Oh god, no,n ot the math. That was my downfall last game! 
Haha… I LOVE the math, as I’m sure most anyone here will tell you who has played with me before. I’m making a concerted effort to cut back, but it won’t go away. 
Seriously? Because I found the pronoun slip silly, I’m not allowed to be skeptical of his role claim? I haven’t even voted for him yet, and I certainly haven’t been “pushing” for his head either. You are way exaggerating my statements.
The pronoun “slip” and the role-claim are two completely different things. Why is it strange to find one unconvincing and the other worthy of special note? I am judging Pleonast by his actions, not him as a person, since I don’t know anything about him (in the context of this game).
That being said, I’ve been mulling over the role-claim, and have decided that I’m going to believe it for the time being. I noted previously that I thought it was a reckless thing for a Townie to do, but upon further reflection I feel that generally, reckless behavior is a sign of Town rather than Scum (GENERALLY, sez I).
Yes, I’m aware I haven’t took a concrete stand on who I find the most suspicious quite yet. I need some more time to re-read the thread, and Mon-Thurs are generally a whirlwind for me in terms of work. I’ll cast a vote before the weekend is up.
Blaster the math you just presented does nothing to clarify but rather further obscures the issues before us. If you can’t bring it down to a level that is understandable don’t bother, it’s just a distraction. Also, you seem to be assuming that the Boss had a 100% chance to successfully recruit last night. You know this is not true.
FOS BlasterMaster
'm fairly sure I voted for **Pleo ** in blue and bold, no less. 
Today is the last day of the month, and I am swamped! I will not be able to get back to the game until Sunday at the earliest. I can read along in snippets, but not do much contribution until then.
Huh? What about my post is not understandable? You have to realize, it IS understandable to me, and that’s how I think. If you have a specific question about my math or my reasoning, state it, but simply saying it is not understandable and FOSing me over it is counter-productive.
My point was not about whether the recruit succeeded or not, simply whether a recruit was attempted or not. I stated my reasoning why I think a recruitment is the more likely scenario in multiple posts, I then used that post to make the point that it’s important for us to use our estimates so we can actually have usable information.
Look at it this way. To simplifiy matters, let’s say we have two possibilities each Night, if we do nothing to eliminate or reduce the weight of specific possibilities, then our ability to determine our situation decreases EXPONENTIALLY. If we treat both situations as equally likely, we have no certainty about whether a recruit was attempted or not, and so we don’t know whether there’s one or two scum, we don’t know whether the Vig attacked or not… we don’t know ANYTHING.
I’m making a concerted effort to reason why I estimate the probabilities for what I think happened last Night, and I provided it as counter reasoning to this obsession over always keeping EVERY possibility in mind. If we don’t make reasoned hypotheses, we’ll be mired under tons of data and still have no idea what to do with it.
Further, I am aware that the Boss did not have a 100% chance to recruit last Night. If you’d like, I can put together a formula that would allow us to determine a more precise percentage. Oh wait, math is not understanable, and making educated guesses get’s me FOSed… :rolleyes:
I’m going to vote Pleonast for the time being.
Now that I’m no longer in the lead, I’ll
Unvote Pleonast
I don’t want to lynch a claimed power unless there’s no alternative (or lynching *me *is the alternative, so expect me to switch back if it’s in my own interest). On reflection, I’m not sure siccing the Vigilante on him would be the right call or not. I don’t trust Pleonast and I don’t trust the Vigilante to choose well without the chance of the victim mounting a defense or a role claim. So to me, losing the Vig is no big deal. But I suppose the “unrecruitable” factor makes it a more important role in this variant…
Anyway, that’s the Vigilante’s call. Not necessarily mine.
STill being early in the game, I don’t yet have anything I’m comfortable in investing a vote on, but a few little FOS’es.
Menocchio, pushing the vig to either kill a confirmed townie and lose their powers to test a role claim is not, IMO, a good use (esp this early). If Pleo is lying, we’ll either come to that through actions and posts, or the mafia will test (unless he’s the boss).
FOS Menocchio
bufftabby I seem to recall you wanted to target lurkers, but jumped right on the Pleo wagon when it got rolling. Doesn’t seem like a good move there , so
FOS bufftabby
For the record, I agree with HNC on the first part of the whole pronoun game; I pointed it out as a silly little thing, and didn’t think much of it really. The role claim, however, is strange, and something I want to think about.
Understandable, but: last time I took a look at the player list, I recalled seeing posts from everyone. So: my research on that is still pending. It’s a bit hard for me to do extensive research on this here Blackberry o’mine. In fact, I’d like to throw this out right now: anyone who would like to take me to task for non-thorough research, a lack of a post-specific link, etc, feel free, but be aware that my primary mode of access is through my lil ol Blackberry, which means I’ve only got one window I can open, and can only see 20 posts at a time. So I’m doing my best on research and such, but I’ve really gotta set aside time to take notes etc to do it thoroughly, which I plan to do this weekend. So don’t worry, Koldanar, I’ll be reporting back my lurker research results once work is not interfering this weekend. 
I wasn’t clear and I’m in a hurry so just one point.
I FOS’d you for failing to take into account recruitment is not 100% successful, not because of the math. You say you did take it into account, I’ll go back and reread, I must have missed it. FTR I am not anti-math (very far from it), I am anti-math that looks like hand-waving from the bleacher seats. I was also suggesting, probably not very diplomatically, that you consider your audience. All those who understood that little mathy diversion raise your hands. Just kidding.
And yes please do come up with a formula for determining the probability of recruitment failure. I think that would enlightening.
This is why I specifically mentioned the option now, so it can be discussed. Obviously, it’s still up to the Vigilante if he believes Pleo’s claim or not and if it’s worth the risk. Hopefully, he’ll make a wise decision.
Huh? How does not taking into account the possibility of recruitment failure mean he’s scum? I would expect scum to be very aware of their limitations. Forgetting a detail about a role tends to imply that player does not have the role.
Well, the point was, yes, I’m aware that there is a failure possibility, and yes, I’m aware it’s higher than a kill possibility. The point of the math was, given my estimate, based on my arguments for why he would have recruited, I can gauge about how much information we have. Whether or not the recruit succeeded isn’t AS relevant as to whether or not a recruit was attempted. Besides, we can safely come up with a reasonable estimate of the success of a recruit, and it’s likely much higher than 50%, thus it would benefit us, from an information theory perspective, to consider that, if a recruit was attempted, it was most likely successful.
I’m willing to come up with a formula that will calculate a successful recruit probability, given some base assumptions… give me a few minutes, as I’ll have to reread on the roles to make sure I’m accurate.
Okay, here’s my crack at it:
let n = the number of players
s = number of scum
D = the Doctor’s protectee (either 1 if someone is protected, or 0 if the Doc is dead or recruited)
V = Vig (1 if the Vig is alive, 0 if he’s dead)
P = number of priests
M = number of masons
C = probabilitiy that it was a Capo recruitment
B = Bishop (1 if alive, 0 if he’s dead)
Thus:
f® = (D + V + P + .5M + (1- p©B)) / (n - s)
Now last Night, we knew n = 24, s = 1, D = 1, V = 1, P >= 1 (I’ll guess 3), M >= 2 (I’ll guess 3), C is likely small (I’ll, in fact, guess 0), and B = 1. So…
f® = (1 + 1 + 3 + 1.5 + 1) / 23 = 0.326
For a given night, we can estimate s by taking the previous estimate, subtracting the one we lynched (if we did), and multiplying the the success rate (1 - f®) by the probability that a recruit was attempted (p®)
We can then use this expected value to get a reasonable guess of how many scum there are, which will be more useful as more days go on. The expected value doesn’t drop below 1.5 unless we think a recruit was attempted last Night with less than about 75% certainty. If I estimate a 90% chance that a recruit was attempted last Night (which I feel I’ve reasonably justified), then our expected value for scum is 1.61. IOW, even including the failure rate, it is still likely that there is a second scum out there.
Hey, NAF. I think that the main difference is the fact that I’m the only one who has mentioned “breadcrumbs” thus far! This used to be a big deal–I’m wondering if it started becoming more trouble than it was worth. But I’m very concerned about losing valuable info. In my opinion, speculation is one thing, but hard-and-fast facts are gold.
Real-life moment: In nursing school, there’s a HUGE emphasis on subjective versus objective data. So I have it on the brain right now. Both kinds of data are necessary, but there’s a big difference between the two. And we need to pick out our facts very carefully so as not to miss anything.
The subjective stuff alone seems to be fertile for speculation: subtext, subtle change in posting style, and assumptions that are couched in probability math.
The facts that are, in some way or another, available to the town, so far, are records of who voted for whom, and the following:
One and Only Wanderers is dead.
One and Only Wanderers was a Townsperson.
What the Vigilante knows about his/her night activities.
What the Detective knows about whomever he/she investigated.
What the Police Chief/Beat Cop(s) know about whomever they investigated.
What the Doctor knows about his/her night activities.
What the Bishop/Priest(s) know about their night activities.
And the following are facts that are compromised by the possibility of a recruitment last night:
What the Masons know about each other.
Only two of these facts are available to all town at this point. The rest are not. The facts are going to multiply rapidly, but we still won’t know more than a small fraction of them. I say that we need to find a way to make sure we lose as few as possible each time someone with undisclosed information dies. I would add “or before they’re made”, but that doesn’t seem to be anything we can do anything about. We won’t know when we’ve lost them to the Mafia unless we’re told by a witness, and even then, we won’t get the info, we’ll just kill them.
So in short, I think that power roles need to have a plan for this. I don’t want to direct your powers, and I don’t want to tell you what you should do to communicate with us, but please consider this!
Well, I’m about to be away for a while - till Sunday evening or so.
I’m going to upgrade my FoS on Pleonast to a vote.
Vote: Pleonast
It’s a weak vote, and I’ll take it off if I see a better case made, but you screwed up early, then made a very convenient role claim early. I know that if you were assigned the priest role, all you could do was claim it. But I feel you’re hiding something, I’m not sure what, but it feels like a discussion killer.
I’ll be back later!
Ok I have a bit more time now. Thanks for including you equation for determining the chance of recruitment failure. That was the same equation I used (behind the scenes) in post 455. Sorry about not using 3 significant figures, I thought it was overkill for the venue. I thought about rounding up to 22% on my first calculation but didn’t want to get called on inflating my numbers.
I guess we will have to agree to disagree. No matter how much you massage the data you are still utilizing your guess at the probability that the Boss would recruit vs kill. I am certainly not as logical as you because with the same success at killing vs failure in recruiting levels I would have killed last night if I was the Boss. To me it would come down to an almost sure thing (killing) vs up to a 36% chance of failure (recruitment). When added to the fact that no real information was revealed about any of us during the discussion Yesterday, for me it would be a no brainer. That just reflects on my risk taking behaviour. (Probably why I am not a millionaire yet.)
I still don’t think we have enough data to determine what happened last night.
un-FOS BlasterMaster
Omitting an important detail when formulating a hypothesis is not something one expects at this level of analysis.