Hockey Monkey!
Borogrove
This is supposed to be a happy occasion. Let's not bicker over who killed who.
Posts: 371
[ Exalt | Smite ]
Karma:
|
Day Two
Sept 25, 2007 20:53:34 GMT -5
Post by Hockey Monkey! on Sept 25, 2007 20:53:34 GMT -5
It's just an observation, and yes I do think how things are worded is relevant. How things are worded is relevant, that's true. However, in order for Hockey's theory to hold any water, drainbead using the word "you" to refer to her night killed action going trough would have to have happened after drainbead was already dead. That was not my point. I'll see if I can be a little clearer. I was hypothesizing that someone who is scum could accidentally take possession of their role with their wording in a hastily made post. All I did was pick up on wording that looked suspicious to me because that was not how I would have worded the question. If this doesn't make sense to anyone else, or if others don't think it's relevant, then I'll make no further mention of it. All I have to go on are these posts so when I see something that pings me the wrong way I'm going to point it out.
|
|
|
Day Two
Sept 25, 2007 21:08:29 GMT -5
Post by ComeToTheDarkSideWeHaveCookies on Sept 25, 2007 21:08:29 GMT -5
ToDay is pretty much an antithesis of yesterDay, so much so that y'all are sounding entirely too reasonable and even-keeled.
I really don't see the fuss about sinjin, and that is the strongest inclination I have at this point, as much as I wish it were otherwise. I'll be looking closely at those who currently suspect her. If any of you wish to save me the trouble of asking for specific clarification as to why exactly you find her suspicious, feel free to step onto the soap box and pontificate.
|
|
|
Day Two
Sept 25, 2007 21:16:26 GMT -5
Post by Boozahol Squid, P.I. on Sept 25, 2007 21:16:26 GMT -5
ToDay is pretty much an antithesis of yesterDay, so much so that y'all are sounding entirely too reasonable and even-keeled. I really don't see the fuss about sinjin, and that is the strongest inclination I have at this point, as much as I wish it were otherwise. I'll be looking closely at those who currently suspect her. If any of you wish to save me the trouble of asking for specific clarification as to why exactly you find her suspicious, feel free to step onto the soap box and pontificate. You can probably just read post 174 of today to get a reasonable synopsis of my suspicions.
|
|
Parzival
Mome Rath
Let's all strive to do our best today![on:forgot to log out][of:forgot to log in]
Posts: 201
[ Exalt | Smite ]
Karma:
|
Day Two
Sept 25, 2007 22:20:25 GMT -5
Post by Parzival on Sept 25, 2007 22:20:25 GMT -5
I just thought of something... panamajack, would you be able to provide more details about the nature of your attack at all? I'm thinking there's not much to read into the death descriptions. Note that even Kat mentioned the grenade for Diggit may only have been to satisfy the explosion-loving contingent. Without quoting it, I can tell you that the description of my attack is vague enough that I don't think anything can be determined there either. It doesn't even mention if a weapon was involved, but it sounds like close quarters. Also, it's clear both that their intent was to kill and that I was unable to get a good look at the person. Honestly, I don't know how much more can be determined from this until future nights. I put it out there just for that (I realize it lets the scum know something new * as well but it helps the crew more) and in my opinion it doesn't affect what I think is the most useful part of 'death analysis' - determining who the scum killed/targeted. As I've already said, I find it unlikely they targeted me since I survived. Roosh, I also saw your point with respect to my thoughts on Diomedes, and believe me, I went through some of that myself. So I realize even if the scum had targeted me, it could have been for any or no reason. There's a lot of speculation that could be going on, but continuing on it just leads to more distraction. *That there's another killing group; prior to my mentioning it they did not know this either.
|
|
|
Day Two
Sept 25, 2007 23:09:28 GMT -5
Post by Boozahol Squid, P.I. on Sept 25, 2007 23:09:28 GMT -5
There's definitely the possibility you were targeted by a doc in addition to a killing role. We can't rule out that you were the alliance hit, nor can we count on the fact that the third killer is only somewhat effective.
|
|
Blaster Master
Mome Rath
The player formerly know as BLAM!
Now 34.788% less repellant to Sharks! :( [on:I WANT TO DIE!][of:I WANT TO LIVE!]
Posts: 0
[ Exalt | Smite ]
Karma:
|
Day Two
Sept 25, 2007 23:53:28 GMT -5
Post by Blaster Master on Sept 25, 2007 23:53:28 GMT -5
OK, let's just start with this: Spaceman Blaster, there is no way I can respond to your argument on the same level at which it was made, because I just plumb don't follow some of it and am unlikely ever to do so (I'm an English major, for heaven't sake). But I think I understand the gestalt of your argument, so I'm going to respond to that. If I'm getting even that wrong, though, feel free to correct me. Here's the thing: I'm pretty sure I disagree with you on this subject. There are two elements to this - the fact of my disagreement, and the question of whether or not my disagreement is scummy. I actually think it is the former point that's more important here, and the one I'm going to focus on. As for the latter - well, two things: (1) Disagreement regarding strategy does not equal scum tell. You know that by now. (2) If I were scum, the absolute last way I'd broadcast it would be to emphatically support what until this game has been a universally unpopular opinion. Every time someone has suggested a no-lynch in previous games, they have been blasted as potential scum. I'd have to be singularly blundering, careless, tone-deaf scum to have made that argument, because I knew it would put me under the microscope. Point taken. However, with regard to 1, I AM aware of this point. However, there is also a fine line between fair disagreement on strategy (sort of like where, it seems, Roosh and I were yesterday) and one person deliberately manipulating points to put forth a seemingly logical strategy that falls apart under scrutiny. It is this second distinction of which I believe you are guilty. Point 2 is 100% WIFOM, so I won't bother to give it any sort of counter argument, because we'll inevitably argue in circles there. You're right, I realized last night that I failed to include one of the most important parts of my argument, namely the whole balancing ratios. I have argued that the town begins with a numerical advantage and the scum begin with an informational advantage, so let's quantify using my previously defined values Town's numerical advantage ratio: AT=(n-s)/sScum's informational advantage ratio: AS=Σnj=1[(Σsi=1Xi,j)/(Σn-si=1Xi,j)]Now, the second formula is pretty unmanageable, so I'll shorthand it to say XT is the information that ONLY the town knows, XS is the information that only the scum knows and X0 is the information that both share. Thus, I can simplify AS = (XS+X0)/(XT+X0). First, as a point I was trying to make Yesterday, please note that as X0 grows larger, while other values remain constant, AS grows smaller. Thus, it IS to the town's advantage to increase X0. We can also theorize that the game is approximately balanced, thus I will assume that at game start AT = AS. Obviously, as long as AT >= AS, the town is doing fine. Thus, I will not substitute in the values to determine a formula for what is an equitable trade... (n-s)/s = (XS+X0)/(XT+X0)1/n = (XT+X0)/[S(XS+2XT+X0)]However, even that isn't a fair representation because it doesn't take into account the fact that there is often a non-zero chance that a given person is scum. So, if we let PT(j) = 1/n Σni=1Ci,jXi,j,~s be the probability of player j being pro-town, then we can summarize it as ΔT = (1-PT(i))(n-s)/(n-1) - PT(i)(n-s-1)/(n-1) + Xj/(XT+X0)What this means is, is n is large then, the first two terms are nearly equal, especially with a the low certainties with which we're dealing in the early part of the game. Thus, ΔT is very close to the information gained through the death of that individual (ie, the last term) Further, as I pointed out earlier, all of the information that we DID gain from Dotchan's death was already known to the scum; that is, they knew she was non-scum and thus knew there was at least one case of a made-up name, etc. If we didn't kill her, the last term would have resolved to 0, thus meaning the ΔT ~ 0 going into the night. IOW, it is clearly in the town's favor NOT to have a no-lynch early in the game, ESPECIALLY in an information poor game. FYI... I started drafting this about 15 hours ago, before a huge crisis broke out at work. I apologize if it is no longer relevant or is a little difficult to follow.
|
|
RoOsh
FGM
Former BatMod
[on:Wanna see a magic trick?][of:See You, Space Cowboy....]
Posts: 284
[ Exalt | Smite ]
Karma:
|
Day Two
Sept 26, 2007 0:05:09 GMT -5
Post by RoOsh on Sept 26, 2007 0:05:09 GMT -5
Fos BlasterMaster.stop with the numbers! I get it! But Numbers scares the others into not posting! No Longer the Gangster of Love.... :does drugs and passes out To awaken to a new role:
|
|
|
Day Two
Sept 26, 2007 0:21:34 GMT -5
Post by CatInASuit on Sept 26, 2007 0:21:34 GMT -5
Dnooman, your avatar is unfair. It makes me trust you more than I should. (Wash makes me drool.) <Starts moving slowly away from wtf >
|
|
Santo Rugger
Mome Rath
The Obviously Innocent Townie
The Rugger formerly known as Pygmy[on:BYAHH!][of:BYAHH?]
Posts: 3
[ Exalt | Smite ]
Karma:
|
Day Two
Sept 26, 2007 2:20:38 GMT -5
Post by Santo Rugger on Sept 26, 2007 2:20:38 GMT -5
Yeah... BlaM? That English thing? We were kinda serious. Well, not really serious, but give me a break. If that's how you think, great, write yourself a word document, complete with equation editor and all. I'd guess that I have a math background greater than the average player here (4 maths after Calc 3), and although I can follow along, it takes me forever.
Just post the darn summary, because I'm going to completely skip over any future posts with subscripts, superscripts, or Greek letters.
|
|
Santo Rugger
Mome Rath
The Obviously Innocent Townie
The Rugger formerly known as Pygmy[on:BYAHH!][of:BYAHH?]
Posts: 3
[ Exalt | Smite ]
Karma:
|
Day Two
Sept 26, 2007 2:21:52 GMT -5
Post by Santo Rugger on Sept 26, 2007 2:21:52 GMT -5
"... takes me forever.", and doesn't really add any substantial analysis that can't be summed up in "five lines of English".
|
|
|
Day Two
Sept 26, 2007 4:42:43 GMT -5
Post by CatInASuit on Sept 26, 2007 4:42:43 GMT -5
More maths and no explanations I think he has something to hide. vote Spaceman Spiff
|
|
|
Day Two
Sept 26, 2007 6:32:15 GMT -5
Post by Drain Bead on Sept 26, 2007 6:32:15 GMT -5
Blam, you're in a game with at least two English majors. We asked you very nicely to knock off the math. At this point, I think you're purposely trying to obscure stuff. Throw out a bunch of math nobody understands and use it to push your theories and intimidate people into not posting.
vote Spaceman Spiff
|
|
|
Day Two
Sept 26, 2007 7:41:39 GMT -5
Post by storyteller0910 on Sept 26, 2007 7:41:39 GMT -5
Point taken. However, with regard to 1, I AM aware of this point. However, there is also a fine line 'twixt fair disagreement on strategy (sort of like where, it seems, Roosh and I were yesterday) and one person deliberately manipulating points to put forth a seemingly logical strategy that falls apart under scrutiny. It is this second distinction of which I believe you are guilty. Once again, you have failed to demonstrate that I am "manipulating" points. Saying something that you don't agree with is not a synonym for manipulating points. I posted a fairly concise summary of my opinion in my last long-ish post: please demonstrate what aspect of that represents any sort of manipulation. Further, the fact that "seemingly logical arguments fall apart under scrutiny" isn't exactly a reasonable scum tell when the "scrutiny" involves math that I don't even understand (at least not completely). Even if I were to stipulate that you were 100% correct - which I am not willing to do except as a hypothetical at this point - isn't it just as reasonable to suppose that I just didn't see what you saw? You know, JSexton made a great post in the last game about the whole WIFOM concept - I'll dig it up if I can, but the essence of his point was that it's misused a lot. I think it's being misused here. Look, let's take all the complex variables out of the equation (heh. Equation? Get it?). Can you just explain, in layman's terms, how it is that you are so certain that any net gain in information is necessarily of greater value than the personnel loss required to obtain it? I mean, presumably you wouldn't deny that the life of each individual pro-crew player has some positive value. Is there literally any amount of information gained that is small enough that you would say it does not justify lynching town to obtain it? If I had a 95% certainty about dotchan (number strictly hypothetical, because there's no real way to quantify this), and if I controlled the world, I had two options: (1) lynch dotchan, and increase my certainty from 95% to 100% or (2) decline to lynch dotchan, but proceed on the basis of that 95% certainty until something else happens to alter it. From a practical standpoint, choosing #2 would have the same effect on my gameplay, at least in the immediate future, as choosing #1 - except there'd be one more pro-town player in the mix. I still don't understand how you are so blame certain that the extremely minor increase in certainty is obviously worth more than the life of a pro-town player. More importantly, I don't understand how you can regard disagreement on this point as scummy, suspicious, or even poor play. I think reasonable people can approach this game in different ways. You seem to think there's only one best way, and anyone who disagrees is either obtuse or sneaky. I don't really know how to address that, because I think it's fundamentally untrue.
|
|
|
Day Two
Sept 26, 2007 8:17:21 GMT -5
Post by The Real FCOD on Sept 26, 2007 8:17:21 GMT -5
That does it. Not only did I find you mildly suspicious yesterday, but here you go again muddying up the waters with your math that only River could understand. (I started watching the show! Now I can make references!) Like the two before me said, say it in fucking English or don't say it. It seems to me that you trying to cover up something or confuse people.
Vote Spaceman Spiff
--FCOD
|
|
|
Day Two
Sept 26, 2007 8:50:42 GMT -5
Post by ComeToTheDarkSideWeHaveCookies on Sept 26, 2007 8:50:42 GMT -5
I'm purposely not reading anything but Spiff's summary paragraphs, which all seem to be saying that when he crunches his numbers it leads him to advocate that "no lynches" are bad, especially early in the game.
From where I'm sitting, those of you fosing/voting for him are doing so due to his posting style and stubbornness rubbing you the wrong way. Unless you're all extremely passionate about a "no lynch" being a good thing, but that has not been included in the reasoning behind some of your votes.
|
|
|
Day Two
Sept 26, 2007 8:56:56 GMT -5
Post by The Real FCOD on Sept 26, 2007 8:56:56 GMT -5
I've played in several games with Spaceman Blam, and in none of them has he started speaking in Greek and writing with subscripts. Why this time? It seems very strange to me. Furthermore, he was asked to post in English, because most of us don't understand what the hell he's talking about. He didn't, and I want to know why.
Personally, I think a no-lynch is a terrible thing.
--FCOD
|
|
|
Day Two
Sept 26, 2007 8:57:58 GMT -5
Post by Zeriel on Sept 26, 2007 8:57:58 GMT -5
Okay, people. I've taken information theory classes, too, so I DO understand Spaceman Spiff's math. He is absolutely right in that by the rules information theory the town gains more from the information we get in a lynch than we lose in numerical advantage right now. He's also very poor at explaining it in any reasonable way, but there are people here like me who can translate.
Dammit, FOS on everyone voting for Spaceman Spiff. This is at heart a strategy disagreement, and he's not obfuscating NEARLY as much as Roosh's 9000-page missives in Day 1.
The short version is simple--in the early game, because we have a nonzero shot at getting a scum with every lynch, the town gains an informational advantage greater than their numerical loss even if we lynch vanilla townies by mistake. At least, by the maths. Subjective considerations may well lead you to different conclusions, and if the game was unbalanced from the start then all the math goes out the window--but if you thought it was unbalanced from the start, why're you playing?
|
|
|
Day Two
Sept 26, 2007 9:04:08 GMT -5
Post by CatInASuit on Sept 26, 2007 9:04:08 GMT -5
I'm purposely not reading anything but Spiff's summary paragraphs, which all seem to be saying that when he crunches his numbers it leads him to advocate that "no lynches" are bad, especially early in the game. From where I'm sitting, those of you fosing/voting for him are doing so due to his posting style and stubbornness rubbing you the wrong way. Unless you're all extremely passionate about a "no lynch" being a good thing, but that has not been included in the reasoning behind some of your votes. Actually, Cookies, from reading his posts from Day 1 & 2. I think he is actually scum using Maths to hide behind. However, top of my sum pile is Spaceman Spiff; A lot of posts and so little substance in terms of working out who the scum might be. I have found more Maths in your posts than in the last set of Mafia games I have seen all of which is very little use to the average townie. More importantly, the following points really irk me. 1) Your insistence that we should never consider someone's rolename as a reason for lynch/non-lynch. I'm sorry, but so far all the names have fitted the characters of those we know about. I'm not saying it will be perfect due to Bastard Mods(tm) but it is a reasonably good guide. Not to mention the fact it is the reason you voted zuma/dotchan at the end, because my name is the only reason I am posting here today instead of zuma/dotchan. 2) Your game information theory and subsequent vote for storyteller. You may not have intended it to seem that way, but that is exactly how it came across and so that is how I will treat it. (and I did it in 5 lines of english as well ) A good scum tell is saying that someone must be scum for a fallacious reason. I consider this your action at this point. 3) Bandwagon jumping: Some very safe votes into the middle of two bandwagons. And your unintentional slip of forgetting who you had voted for. 4) I can find no analysis of other players, other than your attacks on Roosh for his ideas. For someone who is all for getting information to the town I find the lack of any action, other than an end day lynch, to put this into practice really disturbing. In conclusion SpaceMan Spiff: You have some explaining to do. I'm not voting for him, because of his writing style. I'm voting because there is a good case for him being scum and he has yet to answer any accusations from this post.
|
|
|
Day Two
Sept 26, 2007 9:10:11 GMT -5
Post by Zeriel on Sept 26, 2007 9:10:11 GMT -5
If I had a 95% certainty about dotchan (number strictly hypothetical, because there's no real way to quantify this) That's just it--from a mathematical standpoint there is no way whatsoever to quantify that. Blam's arguing that from the mathematical standpoint, therefore, a no-lynch is actively helping the scum by letting them select all the kills to give us as little information as possible. I can't argue with you--I don't like votes based on strategy discussions either. I also can't argue with Blam that mathematically there IS a best way--but at the same time, math alone can at best tie the game, only subjective and inductive stuff to detect scum can win it. The stupid thing about this fight is that you're both right in your own ways. The only point of disagreement is the value of a no-lynch--which I'll reiterate is mathematically worse than any lynch at this stage of a balanced game. The time for no-lynches is when we have lots of information and need to preserve numbers, information-theory-wise.
|
|
|
Day Two
Sept 26, 2007 9:13:02 GMT -5
Post by CatInASuit on Sept 26, 2007 9:13:02 GMT -5
Okay, people. I've taken information theory classes, too, so I DO understand Spaceman Spiff's math. He is absolutely right in that by the rules information theory the town gains more from the information we get in a lynch than we lose in numerical advantage right now. He's also very poor at explaining it in any reasonable way, but there are people here like me who can translate. Dammit, FOS on everyone voting for Spaceman Spiff. This is at heart a strategy disagreement, and he's not obfuscating NEARLY as much as Roosh's 9000-page missives in Day 1. The short version is simple--in the early game, because we have a nonzero shot at getting a scum with every lynch, the town gains an informational advantage greater than their numerical loss even if we lynch vanilla townies by mistake. At least, by the maths. Subjective considerations may well lead you to different conclusions, and if the game was unbalanced from the start then all the math goes out the window--but if you thought it was unbalanced from the start, why're you playing? And the biggest problem I have with this argument is what happens if we lynch a detective role. Does the value of the infomation gained from the lynching of this role outweight the value of the role over the duration of the game. Lynching a detective would provide non-zero value info to the town. However, the value of lynching the role to the scum is much higher than the value of the info provided by the detective's death. Therefore in this situation, it would be better to vote no-lynch then to lynch the detective.
|
|
|
Day Two
Sept 26, 2007 9:28:53 GMT -5
Post by ComeToTheDarkSideWeHaveCookies on Sept 26, 2007 9:28:53 GMT -5
Do you mean killing a claimed detective or a closeted one? As soon as claims start entering the mix, the math that Spiff is crunching will change, at least that is my understanding. (Math ppl?)
The probabilities he's advancing are only valid on populations of the truly unknown. Right?
|
|
|
Day Two
Sept 26, 2007 9:31:08 GMT -5
Post by Yattara on Sept 26, 2007 9:31:08 GMT -5
*nods* Wash is much love. Not that I'm going to say to his face, because I don't want to anger Zoe. I find the math annoying, but then it's never been my strongest subject and I hated that part of psychology.
|
|
|
Day Two
Sept 26, 2007 9:33:40 GMT -5
Post by ComeToTheDarkSideWeHaveCookies on Sept 26, 2007 9:33:40 GMT -5
NETA: So it isn't really fair to dismiss his ideas with "What if we get really unlucky with the probability and lynch a closeted pro-town detective?" unless you also bring along the optimistic side, that there was just as much of a chance that we could hit the scum investigator.
|
|
|
Day Two
Sept 26, 2007 9:41:28 GMT -5
Post by Pollux Oil on Sept 26, 2007 9:41:28 GMT -5
Oh dear, it does look like Spaceman Blam mathed all over the carpet again, didn't he.
*goes to get his clean-up tools*
*grumble*Leaving sigmas and exes all over the place, I just tidied this house up...*grumble*
My interest has been piqued slightly as to how quickly people jumped on voting Spaceman here for another math post, when he said that he was working on it to post before a crisis at work broke out and delayed him. Maybe he hadn't had the chance to reread the thread yet? Could have just been a Post'n'Go.
I need to think about this now.
|
|
|
Day Two
Sept 26, 2007 9:45:39 GMT -5
Post by CatInASuit on Sept 26, 2007 9:45:39 GMT -5
NETA: So it isn't really fair to dismiss his ideas with "What if we get really unlucky with the probability and lynch a closeted pro-town detective?" unless you also bring along the optimistic side, that there was just as much of a chance that we could hit the scum investigator. Ok, an example of the problem: We have 30 mins before the end of day. The person with the most votes roleclaims as a detective. There is not enough time to find another candidate to lynch. What do you do? According to Spaceman Spiff's maths - we should lynch them to get the extra non-zero information, even though in the long term it would be detrimental to the town. I would say that the value of the role outweighs the info gained from the lynch and so we should vote no-lynch. The problem with the info theory is that it does not take into account the effect of town power roles on the game, either in terms of gaining information from above and beyond a lynch or preserving town numbers to prevent the scum getting ahead.
|
|
|
Day Two
Sept 26, 2007 9:52:15 GMT -5
Post by Mad The Swine on Sept 26, 2007 9:52:15 GMT -5
Just post the confound summary, because I'm going to completely skip over any future posts with subscripts, superscripts, or Greek letters. Hell,I skipped over the first one...the only thing I saw is the "vote storyteller" part. Sorry spiffblamboom,it looks like you put a lot of work into it,but it might as well be greek to me(as I see some of it is). Strong HOS on Spiff,but need to look into some of his readable posts first.
|
|
Santo Rugger
Mome Rath
The Obviously Innocent Townie
The Rugger formerly known as Pygmy[on:BYAHH!][of:BYAHH?]
Posts: 3
[ Exalt | Smite ]
Karma:
|
Day Two
Sept 26, 2007 10:04:54 GMT -5
Post by Santo Rugger on Sept 26, 2007 10:04:54 GMT -5
NETA: So it isn't really fair to dismiss his ideas with "What if we get really unlucky with the probability and lynch a closeted pro-town detective?" unless you also bring along the optimistic side, that there was just as much of a chance that we could hit the scum investigator. Bolding mine. Vote cometothedarkside
|
|
Santo Rugger
Mome Rath
The Obviously Innocent Townie
The Rugger formerly known as Pygmy[on:BYAHH!][of:BYAHH?]
Posts: 3
[ Exalt | Smite ]
Karma:
|
Day Two
Sept 26, 2007 10:06:23 GMT -5
Post by Santo Rugger on Sept 26, 2007 10:06:23 GMT -5
NETA: So it isn't really fair to dismiss his ideas with "What if we get really unlucky with the probability and lynch a closeted pro-town detective?" unless you also bring along the optimistic side, that there was just as much of a chance that we could hit the scum investigator. Let's try that again... confound coding! Bolding mine. Edit Note: NAF edit. Added an underline to the bolded word so you can see what Pygmy was trying to say. Bold doesn't really show up well in quoted text.
|
|
|
Day Two
Sept 26, 2007 10:06:27 GMT -5
Post by ComeToTheDarkSideWeHaveCookies on Sept 26, 2007 10:06:27 GMT -5
Oh come on.
|
|
|
Day Two
Sept 26, 2007 10:08:58 GMT -5
Post by CatInASuit on Sept 26, 2007 10:08:58 GMT -5
NETA: So it isn't really fair to dismiss his ideas with "What if we get really unlucky with the probability and lynch a closeted pro-town detective?" unless you also bring along the optimistic side, that there was just as much of a chance that we could hit the scum investigator. Bolding mine. Vote cometothedarksideAre you bolding what I think you are bolding, even though bolding does not come up in a quote. If so, good catch Pygmy Rugger
|
|