Open 199 - Friends, Enemies, and Enemies - OVER

[evilsnail]

Masons will list their partner, as well as a random third person. A Mason kill at night, or even a Mason lynch, will cause a counter-claim between the two "trusted" players -- especially if one of them is non-town. So, this puts us in the same situation we're already in. Our position would be no better than it is now. Neither of the "trusted" players can be fully trusted, and the town will be forced to lynch one of them.

This is a strong chance that both of the trusted players will be townies, though. And, if someone fake-claims Mason, we'll soon find out who the real mason was. Honestly, narrowing down scum to two people is good for town. It increases our chances significantly.

That said, I'm not sure there's necessarily a point to this. Breadcrumbing is much more effective and masons can tell each other how they breadcrumbed during the night. That way, we can be pretty sure all the masons have info to back up their claim.

We should not NL today because of mavsfan. If we can get him modkilled, that's awesome, because it's like having an extra lynch. In no way does that motivate no lynching. Getting mavsfan modkilled means we can get two lynches with one night phase instead of two lynches with two night phases.

Pretty sure CMAR is the way to go today.

Vote: CryMeARiver

[yawetag]

We should not NL today because of mavsfan. If we can get him modkilled, that's awesome, because it's like having an extra lynch. In no way does that motivate no lynching. Getting mavsfan modkilled means we can get two lynches with one night phase instead of two lynches with two night phases.

What's your thoughts behind the chance of losing two townies: our lynch target and mavs?

[evilsnail]

We have that chance if we lynch twice as well. Except then we have an extra night phase tacked on, which means we lose.

Besides, just randomly, there's a 77% chance scum is either our lynch target, mavs or hit on a cross-kill after.

Optimal play is to let mavs get modkilled and lynch anyway. You can do the math yourself.

[evilsnail]

It's perhaps instructive to compare the worst-case scenarios for both.

Lynch + modkill:
Day 2 - Lynch player X and get mavs modkilled. Both turn up town.
Night 2 - Two townies NKed.
Day 3 - LyLo. We lynch player Y. Y is town.
Night 4 - Remaining mafia member killed and one townie. We lose.

No Lynch + modkill/lynching mavs:
Day 2 - mavs modkilled. Turns up town.
Night 2 - Two townies NKed.
Day 3 - Lynch player X. Turns up town.
Night 4 - One townie and the remaining mafia member killed. We lose.

Same outcome, but on the first scenario we have one extra shot at avoiding this result, because we get to lynch player Y.

Look at it this way. On the first scenario, there are two townies left, one mafia member and two werewolves on Day 3. On the second scenario, we have the same group, but in Night 4. It's obviously better for us for it be to Day, in the same way that a Day start is better for town than a Night start.

[xRECKONERx]

Hmm, you're right.

Vote: CryMeARiver

[yawetag]

You can do the math yourself.

I did.

Same outcome, but on the first scenario we have one extra shot at avoiding this result, because we get to lynch player Y.

Try again.

Here are the numbers assuming only townies are killed and lynched (in town/maf/wolves order) if we lynch someone other than mavs:
After D2 - 4/1/2
After N2 - 2/1/2
After D3 - 1/1/2

The same with a no lynch today:
After D2 - 5/1/2
After N2 - 3/1/2
After D3 - 2/1/2

In the first scenario, we lose N3. There's no way to win:
1. Maf kills wolf - wolf kills town: 0/1/1. Town loses.
2. Maf kills wolf - wolf kills maf: 1/0/1. Town loses.
3. Maf kills town - wolf kills town (same target): 0/1/2. Town loses.
4. Maf kills town - wolf kills maf: 0/0/2. Town loses.

In the second, we have a chance:
1. Maf kills wolf - wolf kills town: 1/1/1. Town loses.
2. Maf kills wolf - wolf kills maf: 2/0/1. Town can win.
3. Maf kills town - wolf kills town (same target): 1/1/2. Town loses.
4. Maf kills town - wolf kills town (different target): 0/1/2. Town loses.
4. Maf kills town - wolf kills maf: 1/0/2. Town loses.

Obviously, this hinges on town never lynching scum, but it's worse-case scenario.

Barring finding a true scum, we have two options today:
1. No lynch and get mavs modkilled.
2. Lynch mavs.

Either way, it would be A LOT better if mavs turned scum.

[yawetag]

Besides, just randomly, there's a 77% chance scum is either our lynch target, mavs or hit on a cross-kill after.

I'd love to see your math here. Make sure you include whether or not mavs is scum, as that has A LOT to do with the calculation.

Without counting the cavs modkill, we have a 33% chance of lynching scum (6 townies, 3 scum). The other 66% of the time, mafia has a 28% chance (2 of 7) of finding a wolf, and wolves have a 17% chance (1 of 6) of finding mafia.

Using a simple graph, I got the following possibilities, assuming a townie lynch today, no mavs modkill:
Maf kills wolf - Wolf kills maf: 2
Maf kills wolf - Wolf kills town: 10
Maf kills town - Wolf kills maf: 5
Maf kills town - Wolf kills town (same target): 5
Maf kills town - Wolf kills town (different target): 20

Of 42 possibilities, there's only 2 where a cross-kill will occur. That means, of the 66% of the time we miss scum today, only 5% (2/42) of the time will we get a cross-kill.

My calculations show that a scum will die only about 40% of time. Again, this doesn't count whether mavs is scum or not.

I hate to throw out numbers -- it confuses people. However, I have to point out faulty math when it's used.

[yawetag]

BWOP: I've said it from the beginning of the day, and I'll keep repeating it. Either no lynch or lynching mavs is the smart move for the day.

Vote: mavsfan41

[evilsnail]

Your comparisons are off.

You shouldn't compare both scenarios at the same point (D3 in your scenario), because at this point we have had an extra lynch on the modkill+lynch scenario. To make this work, you're assuming an extra mislynch on that scenario, on D3. If you assume that on the lynch mavs scenario, we also autolose. So we are better off on the modkill+lynch scenario.

My math was quite basic:
Chance of mavs not being scum - 2/3
Chance of not lynching scum, assuming mavs town - 5/8
Chance of mafia not hitting scum, assuming mavs and lynch town - 2/3
Chance of werewolves not hitting scum, assuming mavs and lynch town - 4/5

2/3*5/8*2/3*4/5 = roughly 22%. Ergo, there's a 77% chance of a scenario in which a scum dies.

Please point out the problem in this calculation. I'm not a math genius, so maybe I'm overlooking something here, but I think this holds up.

[yawetag]

2/3*5/8*2/3*4/5 = roughly 22%. Ergo, there's a 77% chance of a scenario in which a scum dies.

Please point out the problem in this calculation. I'm not a math genius, so maybe I'm overlooking something here, but I think this holds up.

You're taking several scenarios and multiplying them together. That's not correct math.

Right now, we have a 33% chance of finding scum; ergo, 66% of the time, my math goes into effect. Everything that happens in that 66% of the time is ADDED to the 33% from D2, but you also have to make sure you multiply any of the N2 actions by 66%. I can give you examples, but it will only fill the thread with useless math. Remind me at the end and I'll show you how it works.

I've said it several times before, but I feel it needs to be repeated: Lynching anyone other than mavs is a bad idea. We risk losing two townies, and that cannot ever be a good idea.

[yawetag]

I meant for both of these to be in the same post. I apologize for the frequent EBWOPs.

Your comparisons are off.

You shouldn't compare both scenarios at the same point (D3 in your scenario), because at this point we have had an extra lynch on the modkill+lynch scenario.

No, it's a very fair place to compare both scenarios. It's the point of losing or being in LYLO. Yes, it's worst-case, but you have to account for worst-case or you lose. That's what MYLO and LYLO hinge on.

[xRECKONERx]

STOP MATHING ME TO DEATH YOU ASSHOLES

[evilsnail]

I wasn't taking several scenarios. I was calculating the odds of the scenario in which no scum dies. To do that requires multiplying the odds of the component parts.

But regardless, my point about your comparison stands. If you think about this carefully, you'll see that I'm right.

[evilsnail]

I meant for both of these to be in the same post. I apologize for the frequent EBWOPs.

Your comparisons are off.

You shouldn't compare both scenarios at the same point (D3 in your scenario), because at this point we have had an extra lynch on the modkill+lynch scenario.

No, it's a very fair place to compare both scenarios. It's the point of losing or being in LYLO. Yes, it's worst-case, but you have to account for worst-case or you lose. That's what MYLO and LYLO hinge on.

You're not getting my point. It's fair to compare both scenarios, but it's meaningless to do so on the same days, because the scenarios entail moving at different speeds.

You're assuming a mislynch on your modkill+lynch mavs scenario that's not in your other scenario. Therefore, the scenarios are imbalanced. If you assume the same mislynch in the other scenario, on D4 with 2/0/1, we also autolose.

[xRECKONERx]

*snore*

You two have put me to sleep, gg

[evilsnail]

I checked my math and it's fine. Yawetag, could you calculate for me the odds of not hitting scum at all on the modkill+lynch scenario? I think you'll find you get the same answer.

[yawetag]

You're assuming a mislynch on your modkill+lynch mavs scenario that's not in your other scenario. Therefore, the scenarios are imbalanced. If you assume the same mislynch in the other scenario, on D4 with 2/0/1, we also autolose.

The first scenario is a town lynch AND mavs modkill showing town. The second scenario I gave is a no-lynch or lynching mavs ONLY and mavs showing town -- no other lynch given.

That's the reason for the difference in numbers. As you'll see, taking the worst-case scenario, lynching mavs today is the only option UNLESS we can find scum. Any other lynch puts us in a very bad situation.

I know the math is confusing people, and I wouldn't normally post it. However, I had to point out evil's mistakes.

I checked my math and it's fine. Yawetag, could you calculate for me the odds of not hitting scum at all on the modkill+lynch scenario? I think you'll find you get the same answer.

Just for today? Sure, and I'll try to make it (somewhat) easy. If we lynch someone other than mavs and mavs gets modkilled, we have the following situations (a total of 72 different ways to happen [1 mafia, 2 wolves, 6 townies]):
1. two scums die - 6 times (8.3%)
2. one scum dies - 36 times (50%)
3. no scum dies - 30 times (41.7%)

Taking those numbers, we will hit at least one scum 58.3% of the time. Not shabby, but significantly less than the 77% you gave.

[Shrinehme]

The wasn't mentioned in the case against SK/CMAR that I just read over and agree with, but Flareonage's vote on him early in that Day Phase...:

Unvote

VOTE: SaintKerrigan

Dana brought up valid points thus your vote is pointless

... his dramatic response...:

SERIOUSLY?!? Doesn't it bother you that DKU took seriously what xReck put forth as a joke?

... and Flareonage's lack of committal with the vote...:

I remember now. Why did he do that?

Unvote

VOTE: xReckonerx

Why were you trying so hard to put suspicion on me?

Looks very VERY much like distancing.


Needless to say, I like this wagon, and I support having a pseudo- "double lynch" with mavsfan's modkill.

Vote: CryMeARiver

.

[Papa Zito]

The oppression of their concrete prison is obviously starting to get to them, because suddenly the discussion shifts away from important things like, say, who's trying to kill them, and into a BFF roundtable. Once math is brought out (prompting the jock to fall asleep) things really start to get strange.

In the midst of all this, suddenly there are a crucial number of fingers pointing at CryMeARiver.

Third Votecount of Day 2

1. yawetag (0):
2. The Tracker (0):
3. cdubs (0):
5. CryMeARiver (4): danakillsu, evilsnail, xRECKONERx, Shrinehme
6. danakillsu (0):
8. mavsfan41 (1): yawetag
9. xRECKONERx (0):
10. Shrinehme (0):
11. evilsnail (0):


Players not voting: The Tracker, cdubs, mavsfan41, CryMeARiver

With 9 players alive, it takes 5 votes to lynch.
The current wagon leader is CryMeARiver, at

L-1

Deadline: February 15 @ 11:59PM CST

[yawetag]

Before someone lynches, PLEASE think about the dire situation we could be in tomorrow if CMaR and mavs are both townies.

[evilsnail]

Not true, you do have an extra lynch. It's right here:

Here are the numbers assuming only townies are killed and lynched (in town/maf/wolves order) if we lynch someone other than mavs:
After D2 - 4/1/2
After N2 - 2/1/2
After D3 - 1/1/2

D3 in this scenario has a townie lynch. This is not mavs or today's lynch (since these take place on D2), so it's an extra mislynch.

Maybe it'll help you to think of it this way. Suppose that instead of a modkill, the punishment for lurking was: a night phase is skipped and the lurker is the next day's autolynch. This is equivalent to a modkill, if you think about it.

That way, the worst-case scenarios are as follows:

modkill+lynch:
D2: lynch player X = town
N2: skipped, because of lurker
D3: autolynch mavs = town
N3: two townies killed
D4: lynch player Y = town/mafia
N4: town/mafia killed, werewolves win

lynch mavs:
D2: lynch mavs = town
N2: two townies killed
D3: lynch player X = town
N3: one townie, one mafia killed, town loses

In the first scenario, we last longer and get an extra lynch. I don't see why this is hard to comprehend for you.

You do see that having multiple lynches is good for town, right? This is why there are 4 scum in a Nightless setup, for instance.

Your math does not add up, btw. You don't even have a situation listed in which all three scum die (which can happen if one of CMAR/mavs is a werewolf and scum cross-kill each other). Clearly, you're doing something wrong.

[yawetag]

In the first scenario, we last longer and get an extra lynch. I don't see why this is hard to comprehend for you.

I'm done arguing about the scenarios. My situations had townies killed at all times; situation one had two lynches today (a target and a modkill), and situation two had one lynch today (a modkill). There's no other tricks about it. Let's agree to disagree and go back to what we're supposed to be doing... finding scum. Right now, my biggest FoS is on CMaR. I didn't like SK's actions during D1, and CMaR hasn't done much to defend the spot today.

You do see that having multiple lynches is good for town, right? This is why there are 4 scum in a Nightless setup, for instance.

It's obvious that I disagree with you, and I always will. Lynching today is not a good idea at all. If we lose two townies (which will happen 40% of the time), the town loses. If we no lynch (or even lynch mavs), we have a chance to win if he turns up townie.

Obviously, NONE of this matters if mavs is scum.

Your math does not add up, btw. You don't even have a situation listed in which all three scum die (which can happen if one of CMAR/mavs is a werewolf and scum cross-kill each other). Clearly, you're doing something wrong.

No, you're looking at it wrong. You wanted to know the odds of hitting scum with a lynch and a modkill. This has NOTHING to do with night actions, only the actions of the town during the day and the modkill that will presumably happen before night begins. In that case, there are only three possibilities: scum/scum, scum/town, town/town. Nothing else.

[yawetag]

In the first scenario, we last longer and get an extra lynch. I don't see why this is hard to comprehend for you.

I can't ignore this. You are wrong, plain and simple.

Situation #1.
On Day 2, we are at 6/1/2. We lynch someone and mavs is modkilled. Both turn up as town. We lose two townies.
On Night 2, we are now 4/1/2. Both mafia and werewolves kill a townie. We lose two townies.
On Day 3, we are now 2/1/2. We lynch a townie.
On Night 3, we are now 1/1/2. There is no way for the town to win.

Situation #2:
On Day 2, we are at 6/1/2. We no lynch and mavs is modkilled (or we lynch mavs -- it makes no difference). mavs turns up town. We lose one townie.
On Night 2, we are now 5/1/2. Both mafia and werewolves kill a townie. We lose two townies.
On Day 3, we are now 3/1/2. We lynch a townie.
On Night 3, we are now 2/1/2. Town has a chance to win if mafia and werewolves attack each other.

Where did I mess up? Yes, every step assumes a townie is killed, but you HAVE to look at worst-case scenario when determining an action. If you didn't, you shoot yourself in the foot.

[evilsnail]

You are wrong because the Day 3 townie in situation #2 is lynched on Day 2 in situation #1. There is an extra townie lynch, the Day 3 lynch, in situation #1. If you add this to situation #2, on Day 4, we also autolose. Situation #2 is then worse, because we have a chance of not even getting to the point where we can get that lynch.

And it is important to have this argument, because there's a chance someone might believe you and it would be bad for town not to take advantage of this opportunity.

No, you're looking at it wrong. You wanted to know the odds of hitting scum with a lynch and a modkill. This has NOTHING to do with night actions, only the actions of the town during the day and the modkill that will presumably happen before night begins. In that case, there are only three possibilities: scum/scum, scum/town, town/town. Nothing else.

I wanted to know the odds of hitting scum with a lynch, modkill and cross-kills. This is then where our calculations differ. I think you'll see that if you include cross-kills, you get that 22% I got. At which point you'll also (hopefully) finally realise that my math was not faulty, as you claimed.

[yawetag]

You are wrong because the Day 3 townie in situation #2 is lynched on Day 2 in situation #1.

This makes absolutely no sense. Your argument is that I'm counting a day 3 lynch in situation #2. Of course I am... we will have to lynch someone on day 3 in BOTH scenarios. There's also a lynch on day 3 in situation #1. Adding an extra kill on day 2 doesn't eliminate a day 3 lynch.

Step back for a second and think about the whole situation. Your contention is that by eliminating an extra player from the field (who we both are making a townie for worst-case scenario), we gain extra time. This makes no sense, no matter how you dice it. Each day, we're lynching one person, except for today, when you're arguing that we should lynch TWO. There is no situation where lynching two townies is better than lynching one.

And it is important to have this argument, because there's a chance someone might believe you and it would be bad for town not to take advantage of this opportunity.

So you think you're right, but your hope is that someone believes me and takes advantage of it? Do you normally double-talk? I agree that the argument is nice, but it's getting to the point of hurting town more than helping. You've given your opinion based on your math, and I've given mine. Unfortunately, we've had very little discussion from other members and we've got a target on L-1. The fact that the hammer hasn't dropped yet tells me we're sniffing up a good tree, but I don't think it's in the best interest for the town.

I wanted to know the odds of hitting scum with a lynch, modkill and cross-kills. This is then where our calculations differ. I think you'll see that if you include cross-kills, you get that 22% I got. At which point you'll also (hopefully) finally realise that my math was not faulty, as you claimed.

Wow. This calculation is a lot more complex than what you posted earlier. You have to go into the percentages of how many scum could be killed by lynching, which effects the percentages of how many scum targets there are for scum to hit. With that said, and without actually calculating it, I can agree that the chance of at least one scum dying between now and the start of tomorrow would be close to that. Basically, you have a (6/9)*(5/8)*(4/7)*(3/6) chance of only townie kills with the lynch, modkill, and two night kills, which comes out to approx 12%; that makes 88% chance of hitting at least one scum in those four kills. However, you're taking the chance of up to 3 townies dying in the process -- that's a horrible situation for the town.