finishing a townie off.
avatar
Bookwyrm627: If you can satisfactorily demonstrate to other players that there are town investigative roles that can determine who is scum in time for the scum to be eliminated, you might get more traction for a No Lynch...Scum would love to know who has a power so that the power can be neutralized before it becomes a threat to them.
Having power sounds awesome.

I don't particularly care about traction. But that was a wonderful lecture, summarizing points already made multiple times as well as basic rules about the game. I learned so much. ;)

I've never been under the illusion that anyone else is likely to vote no lynch today, whether or not I did. As whoever noted, a no lynch result in this day seems unlikely to come from any no-lynch vote, but rather from gridlock. And on that score, we seem able to compete with the U.S. Senate.

On the bright side, I do think I'm close to deciding whom I'm casting my first serious vote for. But since everyone else is just sitting pat and stalling til the deadline posts, I'm not inclined to rush it.
Got home from work, tired, skimmed through the arguments...I'll have to re-read these in the morning...

Only thing I have to add right now is my suspicion about Vitek's start of game delay. While unusual, how many of you have been using "Chat" a lot? I'll just say from my giveaways, messages do not always show up immediately and the way its set up, especially if you are messaging someone for the first time, it's not very clear who you are messaging (hope you clicked correctly). All this said...

My guess is he wanted to make sure everyone actually received their PMs before the game started. Because our chat system is "problematic". That's it. I wouldn't read too much into it. Just as I wouldn't read too much into his rules in the OP. In short, Vitek seems to be intentionally vague and broad at the same time. In other words, he is leaving all options open. Just because he has left an option open doesn't mean he's necessarily utilizing it.
avatar
bler144: On the bright side, I do think I'm close to deciding whom I'm casting my first serious vote for. But since everyone else is just sitting pat and stalling til the deadline posts, I'm not inclined to rush it.
This I don't understand, why is everyone inclined to wait for a deadline to be posted to vote? Does everyone expect a magic fairy to pop in and start handing out information? Day 1 is little information and that is not going to change. All we are doing now is creating more WIFOM. People just need to get over it and vote so we can move the game along. It's pretty simple, think someone is acting scummy? Vote! If you think there are multiple people who are acting scummy, pick the one who is acting the scummiest and Vote!

I guess for some reason though most seem inclined to wait until there is little time left so we can rush a vote.
avatar
trentonlf: I guess for some reason though most seem inclined to wait until there is little time left so we can rush a vote.
Scummy players don't want to seem over-eager, and Town players are awash with uncertainty and don't want to choose incorrectly. At least that is my interpretation of most games on Day 1. It usually takes a big slip up, real or imagined, or some sort of external factor, such as a deadline, to get anyone moving. Either as an excuse, or as extra motivation.

As for me, I am always bloodthirsty, regardless of alignment, so it is hard for me to relate. ;)
avatar
Krypsyn: Scummy players don't want to seem over-eager, and Town players are awash with uncertainty and don't want to choose incorrectly. At least that is my interpretation of most games on Day 1. It usually takes a big slip up, real or imagined, or some sort of external factor, such as a deadline, to get anyone moving. Either as an excuse, or as extra motivation.

As for me, I am always bloodthirsty, regardless of alignment, so it is hard for me to relate. ;)
I decided a few games back that being uncertain caused me too many problems. If I feel someone is scummy I'll hound them as much as I can. I find it's better than posting a wall of text that amounts to "I'm not sure but maybe"