Pick Your Power II - Looks like the wine is gone (SCUM WIN)
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
But it's obvious that four scum wouldn'tJack wrote:Let's save the speculation on which numbers the mafia picked until later. It sucks as a game starter, because scum are just as good as town at strategic type arguments, but people get stuck on proving their theory and end up omgus'ing and voting based on bs. Then once that talk is exhausted, there isn't anything real to go on, and the game stymies.
Vote:Tony Montanafor picking (1,1) in the draft = mafiaallpick 1 or 6 as their X number. The risk of leaving the town with too many powerful roles is too high. Therefore singular and doubled numbers have the greatest chances of containing scum.
Kind of weird you're voting based on numbers (even if it's silly).-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Even if that is true, it still lowers the people in brackets of 4's chances of being scum. I doubt scum would want to double up more than once though, because they basically screw themself over with shitty picks.Socrates wrote:Hoopla, Hoopla, Hoopla, if there is anything that Mayor mafia should have taught you its that mafia are absolutely terrified of sticking out in any way, and are much more at home in the crowd.
The scum doubled up on one of their picks, mark my words.
I hope I don't hear any of that 'speculation, wifom!' nonsense, because number analysis will be a VERY viable method of catching scum, because we know there are certain things they won't do.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
dramonic wrote:Do we have any reason to believe scum had the possibility to actually communicate legally before sending their draft/role?
Although, I'd normally be skeptical about someone making statements like that (a similar one is, 'oh I didn't know scum could daytalk' ), this seems genuine.SpyreX wrote: The Draft:[/b]
Once the order is done, you will then straight down the list attempt to pick a power role.If someone above you on the list has picked that role you will be vanilla.
Note: The mafia will be able to talk before the draft begins for a period of approximately 24 hours.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
What would you do if you were scum organising your team?RayFrost wrote:
spreading seems possible, depends on how they think things'll go down and such, so yeh.Hoopla wrote:
Do you think scum would have deliberately chosen to double up on a number somewhere to distance, or spread their choices over 5 different X numbers to improve their chances at better roles?RayFrost wrote: small numbers, higher draft choice, get PR, be happy, rock out
dunno.
either seem plausible.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
In contrast to dramonic's (seemingly genuine) clueless response about the drafting system, this seems like Fate is piggybacking on this mindset to look town. The lame, forced question at the end quantifies this, and just looks like unnecessary words to fill a post that might otherwise had stood out.Fate wrote:
This. I must have missed a rule... IIRC, he said we would all get a PM with our alignment, and then have to send back a reply with the picks.dramonic wrote:Do we have any reason to believe scum had the possibility to actually communicate legally before sending their draft/role?
Does alignment include list of team members? Hmmm... Any scum want to claim and tell us?
Vote: Fate
For being a singular number, too.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
People who are rejecting the concept of number analysis are idiots. It is as much wifom as vote/wagon analysis and other mainstream scumhunting techniques.
We know scum were able to coordinate before/during the draft (despite a few people pleading ignorance to this, some genuine, some probably not). Lets think about it logically though. Some people are trying to sweep this under the wifom rug, and although there are options for scum to take, some are illogical and detrimental to scum's win condition.
Optimum scum strategy is to spread your numbers across 5 different X values, and try to pick up as many of the powerful roles as possible. This is probably the easiest way to cripple the town, by getting power on your side. However, scum may suspect town will figure out what their optimum strategy is, which lowers the effectiveness, and in turn scum may choose to throw the town off. However, there is only so far they can go, before giving too many roles to the town is a negative net result for them.
The good news for town is, we can already pick out unlikely combinations of scumpairs/teams based purely on numbers. At the very most scum would have doubled numbers once - any more than that leaves them too far down the draft. And frankly, if they have done that I'm thrilled because our roles will probably catch them before they can the reap the benefits of illogical number partners.
Right now, the best assumption is that scum have spread across at least four different X numbers. The beauty of this is, if we can keep groups of 3/4 players alive, we have a solid chunk ofprobablytown players. This forces scum to kill in these pools, the places where our good roles probably aren't.
Trust me, number analysis will win this game for us.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
This might read ambiguously. I meant, if we keep groups of 3/4 playersHoopla wrote: Right now, the best assumption is that scum have spread across at least four different X numbers.The beauty of this is, if we can keep groups of 3/4 players alive, we have a solid chunk ofThis forces scum to kill in these pools, the places where our good roles probably aren't.probablytown players.with the samenumber alive, it means there will always be mostly town there.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
You're not a purist. You're a drone incapable of applying critical thinking to a game of mafia - using wiki tells isn't playing mafia.Fate wrote: So if two flip scum the third is automatically cleared? Call me a purist, but I rather rely on other, more traditional techniques. Instead of "follow the numbers." Also, if one of the "doubled up teams" (Pomegranate+Faraday, WRM+fifi, etc.) flips scum, we're supposed to grill the other one?
This game is a unique concept, rich with juicy information just ripe to be analysed, and you want to throw it away?
Again, I hate to break it to you, but your 'traditional scumhunting techniques' are based entirely on assumptions. Traditional concepts, such as bussing, stem entirely from guessing games back and forth between optimum scum strategy, and what townFate wrote: Assumptions are bad. If someone flips scum in the (1, 1) group, should we lynch you based on your own policy? What about the other pairs?thinksoptimum scum strategy is. Number choices are no different - you've just not encountered them before, and refuse to use your brain to figure it out and make rational predictions.
But scum have the advantage ofFate wrote: Another problem is that "optimum scum strategy" is thesameas optimum town strategy. I picked numbers I didn't think anyone else would pick, worked out pretty well for me.knowingfive sets of numbers before they enter, which can dramatically improve their chances of winning, or placing better overall compared to a random sample of five townies.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
To respond to a couple of your questions;
1) No, the opposite. As it means it's highly like the players in the '1' group are town. This is based on scum not having 2+ of their members bid the same number. I maintain, that scum will have spread their numbers across 4 (probably 5) different X numbers.Fate wrote: 1) If someone flips scum in the (1, 1) group, should we lynch you based on your own policy? What about the other pairs?
2) You really think scum is going to shoot 'probable townies' (based only off number choices, I might add) in lieu of probable power roles?
But I want to ask what you think. As scum, would you deliberately bid the same number as a teammate? How would you organise your team's number choices? Because you know if you picked the same numbers, it's automatically going to ruin your chances of good roles. Do you think it's likely scum doubled somewhere? Or they are spread across 5 different numbers?
2) If they were smart, yes. This would be a pretty bad situation to be stuck in for them (the result of going after the singular/doubles);
StrangerCoug (3,1)
The1fifi (3,7)
wolframnhart (3,7)
Devotress (8,3)
Porkens (8,9)
DocPotter (8,10)
Ellibereth (1,2)
Dramonic (1,1)
Hoopla (1,1)
TonyMontana (1,1)
Farside (6,3)
curiouskarmadog (6,12)
Pomegranate (6,4)
Town could easily eliminate A LOT of possible scumteams if there are fewer brackets of numbers. I repeat, the scum WON'T be in the same 2-3 numbers.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
It applies now because we don't have any other information to work with. We don't have wagons/votes to analyse yet or any confirmed alignments. Number analysis eclipses an unnecessary RVS.Ellibereth wrote:I don't get how the whole number thing appliesnow. Can someone explain it to me like I'm an idiot?
The facts are, singulars/doubles have the highest chance of being scum due to scumprobablyspreading their choices over 4+ different X numbers. The only way this point can be disputed is if you think it's likelier scum would have bid the same 2-3 X numbers. And if so, I'd like to hear your logic behind that.
Do you have any thoughts on the number issue?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Only about the Y number which doesn't count for very much of your chances of a high pick. My tactic was to come in and claim '1' early to scare everyone off it (an assumption based on nobody wanting to sacrifice their own draft position). However, most people didn't understand the system or probably submitted their numbers before reading my post.Ellibereth wrote:We have a wagon on Fate.
I get that it's most likely that scum are on the single numbers, but that doesn't reduce the list by that much anyway.
I'm curious why you lied about your picks in the signup thread.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Jack, I think you're taking my words too literally. There are never times when you have 'no' information. You always start a game with knowledge of a player, your role and the previous posts. I'm stating that numbers are a primary source of information - they should hold more importance than whatever scumtell you can generate from a few pages of voting.
Thank you for conveniently cutting out the bit about numbers from that quote you posted of mine, to make it look likethatwas my sole reasoning for my vote.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
It was a combination of reasons. I will likely want to lynch you or Socrates or a double today, and will probably pick the scummiest one of that set of players (barring significant slips from low draft picks). You are currently the scummiest of that set, for reasons outside number analysis.Fate wrote:
Simulpost. So you admit you really did vote me for being a singular number? I was hoping that was a last bit "RVS' joke...Hoopla wrote:Thank you for conveniently cutting out the bit about numbers from that quote you posted of mine, to make it look likethatwas my sole reasoning for my vote.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
It would take a mature mindset to be willing to sacrifice your chance at a role, to be hidden from number analysis. But is it even worth it for scum? As scum, you cannot afford to let town get many of the good roles by trying to get cute.Fate wrote: I would definitely not pick the same number as my teammates. That is what I would do, but I don't know what scum inthisgame has done. Maybe they did spread their choices out, maybe they did double up. But people are selfish, who is going to agree "sure I'll double up and likely not get a role." Also the fact that they only had so much time to choose their numbers, I don't know if they debated the issue enough.
Perhaps a more visual approach to my theory would be helpful;
We have 9 sets of numbers, yes?
1) Socrates
2) Fate
3) bouncy.bouncy, Cobalt
4) RayFrost, Redcoyote
5) Jack, FeFiFoFum
6) StrangerCoug, The1fifi, wolframnhart
7) Devotress, Porkens, DocPotter
8) Ellibereth, Dramonic, Hoopla, TonyMontana
9) Farside, curiouskarmadog, Pomegranate, Faraday
Now, there is a good chance, that 5 of those 9 groups contain one scum each. An outside chance that only 4 of those groups contain scum, but one group has two. But for the sake of this exercise, lets assume it's best case scenario for town, and scum decided to spread themselves across 5 different numbers.
There is little logical reason to try for a lynch in Group 8 or Group 9. At best there is a 25% chance of a scum lynch, and a decent chance none are there at all. Groups 6 and 7 are slightly better, but again, if we go higher up the list, we have better chances of hitting in smaller groups.
The beauty of this is,ifscum spread their choices across 5 different numbers, each group has equal chances of containingascum. Basically, a better than 50/50 shot that a group has a scum in it. So by lynching Fate or Socrates, we have a ~50/50 shot of scum being in that group, but because they're the only ones in that group, it is 100% going to be them. As opposed to a 25% chance in a Group 8 or 9 lynch.
Even if scum have doubled up once, the same principal applies (higher up the list, the scummier), but with slightly weaker odds. But really, from reactions of the questions I've asked others about what they'd do as scum, I think it is more than likely than scum are in fact in 5 different groups (which is such a good position for town to be in)!-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I've changed my mind, based on the sheer ignorance of the town towards the numbers system. Multiple people haven't understood the draft, multiple people think this is wifom territory and not worth touching, and multiple people seem clueless when it comes to even thinking about number analysis. See post 102 for the latest example.Fate wrote: Wow, you're serious? I found a contradiction ladies and gentlemen:
The good news for town is, we can already pick out unlikely combinations of scumpairs/teams based purely on numbers. At the very most scum would have doubled numbers once - any more than that leaves them too far down the draft. And frankly, if they have done that I'm thrilled because our roles will probably catch them before they can the reap the benefits of illogical number partners.
I now doubt scum would have doubled up - and if they have, they shouldn't have because the town is too closed minded to attempt catching them in this way.
Yeah, that scum vig or bomb is really beneficial!Fate wrote:Yet you yourself just advocated lynching either Socrates or myself, who are the probably the ones with roles most beneficial to town.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I'm not. I think you're missing the point of that post. I was showing what would happen if we took out the top drafted players (scum won't do that), because it leaves the players I listed before. Which makes potential scumpairs easier to spot.farside22 wrote:Hoopla why are you ignoring this group in your number analysis?
Socrates (12,1)
Fate (13,4)
bouncy.bouncy (2,1)
Cobalt (2,10)
RayFrost (4,6)
Redcoyote (4,9)
Jack (7,2)
FeFiFoFum (7,11)
See my post 100 for a more rounded view of my opinions.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
It's basically like a one-shot guarenteed vig shot for scum. The only way they don't get to use this power is if that scum is vigged themself - but there is a chance the vigilante is scum, which means double scum kills. I really think we need to keep an eye on the secondry kill choices to assess if we think the vig is town or scum, because allowing scum double kills for 3+ nights will screw us over badly.DocPotter wrote:Someone brought up Vengeful townie in relation to Fate. Is there a good case for scum to take it?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
This is kind of true, and I want to try and explain my point one more time before I give up, because I think people don't understand my posts very well.DocPotter wrote:I'm kinda thinking that the nine groups are kinda like neighbourhoods. I know they can't communicate or such but, some will contain no scum, some one scum, maybe one with two scum.
Not a reason to vote for somebody, but a reason to study groups carefully.
~~
The X number determines what group (or neighbourhood) you're placed in. At the very least, I think almost all of us have agreed, at the very least scum are on 4 different X numbers (with the possibility of 5).
Because the scum couldn't know what town picked, they don't know how many groups of what size there will be. They can only control how many groups they want to be in.
If there is scum in 4 X groups, this gives eachgroupa 4/9 chance of having scum in it. If there is a scum in 5 different groups, eachgrouphas a 5/9 of having a scum in it. Therefore, it is logical to lynch from the smallest group if all groups have equal chance of havingascum in it.
~~
An example assuming scum are in 4 different groups;
Group 1; 4/9 chance of a scum being here. Then, a 1/1 chance it is Socrates (~44%)
Group 9; 4/9 chance of a scum being here. Then, a 1/4 chance it is Faraday (~11%) (Though there is a slim chance the double up is here, which makes it hard to calculate, but I would put it at no more than a 5% increase)
An example assuming scum are in 5 different groups;
Group 2; 5/9 chance of a scum being here. Then, a 1/1 chance it is Fate (~56%)
Group 8; 5/9 chance of a scum being here. Then, a 1/4 chance it is TonyMontanna (~14%)
The only way this is not correct is if scum have tripled up somewhere, or doubled twice (meaning their numbers are in only 2/3 groups). You must prove this is a viable possibility for scum before disputing my numbers (wifom is not an accepted answer).-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Each group has equal chances of containingRedCoyote wrote: Hoopla, assuming I agree with your post 137, which I conceptually do, how do you propose we choose which groups (or neighborhoods) have scum in them? I mean, is Socrates right? Do you really propose we lynch straight down the list until we hit paydirt?ascum, as scum can't know what numbers town would have selected. However, the less amount of people in the group, the less places that scum could be (if there is a scum in the group). Ideally I'd like to lynch singulars/doubles until we've found two scum, and then assess the game from there, but I know nobody here wants to delve into the wonderful world of numbers.
I think the most realistic option we have to bring us closer to playing optimally, is lynch only today based on numbers. On Day 2 we have role information, alignment flips, vote analysis and other reliable sources of information available. Number analysis should still be used, but it may take the backseat to everything else, which I'm fine with. For now, we should lynch the scummiest (in thetraditionalway) singlular/double.
Today we have a ~50 chance of hitting scum if we go for Socrates or Fate. Number analysis gives us these odds, but nobody wants to play them. Do people really value their own scumhunting abilitiesthatmuch, that they'd rather lynch in a traditional manner, and have a maybe 20% of hitting scum? Let it be known, I have a bleak view of Day 1 normally, and believe it is a lot more random than some think. But in this game, we actually have information to start with! I'm so frustrated nobody wants to use it. You can do your iso-whatevers, buddying tells, wagon analysis etc tomorrow when that information surfaces, but now, numbers are our best option.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Do you understand my logic in post 137? If you don't, I will try and betterly explain it for you. If you do, please tell me where my logic fails, and you get the impression I'm making it up.Devotress wrote:We actually have a 4.5% chance of hitting scum, if anyone was wondering.
There are plenty of people who want to lynch fate though(me), but your numbers case is bad. You're overblowing the numbers way too much, and you're doing the fate wagon no favours with all this numbers talk. No matter how many percent chances you decide to make up(and making up IS what you are doing with "50%"), it just sounds like you're trying to justify lynching all power roles.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
In the theory I illustrated the only unproven assumption is the scum are on 4 or 5 X numbers. If that is true, then there is no other variable, and the numbers are reliable. However, I stated this at the end of my post;Devotress wrote:
Your logic doesn't fail(though I personally don't want to lynch that way), and I see where you're coming from. It fails when you start to call it a mathematical statistic. It isn't. It's based on an assumption of an unproven theory. The theory can be sound and accurate, but when you start assigning math to it you're making stuff up. Any number you assign to it other than 4.5% is wrong.Hoopla wrote: Do you understand my logic in post 137? If you don't, I will try and betterly explain it for you. If you do, please tell me where my logic fails, and you get the impression I'm making it up.
You get what I mean?
This says the numbers are different if scum are in only 2/3 sets of X values, which I deem very slim, as it is an illogical, nonsense gambit for scum to even consider, as it guarentees very low draft picks.Hoopla wrote: The only way this is not correct is if scum have tripled up somewhere, or doubled twice (meaning their numbers are in only 2/3 groups). You must prove this is a viable possibility for scum before disputing my numbers (wifom is not an accepted answer).
In the slim chance scum have done just this, the town has enough potent roles to compensate for a D1 numbers based lynch. If the scum are on 4/5 X numbers, then my percentages are correct, and not a lie.
I see where you are coming from though. They are numbers based on an assumption. But in this assumption, the numbers are truthful. I guess it depends on how likely you deem this assumption to be. If it is highly likely to be true, doesn't it make it worthwhile to pursue such good odds on D1?
~~
Do you think scum have doubled up? Would you have suggested doing as scum? What about tripling up? How many different X numbers do you think scum chose?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
That's a shame you don't want to use your brain, when it is the most important information we have.Ellibereth wrote:All the posts with number strategies percentages and whatever are annoying the hell out of me.
I'll muck my way through it tomorrow and post something real tomorrow. It's my impression (and correct me if I'm wrong), that not too many people actually care about all the number/strategy/whatever stuff, so stop posting it and clogging the thread kthx.
~~
Pom, the numbers are accurate;
If scum are spread 1-1-1-1-1 over five different numbers, then Socrates and Fate have a ~56% chance each of being scum.
If scum are spread 1-1-1-2 over four different numbers, then Socrates and Fate have a ~44% chance each of being scum.
They have lower odds for combinations like 1-2-2 and 1-1-3 etc, which should be considered if you think it's a realistic possibility scum would do it. However, this quote suggests you don't think that;
Which is a good point, but contradicts being against the numbers. Because the numbers for 1-1-1-1-1 and 1-1-1-2 are solid.Pomegranate wrote: I would be extremely surprised if scum decided to all pick the same numbers- getting PRs is better for them than having that defense.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
The risk is (generally) relative when it comes to lynching a claimed powerrole, as roles aren't in anyway related to alignment. You have to weigh up the value that role has for town, and then for scum. If it favours scum, or is neutral, then it isn't a bad lynch. Remember, they will only be in the position to claim if they are the scummiest player in the game.Jack wrote: I don't think it's good play this early to force a claim from the 2nd drafter. Are we going to lynch him if he claims one of the top roles? No.
Some examples;
The consensus decides Z is the scummiest in the game. Z claims vig - it is worth lynching as extra town kills aren't worth the possibility of scum having double kills. That is such a ridiculously powerful ability that can cripples towns so quickly.
The consensus decides Y is the scummiest in the game. Y claims cop - it is probably not worth lynching as it's town value is far better than it's scum value (nothing).
The consensus decides X is the scummiest in the game. X claims Governor - it is definitely worth lynching as it has so much more scum value than town. Preventing lynches deprives town of information, and gives scum another free nightkill. The only town motivation is preventing scum from having this role, but it is not worth the risk of leaving a potential scum governor alive.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
At the moment, I want to lynch you over Socrates, and any of the doubles. I only want a claim if there is enough town support for your lynch (that means L-1 and at least one person prepared to hammer). Please don't claim before then.Fate wrote:^For the record, are you saying you want to see me put at L-1 and claim, Hoopla?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Scum are scared of the numbers and are bussing Fate hard. Either that, or they're wagoning Fate to protect Socrates - I haven't come to a definite conclusion just yet. Either way, I agree that the speed of this wagon implies there is quite probably a decent amount of scum on it. I hope it's because of bussing.RedCoyote wrote: I clarified it in the following post. I just meant that it seems as though people are opening up to Hoopla's plan, because I certainly do not see any other reason for Fate's wagon to be as large as it is. There was no scumslip, so far as I can tell.
~~
A sidenote; it's funny how all the people wanting to do traditional scumhunting aren't really doing much of it at all.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I agree with you - I think multiple scum are probably on your wagon. If you're being bussed, it is probably later in the wagon after it's gathered enough steam to be viable. This makes me think (if you're scum), you're either partners with Socrates, are the cop (no scum use), are the bulletproof or bomb (with a scum vig), or are vanilla after going for the same role as Socrates. There are plenty of possible scenarios where bussing isn't a bad play.Fate wrote:First off, this is bullshit. "Scared of numbers?" Scum are bussing me hard? Because the "slip" I made (what was that again? I don't remember because I'm not scum in the first place).
Why would scum buss their (most likely, unless you want to think both Socrates and I are scum) TOP draft choice?
Answer that, please.
I think most of the people on your wagon are using some form of number justification - you cannot say that they aren't influential. I hope you don't take it personally, but I believe scum are spread 1-1-1-1-1 or 1-1-1-2, which means for me, you or Socrates are the highest percentage lynch. I play mafia to win, and your lynch gives the best odds for Day 1.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Ray, are you seriously voting Fefi for that reason when your game contributions have been thus;
RayFrost wrote:vote: jack
answer the questionRayFrost wrote:
small numbers, higher draft choice, get PR, be happy, rock outHoopla wrote: Why'd you pick 4,6 as your numbers?RayFrost wrote:
spreading seems possible, depends on how they think things'll go down and such, so yeh.Hoopla wrote: Do you think scum would have deliberately chosen to double up on a number somewhere to distance, or spread their choices over 5 different X numbers to improve their chances at better roles?
dunno.
either seem plausible.
One vote on page two, and nothing else without being asked first.RayFrost wrote:
Prob go for more powers, I guess.Hoopla wrote: What would you do if you were scum organising your team?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Well, here is my last town game (Mini 892: Mayor Mafia). Monkeyman was wagoned at a similar pace to you, and claimed cop on D1. We lynched him because the role was mostly valueless now that he was outed - we deemed it the better percentage play to hope he was lying. Towns don't win games by being too pussy to lynch claimed powerroles.Fate wrote: Here, let me try another approach to get through to you. Hoopla, in regular mafia games, what is your approach to Power roles? Let's put all the numbers and PYP and everything aside for a moment, just to humor me, and tell me: What would you do in this situation?:
1. We have Fate, at L-1, softclaimed PR.
2. We have many votes on him that are suspicious.
How do you proceed?
You're also missing an important factor - power roles aren't indicative of alignment which you seem to be pushing. A scum vig is equally (or probably more) beneficial to scum, than a town vig is to town. We should be lynching players with the highest odds of being scum even if that means lynching a powerrole.
They aren't statistics - I'm not doing studies and recording/displaying data. We're talking aboutFate wrote: Stop. Tunneling. Get off your high horse. We get it, your statistics are right. Using them to advocate lynching me may have even made scum jump at an "easy lynch"-a justifiable target. Dramonic seems the biggest offender of this.probabilitiesin the different scenarios scum could choose to adopt. The probabilities in some scenarios are more than double of what you would get from a random or normal lynch on Day 1 in any other game. If the town deem certain the scenarios (where we have high odds for a scum lynch) likely, then we should pursue them.
The stigma against numbers is quite amusing to me, because I treat it as a viable form of scumhunting. If you strip back the fancy name and genralised tells, scumhunting is the term used to describe figuring out who scum is. People have different techniques, some are very logically or psychologically inclined, while some favour gut - but what all these techniques attempt to do, is find out information toimprove your oddsof lynching scum.
Nobody is perfect and we all have different interpretations of what is scummy, and what isn't. This is shaped through the success and failures of various techniques, seeing what works, what doesn't, and this always evolves as there is no set tell that is definitive. Once a tactic becomes common, most people will add it to their arsenal and use it/slightly tweak it to scenarios where they deem it useful. Vote/bandwagon analysis, bussing/distancing are all stock things that are debated in a normal game - but this is still guesswork based on the patterns you've established scum will follow.
I am using probabilities of scum positions to try and improve the odds of lynching scum on Day 1. The initial assumption (of whether scum doubled/tripled up or not) is guesswork based on what I think scum would/wouldn't do. Every alternate tell you use, uses an element guesswork, trying to decide if it is something scum would do. The reason I am favouring probabilities based on my assumption, is because I think I can guess what scum are doing with more likelihood, than guessing using more common techniques.
You'd be surprised how difficult it is to catch scum without any starting information - it is certainly a lot more random than people believe it to be. And without trying to sound too hypocritical, I think people overvalue their own personal reads/thoughts a lot (especially on D1).
I'm sorry for the rant, but I dislike numbers being trash-talked when the numbers are fine. It's the previous assumption the numbers are based on that should be disputed - and I have yet to see compelling arguments that scum have done anything other than 1-1-1-1-1 or 1-1-1-1-2.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
This needs to be taken into account when assessing anyone's play - whether it be based on numbers, 'scumhunting', role-information or whatever. It's kind of a time-waster (and insulting of people's intelligence) to go, 'hang on, before you take in this information, just remember i might be scum!'DocPotter wrote:You know Hoopla, and Happy birthday btw, much as I like talking theory and such, there is one basic assumption in your probabilities that you have failed to address. Your own alignment.
If you are town then your other assumptions are fair, but if you are scum then you'll already know how many 'X' numbers the scum are on and can tailor your assumptions to suit.
So rather than lynch the top, lets lynch you. If you flip town we know we can trust your assumptions and place a little more faith in your theory/plan. If you flip scum then ...
I have explained my theory as eloquently as I can - you just have to decide if my motivations behind this theory seem sincere, scum oriented, or not important/not understood. At the moment, I think many people don't care about my numbers as much as me, which makes the 'lynch to prove my theory' irrelevant. If Iknowthe town will follow my plan I will sacrifice myself to give credit to my theory and show it was town motivated.
But I certainly expect numbers to play a lesser role when we have other genuine information D2.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
A watcher watches two players visit the house of a townie who is murdered - don't tell me you wouldn't use that information against a player, even if there was barely anything scummy about them.Fate wrote:
I'm not forgetting that you're #2 in the draft order, as there was barely anything scummy about you, Fate*Hoopla wrote:Wow, I'm dazzled by how quickly Fate's wagon has evaporated. I still wouldn't be opposed to lynching him, but it makes me think scum were trying to protect Socrates via a Fate wagon;
Unvote, vote: Socrates
I'm not forgetting you, Fate.
*Fixed.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Fate, in this hypothetical, you know those two players have a 50% chance each of being mafia. Therefore they have equal chances of being scum variants of whatever their role is. The town will not win if we're too scared to lynch powerroles.
How do you think the town will win if scum have the vig and cop?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
That act is becoming lame and unhelpful.Devotress wrote:Hoopla wrote:
Is there any way you can see the number choices improving the odds of particular players being scum/town? Or is it all ~WIFOM~ to you?
It does increase there odds, I said so in my last post.
By 5.1 percent.
(as per jacks very helpfull equation post)-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I love you too. But you don't understand, my brain automatically callibrates things into percentages and numbers when I take any information in, you know? Like, even regular information from scumhunting filters through a part of my brain that turns it into psuedo-percentages (even if they aren't always quantified). Something can be a weak or a strong tell, we may have multiple tells we're working with, and some data/tells we use are more reliable than others. Whatever it is, we naturally add up in our subconcious all the information (of varying importance) and rank it against everyone else. This is a very mathematical process, even if we don't understand (or realize) that we're giving every player in the game 'odds' of being scum.farside22 wrote: Hoopla I love you when I say this and I'm not saying your numbers analysis is shit at all but I think we should use a combo of scumhunting and number analysis to find scum. I would like you to actually read the game and point to those you think are scum and why that have nothing to do with the numbers right now.
So, because I put so much stock in the value of this number analysis, it's clouding my ability to see past it, and set it aside. For example, Pom is probably scummier than the percentages purely from number analysis gives me - but then I remember she is one of 4 people who picked the number 6. It's highly likely that there is 0 or 1 scum in this group, so even if she is scummy, it does not make it a good chance lynch for me. I'm sorry, it's just stupid.
I'll give you a list of players who are scummier than number analysis implies, but giving such a list implies they are my top choices for a lynch, which is in most cases completely false. But here you go anyway;
Pomegranate- myself and others have brought up contradictions with her talk of numbers, and then about not wanting to use numbers. Then her request of Fate to claim was quite fishy - I don't think she has been very sincere.
Fefi- Undercontributing, and made a very bad vote to get onto the Fate wagon. I think a lot of scum were on Fate's wagon, and he's probably one.
RayFrost- Is suspicious for targetting a lurker, whilst being one himself. It was a ridiculous attempt at a vote.
~~
But as I said, these reads from 'traditional scumhunting' aren't very valuable compared to number analysis for me. I know I'm pissing a lot of people off because of it, but this is such good information. Fefi and Ray are the two doubles I'd consider lynching first if we can't lynch Socrates/Fate.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I suspect you on the basis if scum went 1-1-1-1-1 or 1-1-1-2, you have a 54% and 46% chance of being scum (respectively). Although your wagon is probably scum driven, the odds I believe (numbers) probably eclipse this other belief (a lot of scum on your wagon). It's possible they are bussing, it's possible they are protecting Socrates via a Fate wagon, or it's possible you're both town. Bussing is a viable possibility late in the wagon, maybe 7th onward - it's a reasonable expectation for a wagon to go through when it reaches that point, and scum may have thought it was better to cut their losses. I don't see scum starting your wagon though.Fate wrote: @Hoopla:
Again your logic escapes me. A lot of scum were on my wagon, yet you still suspect me? Hardcore bussing of #2 power role to look good later on (even though half the wagon, had any real reasons that they could later look back on and say "hey I was right!") doesn't seem plausible to me right now, no matter what my role is. Even if I'm vanilla scum such a buss wouldn't be the right play as the first wagon of the day...Fefi - Undercontributing, and made a very bad vote to get onto the Fate wagon. I think a lot of scum were on Fate's wagon, and he's probably one.
I've recently admitted Socrates is a better choice than you now, due to my hunch that your wagon was scum-fueled, but it still doesn't hide the numbers which override a lot of other less significant data.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
No, I propose we lynch a singular number for just Day 1, when the lynch is most random. As days go by, we will generate plenty of non-number information (role info, bandwagon/vote analysis, confirmed alignments etc.) that I am happy to use. But as it stands, we have very little of that, and I deem number analysis the best information we have to work with. I'm using other methods outside number analysis to determine who I would prefer to lynch if the numbers mean multiple people are tied due to just numbers.Fate wrote: Since numbers "override" a lot of other data, i.e., everything fucking else about mafia.
Your proposal is that we lynch straight down the list all singular numbers right? And you think this is a viable way to play D1?
For example, you and Socrates should have equal odds of being scum because of numbers. Earlier I deemed you scummier (for reasons other than numbers) than Socrates, but it has since swung now that your wagon looks scum-driven. Then if I can't get this to happen, I will vouch for the scummiest double player.
~~
Again, if we're too scared to lynch power roles, we will not win this game. Are you going to let a scum vig or scum bulletproof live for multiple days just because of their role?DocPotter wrote:It's viable. But there's a decent chance that neither Socrates nor Fate are actually scum, so we'd just probably be wasting our best two power roles.
The risk is relative, and grounded in simple theory. Most roles are of more of less equal value to scum than to town (with some significantly more powerful for scum, aka vig) - therefore the risk associated with lynching a vanilla scum/vanilla townie is linear in relation to lynching a scum bulletproof/town bulletproof. If most roles stand to benefit each side roughly equally, we should be lynching based on chances of scum, not whatever role they claimed (with small exceptions).
Remember, all roles are open to both alignments. Alignment has nothing to do with role.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I don't understand this summation. I'm not talking about lynching roles, explicitely, but it does coincide with the top of the draft. The theory is scum are likelier than random to be higher up the draft order due to potential planning. Therefore, if there is near equal value for town/scum for most roles, it shouldn't worry us about lynching a vig or a bomb etc.DocPotter wrote:You want to use theory to lynch roles on the (decentish) chance they are scum, I want to use theory to help lynch scum and don't care if they have a role.
Fair summation?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
That's a valid concern, and I hope my theories won't be forgotten (although I know they will be) if/when I die. I feel like my numbers are good, because even if I am wrong and scum went 1-2-2 or 1-1-3, the town is still in a really good position, with scum likely having no/few good roles. I just don't see why scum would sabotage their chances at doing well in the draft to double/triple up - when all they get in return is a wifom defense against number analysis.DocPotter wrote: ATM, I'm flippin between Jack and Hoopla appearing slightly scummy for pushing optomised math and not looking at the their assumptions properly. Could be scum trying to push theories affected by known scum choices.
And when you look how swiftly people have called in wifom/unimportance about trying to analyse possible scum tactics, I sure hope they were foolish enough to double/triple up out of fear, because the town won't ever do anything about it, as it's an unconventional way of guessing who scum is.
~~
Due to a looming deadline, I'm willing to drop Scorates and Fate due to lack of interest (and fear of lynching powerroles ), and lynch from the pool of doubles since there are still decent odds there. I personally refuse to be apart of any lynch of a player in a group of 3 or 4, though.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I'm confused by bouncy's D1 defeatist attitude considering his play in the game he replaced in. It might just be a lack of interest (not that it's acceptable), but I get a town vibe from him. I'm also kind of curious to see if he'll open up on D2.
Fefi and RayFrost are very good lynch choices, due to their underperformance. RayFrost is especially concerning when you look at his play and realize his only activity has been attacking a lurker.
Fefi needs to die for his ill-timed, opportunistic vote on the Fate wagon.
Vote: Fefi-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I shouldn't have to - the numbers should stand alone. It's the assumption beforehand (guessing how many different X numbers scum picked) that needs to be debated.Fate wrote:@Hoopla: Don't act as if dying is going to prove your 'numbers' theory is spot on. Scum can kill you just to make it seem that way and mislead town... (Hint: Watch for Hoopla dying and Dramonic trying to take up the mantle to lynch me tommorow).
I agree me dying doesn't prove a theory, but it would show my motivations were truthful, which seems to be the primary concern of people in the know. Now you're adding another clause about scum killing me to manipulate numbers? What you don't realize is that every piece of information in this thread is subject to scum manipulation - it feels like you're arguing as if numbers are more likelier to be tainted than any other piece of information we get. You're discrediting numbers using logic that basically equates to, 'but it might be scum manipulating us!'
It's like saying, 'don't think activeness and genuine-seeming scumhunting is pro-town, it might be scum trying to do it!' We're all limited in our abilities to break down and scan information, but we must always make decisions based on our best predictions. Provide reasons why scum aren't likely to do what I am predicting, without saying 'scum might be doing it', because as soon as you jump into that line of thinking, you also remove any chance of catching scum using that information.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
This feels fake to me. It's like a statement of disapproval, but it's only made when it's certain the wagon is going to go through, which makes it read like he's fishing for town credit for being anti-Fefi wagon.Socrates wrote:8 votes in a little over 1 page? The wagon is an apathy driven lurker lynch? We are less than 24 hours from deadline?
FeFiFoFum is going to flip town. Wheee!
I also think scum are more likely to make these sort of declarations prior to a lynch, because they can orchestrate their stance on the situation accordingly, due to knowledge of what the flip will be. Why would a naive townsperson make such a grand statement? It's an unnecessary risk - why risk Fefi being scum, as it incriminates you. But then if he flips town, you don't get that much credit anyway, so why risk it?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I agree scum likely piled on toward the end - the weird thing is the last two players on the wagon flipping town. I'd give decent odds to the remaining players on the wagon (especially the latter half of the wagon) being scum.Socrates wrote:
Uh, because I believed it? That many votes in that little time for that kind of reasoning never actually results in a dead scumbag.Hoopla wrote:
This feels fake to me. It's like a statement of disapproval, but it's only made when it's certain the wagon is going to go through, which makes it read like he's fishing for town credit for being anti-Fefi wagon.Socrates wrote:8 votes in a little over 1 page? The wagon is an apathy driven lurker lynch? We are less than 24 hours from deadline?
FeFiFoFum is going to flip town. Wheee!
I also think scum are more likely to make these sort of declarations prior to a lynch, because they can orchestrate their stance on the situation accordingly, due to knowledge of what the flip will be. Why would a naive townsperson make such a grand statement? It's an unnecessary risk - why risk Fefi being scum, as it incriminates you. But then if he flips town, you don't get that much credit anyway, so why risk it?
Also note that Bouncy.Bouncy had more votes than FeFi did when people started pileing up on FeFi and is also guilty of pretty much anything you could argue that FeFi was and it it shouldn't be hard to connect the dots.
I might be open to a bouncy lynch just based on that occurance alone.
Interesting theory about Bouncy though, it makes sense.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Which of his town games are you getting this meta from? Does he have more than one?Ellibereth wrote:For the nth time, I don't see bouncy scum making this post:
http://www.mafiascum.net/forum/viewtopi ... 79#2108179
Why is it alignment telling, rather than a player just being behind and taking the easy way out?-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Good question Devotress! Short answer is yes, and I'll show some updated groups;
Group 1:Socrates
Group 2:Fate
Group 3:bouncy.bouncy
Group 4:Jack
Group 5:RayFrost, RedCoyote
Group 6:StrangerCoug, The1fifi, wolframnhart
Group 7:Devotress, DocPotter, Porkens
Group 8:Hoopla, TonyMontanna, Ellibereth
Group 9:Farside, curiouskarmadog, Pomegranate, Faraday
There are now 4 players who are the only ones in their group, where 'group' represents the 'X' number they chose. I think this will gradually get easier to understand and rationalize as the game progresses. Unless people want me to talk about numbers, I won't bother, because I know you all hate them/don't understand/think they're wrong. But I think closing down entire groups is a sensible idea, because it means scum can only hide if they're all in 2-3 groups.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
I tend to disagree. This analysis distorts the events, and makes it seem like the bouncy wagon was stagnant. But itfarside22 wrote:I have to disagree with RayFrost. The wagon on bouncy wasn't any closer then the wagon that was on FFF.
from one page it went from 2 votes on FFF to 9 votes;
- (L-9)Bouncy.Bouncy - 3 - RedCoyote, Faraday, Fate
Cobalt - 1 - The1fifi - (L-11)
Dramonic - 1 - TonyMontana - (L-11)
Fate - 3 - Bouncy.Bouncy, Dramonic, FeFiFoFum - (L-9)
FeFiFoFum - 2 - Rayfrost, wolframnhart - (L-10)
Jack - 1 - DocPotter - (L-11)
Pomegranate - 2 - Socrates, Farside - (L-10)
Socrates - 2 - Hoopla, StrangerCoug - (L-10)
The1fifi - 3 - Cobalt, Porkens, Devotress - (L-9)
wolframnhart - 1 - Jack - (L-11)
Official Vote Count
Players needed to lynch: 12
TM is definately the worst offender when it comes to jumping on a BW to vote for FFF. Redcoyote and faraday were both on bouncy before moving to FFFF. So if bouncy isn't scum I could see those who were on the bouncy wagon jumping onto the FFF wagon to give it the fuel to go.Bouncy.Bouncy - 3 -- (L-9)RedCoyote, StrangerCoug, Porkens
Fate - 3 - Bouncy.Bouncy, Dramonic, FeFiFoFum - (L-9)
FeFiFoFum - 9 - Rayfrost, wolframnhart, Faraday, Hoopla,- (L-3)The1fifi, Ellibereth, Jack, Fate, TonyMontana
Jack - 1 - DocPotter - (L-11)
Pomegranate - 2 - Socrates, Farside - (L-10)
The1fifi - 2 - Cobalt, Devotress - (L-10)
Players not voting: curiouskarmadog, Pomegranatewasin competition when you look at 5 different names (in blue) were on his wagon over the same, short amount of time. The timing of vote counts doesn't reflect that though.
When you take that into consideration, it really looks like scum were trying to muscle a Fefi lynch ahead of bouncy. I'm prepared to vote for bouncy based on this evidence (with the added bonus of numbers playing to his advantage). But I also think there is at least two scum in the names in red I listed, which make good back-up lynches.
Vote: bouncy bouncy-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
It's a tough one to balance, I suppose. It's all well and good to be wary of lynching town power roles, but what are you going to do if scum have a vig, a tracker and a bomb, for example? How do you seriously expect to win a game if you're not going to lynch these roles, if they're scummy enough in the first place (whether that's by numbers or anything else).RedCoyote wrote: I'm kind of throwing a lot of pronouns in this paragraph, but it's hard for me to explain exactly how I feel about the issue. If anyone is confused, I just mean to say that I like Hoopla's plan as long as it's considered another piece of scumhunting rather than a universal tool.
I'm most paranoid about the prospect of scum having a vigilante - I don't think some people understand the amount of damage that can do if it's in scum's hands. And let's be honest, it's not such bad odds. If the vig misses tonight, I want it to claim tomorrow so the town can direct the kill.
Okay, back on topic, I agree with your post and I think I have compromised my position to allow numbers to be just one facet of my lynch choice now. We have enough information now to not have to solely rely on number analysis.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Claim or die more like it.Faraday wrote:Vote Bouncy.bouncy
Contribute or dieeeee.
A thought: for claims in the future (and now), if they're vanilla town, I think it's wiser if they don't claim what role they tried to get. Or at least, someone this high in the order anyway. We don't want scum figuring out where specific roles lie.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Yeah, I have no qualms with that.farside22 wrote:
That high in the order I can agree with but last game we found the information from the VT's helpful later in the game and it kept scum from lying about their role.Hoopla wrote:
Claim or die more like it.Faraday wrote:Vote Bouncy.bouncy
Contribute or dieeeee.
A thought: for claims in the future (and now), if they're vanilla town, I think it's wiser if they don't claim what role they tried to get. Or at least, someone this high in the order anyway. We don't want scum figuring out where specific roles lie.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
Eh, initial thoughts; it's probably more likely to come from town. I think there are probably better roles scum would have gone for with the 3rd pick. Although, I want to wait for a potential counterclaim.bouncy.bouncy wrote:
Role copHoopla wrote:Claim or die more like it.
Go ahead and hammer me if it helps the town, my role is pretty much useless for town with the PYP mechanic and I only picked it because I didn't want the mafia to have it.
Should I post my result from last night? Idk if that would help the town or the mafia.
I don't know about the result. I'm leaning towards no, or maybe tell us the player andthenthe town decides whether or not you can say the role from then. But lets wait for everyone else.-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008
-
-
Hoopla
- Posts: 10788
- Joined: October 12, 2008