Open 316 - Hard Boiled (Day Two)

[zMuffinMan]

@Hoopla,

One last thought. Your objection to the Hider claiming who he would hide behind is that scum would be able to take out the VT he hides behind and we'd have one less mislynch/misvig available, right? But if scum take out one of the potential mislynch/misvigs anyway, is this really a problem?

[Amrun]

Yes, because that's two town deaths vs 1.

Not directed at me, but that's my issue.

:/

[Hoopla]

No, because a confirmed clear will go a hell of a long way.

The reward of gaining a confirmed innocent isn't worth the risk of cutting into our total mislynch/misvigs. I'll explain;

In a situation where we don't use the Hider at all (other than being bait), and use the vig to get us three NK's - the only way we lose is if scum take a risk and go for the vig (and succeed), or if we're wrong four times in a row (D1/N1/D2/N2) before hitting scum. We essentially need to nail

all

the townies and leave the the last scum alive. If we get down to the last scum, that scum player needs to outlive EVERY other townie in the game (it should come down to a 2:1 with a confirmed PR at worst). This means we have to gamble on one alignment of a player in the game. What is the point of gambling with the Hider straight up, than in a 50/50 in lylo? We might not even get to that stage where we need to gamble on someone's alignment. Doing it earlier in the game with less information is suboptimal, and doesn't really reap any benefits. It only has risk.

[Hoopla]

@Hoopla,

One last thought. Your objection to the Hider claiming who he would hide behind is that scum would be able to take out the VT he hides behind and we'd have one less mislynch/misvig available, right? But if scum take out one of the potential mislynch/misvigs anyway, is this really a problem?

My analysis factored in the scum nightkill. They only get to kill one player each night. The Hider taking a player with it at night is the additional kill messing up the cycle. We don't want that extra kill.

We should essentially be treating the Hider/Hider Tracker/Det-Psych as named townies, and disregard their abilities.

But losing the utiliiity of the hider and ht also doesn't seem good. :/

This is tempting logic to follow, and using the Hider as NK bait to protect the vig could even feel counter-intuitive to some, but the plan is sound. I think all you need to do is look at the last two Hard Boiled games to see how the PR's in this setup can go wrong. Securing a stable, predictable amount of benefit from the PR's we have (in this instance, using them as NK bait to protect the vig) is a superior option to the boom/bust nature of unplanned/unclaimed actions.

[Amrun]

I played in the last two hard boiled - once as pants on head town that lost, and once as lolvig.

Eeeeh, so, I said I'd give this a shot, so let's do it - hider doesn't hide.

[Hoopla]

I think we should wait and see what everyone else says. I have some pretty good reads though. I'm hoping the obv-townies aren't the PR's.

[Amrun]

Wouldn't that be nice? -.-

We need more posting from other people. I agree.

[andrew94]

okay i understand that plan.
/agree

[Quilford]

Whoops, I forgot I was in this game.

I understand and agree to Hoopla's plan.

[DarkFlashlight]

This was quite a hefty read for anyone who missed it, so slow responses are pretty normal.


Okay, let me try to break down the plan so that I know I understand it, and anyone else who reads this. Steps are written in order they would happen:
1. Everyone goes through and claims either det/psy or

not

det/psy. If there are no CC's we move on.
2. Hider tracker claims [in the same fashion]. If not CC'ed, move on.
3. Hider claims [also in the same fashion].
4. Four of the leftover [potential] 9 claim

not

vig.
5. The other 5 are in the vig pool, 4 Mafia/VT's+the real vig.
6. Lynch as normal, but only out of the 4 who claimed not vig.
7. Vig shoots one of the 4 other people in the vig pool.
Repeat steps 6 and 7.

Also throw in some possible help from the other PR's.


Now this ordeal over Hider is the biggest discussion thing it seems. Hider should definitely hide. Whether or not they say who they're hiding behind is more complicated. I have an idea. The likelihood of scum killing HT while at the same time hider hides with scum seems pretty little, but if you're worried why doesn't the tracker post two names. Who he'll hide with is between those two. Maf won't bother trying to hit him because assuming that both choices are townies, they don't want to be wrong and deplete part of the non-vig tank which doesn't have that many people in it and will assumedly have some scum. They don't want to narrow down the choices against their own odds. Now imagine there's one scum and one VT in the possibilities. This can only backfire if hider goes with the VT. Otherwise, scum will know that they can't kill scum, so they'll go with the VT if either of those two. Now stop and think about that. Hider died, and it might look like he died because he was hiding behind VT and VT was killed by scum, but wait! If scum used their kill on the VT, then HT can confirm that hider actually hide with the other, meaning that he's obviously scum. So this plan succeeds if the two are town, or if it's 50/50 and he hides with the scum, but fails if both are scum, or it's 50/50 and he hides with town (only if scum kill the VT). Also, obviously HT has to be alive for one of those scenarios.


The thing I can't seem to shake about your plan is, what if scum DO slash in the vig pool, and DO kill vig before D3? Then we'd have to dump the vig pool into the non-vig pool and all of our PR's would have been outted, resulting in a normal game, except with all of our PR's outted.

So I'm not necessarily in for this yet.

okay i understand that plan.
/agree

Lame.

Whoops, I forgot I was in this game.

I understand and agree to Hoopla's plan.

Lame.

[neil1113]

Again, I really don't like the idea of the plan. Dark explained why a little better then I did. It's a good plan, but not a sound one. And with one lucky shot, our vig could wind up dying night 1. Then what Hoopla? Also, what if the vig doesn't exactly know of the pool of people who to vig? Just take a lucky guess and cross your fingers?

[Amrun]

Dark: I like this proposed Hider plan best so far, I think.

neil: Why wouldn't the vig know in which pool of people to vig? We can assume the vig can read English.

[neil1113]

Amrun, I didn't say the vig wouldn't know which pool to vig from. I said what if the vig doesn't have a clear choice, out of the people in the pool to vig? Does he just make a random shot, and cross his fingers?

[Amrun]

He uses his best scumhunting abilities, using the pool to narrow his choices. Of course.

[Hoopla]

Again, I really don't like the idea of the plan. Dark explained why a little better then I did. It's a good plan, but not a sound one. And with one lucky shot, our vig could wind up dying night 1. Then what Hoopla? Also, what if the vig doesn't exactly know of the pool of people who to vig? Just take a lucky guess and cross your fingers?

We have two pools - 9 players in the non-PR pool, and 3 players in the claimed PR pool. For all intensive purposes, those three claimed PR's are confirmed town if nobody counterclaimed. It is like a 3-player open masonry. This is an especially threatening prospect for scum, because if they don't get rid of this "masonry" quicker than we can pare down the other pool, then we win - this is the whole premise behind the Friends and Enemies setup. If scum attempt to shoot the vig, it will lose them the game if they miss. They missed an opportunity to remove one of our 3 PR's,

and

helped further town's goal by minimising our pool, which is all we need. We essentially force them to play our game unless they want to gamble.

Having said all of that, it might be more efficient to keep the vig pool slightly larger so scum have less idea about who the vig is. 6 in the vig pool and 3 in the lynch pool (for D1 at least) seems wiser.

[Hoopla]

Amrun, I didn't say the vig wouldn't know which pool to vig from. I said what if the vig doesn't have a clear choice, out of the people in the pool to vig? Does he just make a random shot, and cross his fingers?

Sure. Or we can nominate players from the vig-pool who would make better options, and the vig can go off that advice.

[zMuffinMan]

or if we're wrong four times in a row (D1/N1/D2/N2) before hitting scum.

This is the problem, and this is what I want to think carefully about. And we can't be wrong 4 times

before hitting scum

because come D3, if no scum are dead, it's game over. And this is the same in both your scenario with 9 alive tomorrow and my scenario with possibly 8 alive tomorrow.

The Hider taking a player with it at night is the additional kill messing up the cycle. We don't want that extra kill.

This I disagree with, because if scum are taking out one of the VTs, it means we don't actually need the additional kill in the cycle. I want to run through some scenarios and think about this before I say anything else.

[Hoopla]

The Hider taking a player with it at night is the additional kill messing up the cycle. We don't want that extra kill.

This I disagree with, because if scum are taking out one of the VTs, it means we don't actually need the additional kill in the cycle. I want to run through some scenarios and think about this before I say anything else.

It's not scum taking out the VT. The scum only get one kill each night. If they choose to nightkill the Hider, that's all they get. The Hider electing to hide behind a player is what is causing the extra death. The scum still kill their PR, but the other death is guaranteed. It also runs the risk of the Hider claiming to hide behind a scum, and the mafia killing another PR - meaning we lose two in one night. Yes, we get a scum lynch the next day, but the scum have all our PR's dead by N2 (except the vig).

Both plans are railroaded to get four chances of being wrong

initially

, but once one scum is down, we get two more chances to be wrong under my plan before we need the next one dead, as opposed to one. That's because using the three nightkills as all possible vig scum kills is better than using the three nightkills for two vigs and one Hider-related death, as the Hider related death guarantees a town player's death regardless (and the possibility of a PR). They look similarly beneficial when you project to Day 2, but it is less optimal the deeper you go into the game.

[zMuffinMan]

You're right that if we do hit scum in your scenario, we have a bit more leeway, but the chance of hitting scum by D3 in my scenario is also much higher. Which is why I want to think about it first.

[Hoopla]

You're right that if we do hit scum in your scenario, we have a bit more leeway, but the chance of hitting scum by D3 in my scenario is also much higher. Which is why I want to think about it first.

That's valid - it has a slight benefit for the short-term, but is worse in the long. I think if we aren't able to hit 1 scum from 4 tries with 3/8, 3/7, 3/6, 3/5 odds we deserve to lose. I don't see how we wouldn't hit scum at least once with those odds - especially when you factor in reads.

[zMuffinMan]

I don't think it's as simple as looking at odds and 'reads'. You're discounting how scum would interact based on knowing a mass claim is likely at this point, and knowing they may have to bus their scum buddies. An example of this is you; I assume if you're a VT, you'll be nominating yourself for the non-vig pool, right? But just because of play so far, you'll be much less likely to be lynched than other players. You can say, "well, you can lynch me if you want", but you know that sort of attitude makes it less likely you'll be lynched. Not to mention if you are scum, your buddies probably knew you'd be doing this, and any reads based on the game so far aren't really reliable.

I'm being really, really paranoid here, but this should explain why I'm not liking play based solely on odds.

To be honest, I'd much rather play the game as is and consider mass claiming again further down the line.

[Hoopla]

I don't think it's as simple as looking at odds and 'reads'. You're discounting how scum would interact based on knowing a mass claim is likely at this point, and knowing they may have to bus their scum buddies. An example of this is you; I assume if you're a VT, you'll be nominating yourself for the non-vig pool, right?

It absolutely is. Playing independent of a coherent plan is just as much leaving it to chance, if not more (I believe a lot more). Playing the game out normally, we have to gamble on some players being town throughout the game - we cannot kill everyone. I've devised a plan that enables us to kill more than usual and guarantees a set, predictable return from the power roles, as opposed to leaving it to chance that they don't kill each other.

If I'm a VT, I'm not sure what I would nominate myself for. I don't think I need to commit to any stance right now.

Not to mention if you are scum, your buddies probably knew you'd be doing this, and any reads based on the game so far aren't really reliable.

If I'm scum, I wouldn't lock us into a suboptimal play. I believe this game is scum-sided when played straight, and breakable for town when a plan is formed. That is why I chose to pick this game - because I had good odds either way regardless of what role I was given. I just wouldn't have organised a plan if I were scum.

[zMuffinMan]

we have to gamble on some players being town throughout the game

You call it gambling, I call it reading.

Your plan doesn't eliminate gambling in any way, though. There's still a gamble.

If I'm scum, I wouldn't lock us into a suboptimal play.

You call it sub-optimal, but it really isn't. You're nullifying all the town PRs except for the vig (essentially they'd all just be confirmed townies) and pretty much ensuring you get to at least LyLo because you probably wouldn't be vigged or lynched over other candidates. You'd be removing most of the PR luck elements that could potentially destroy scum before D2.

I don't actually believe this is a scum-sided game when played straight. Although it really comes down to town PRs playing well, and in past hard-boiled games I've looked at, town have misplayed or scum has been a bit lucky (e.g. in one game, the last remaining scum took out the tracker on the night he tracked the last remaining scum).

All this said, I am a sucker and I do believe you're town, but I really don't like that you said that you must be town because you wouldn't suggest this if you were scum.

[DarkFlashlight]

I'll go through with this plan we split the non-vigs into 3 and vigs into 6. That way, even if Maf still try to take out vig, by N3 with two vig kills and two Maf kills, they still only have a 50/50 chance of hitting him.


So I take it you two didn't like my hider idea?

[zMuffinMan]

Actually, 3 and 6 is a good idea. A really good idea. Not because it hides the vig, but because it makes mafia's decision really, really hard. 1/3 chance of getting lynched if they nominate themselves, and 3/5 chance of getting vigged if none of them nominate themselves. could also get behind a hider using his ability in this scenario.

I didn't mind your hider idea, but we were talking about it from the perspective of numbers, and also there's a possibility that any two players the hider chooses could actually both be mafia (not an extremely likely possibility, but still a possibility none-the-less).