What happened in M2? I didn’t follow that game; would you please post a recap?
Speaking of voting, something I noticed in M3 that was driving me bonkers was the very low vote counts that were getting people lynched. Lack of consensus is bad. I don’t know how we can force ourselves to reach a consensus in only 5 days, but letting a relatively small number (25%!) of votes determine the lynch was very bad for the town. *Everyone *needs to vote even if ‘unsure,’ otherwise we risk giving scum easy cover. So I state right now, I’ll suspect anyone who gets to the end of the day with no registered vote. We still may end up with relatively few determining the dunk, but I can’t see a way around that. Unless we as a town agree to chatter and discuss and cajole so that we can reach a true majority on each vote.
Oh, **chrisk **posted a strategy suggesting everyone post results of fake investigations, so that the real cop could post info in disguise. The strategy was flawed.
I have an idea I’d like to discuss, that *could *be flawed, but won’t damage the town simply by discussing it.
As per usual, this topic always surprises me with how much I have to catch up on when I get on in the day.
Um, so…how is talking about methods used for determining who gets a vote not a discussion of strategy?
WHERE HAVE YOU BEEN? Lurking???
It seems to me that the flaws are (1) the first person to post is almost certainly NOT the cop/detective/Oracle/Zadoc, and (2) nobody can tell which “investigation” was legit until the cop has been killed off. Do I have that right?
Re Random Voting/Unvoting/FOSing:
I don’t really see the harm in random voting the first day. That said I really don’t understand why someone would random vote only to later recant said vote in the very same real life day. What is that going to accomplish? The stated defense for random voting is getting people to talk as they are afraid that they will die. So why unvote so quickly?
But Clockwork Jackal didn’t make a revenge vote. All he said was that he was inclined to FOS you for saying that speculating on the distribution of players is a scum tell, which seemed to imply that you were discouraging such speculation as unnecessary. (I see that’s not what you meant, but it initially appeared that way.) He didn’t even point an actual FOS your way. If anything, it would seem that your vote for him is something of an over-reaction.
I do agree that discussing the distribution of players is not very fruitful this early on in the game, which is why I will withhold pointing a FOS at you for now.
I should have quoted more of Pasta’s post to be more clear.
I’m talking analysis of the rules and game mechanics and exploiting them to the town’s advantage. I don’t mean to distract from the scum hunting and voting fun, but I think it is a worthwhile discussion. If everyone else disagrees, I’ll let it go.
You are right.My apologies for saying revenge vote when it was just a revenge FOS.
I am satisfied with my vote, for now,however.
Well, I’ll take a shot at it. (Usual disclaimers apply, degree is in English, put too much faith in this at your own risk.)
OK – so we’re looking at a maximum of seven scum, maybe eight in a pinch. The weird part is that we have … um … something on the order of one and a half doctors and one and a half detectives. I’m going to figure three scum to balance the Oracle and Apprentice; could be one or two to balance the Priest and Disciple. There are probably approximately three Monks (I’d be surprised if there were as many as five), so I’m figuring another scum there. The number of vanilla townies is probably closer to twelve than six = two more scum; one Prophet = -1 scum. I’m not sure how the Avatar would affect the balance, but I can see ways his existence might benefit either side, so I’m going to guess that he’s a null, like the Crusader and Psychopath.
However, there have to be enough nonbelievers for them to have a plausible shot at winning, so I’m guessing the numbers of scum and citizens are probably lower than they would otherwise be.
So, I’m estimating that the balance probably looks something like this:
Scum: 6, of whom two have power roles – but I’m willing to buy seven.
Oracle, Apprentice, Priest, Disciple, Crusader: one of each
Monks: 3-ish
Citizens: 10-ish
Alchemist, Psychopath: one of each
Regular Non-believers: 4-ish
Oh I didn’t mean to discuss chrisk’s strategy specifically. I just didn’t want to have the whole thing blow up in my face like his did. Which I seem to have managed not to avoid.
Here is my thinking: ignore if you so desire.
We have two investigative roles (I’m assuming there is one Oracle and one Apprentice). For the town, it is best to maximize investigative power by minimizing overlap of investigations. I propose that we divide the set of players into two groups and arbitrarily assign one group to the Oracle and one group to the Apprentice. That way we can be sure that investigations are disjoint. This method does nothing to reveal roles and gives sufficient coordination to avoid overlapping investigations.
In fact, I’ll go as far as to say I propose the Oracle investigates players 1-15, and the Apprentice investigate players 16-30. Feel free to ignore me if you like. It was just a thought. We don’t really need to discuss it much more and just let them decide on their own.
Now this line of thinking got me thinking about the Priest and Disciple as well. I thought about it and decided the benefit to town woefully do not outweigh the benefit to scum in employing a similar strategy.
Yeah. A dead player’s words are a lot more meaningful than a living player’s.
–FCOD
As far as our investigative arm(the Oracle and his/her Apprentice(s)) are concerned, the Non-Believers are a superset of the Cultists. So rooting them all out should be their priority. Now the question for the town becomes, how do we sort the Cultists from the Non-Believers. This is an important exercise for three major reasons. Firstly, the Cultists are the ones trying to kill us. They should be the targets. Secondly, the Non-Believers count towards the town’s side in the equation to determine a Cultist victory or not. Killing Non-Believers(as opposed to just those who investigate as Non-Believers, which includes most Cultists) advances the Cultist’s win condition. Thirdly, while it is theoretically possible for the Non-Believers to take a victory away from the Believers, it’s pretty damn unlikely. Let’s not repeat the mistake of going for the Mason win in MIII by focusing on unlikely events when the most realistic threat is still out there.
On the other hand, the real value of investigation isn’t who it implicates, it’s who it exhonerates. Each verified Believer becomes a trusted townie. Given the ratios of Believers to Non-Believers, even with the confounding factors of the Prophet, and his/her convert(hmm, typing his/her made me think, what if the Prophet is a woman? How would she implement the conversion…) showing up as Believers and the pro-town Non-Believers being lumped in with Cultists, the odds are still heavy that the first few investigations by our Oracle and his/her Apprentice(s) will turn up Believers. These basically become globally known Masons/Monks once the Oracle is sniffed out(or snuffed out :() and their breadcrumbs are deciphered. These form a solid base of trusted players and can see the game through to the endgame. Think jsgoddess in the original Werewolf game. This is confounded somewhat by the Prophet and his/her convert registering as Believers, but we’ll just have to remember, if we haven’t found the Prophet and/or convert, when the endgame comes, even a “verified by the Oracle” notch on someone’s belt loses it’s protective power.
Also, lynching non-believers flat out kills the Psycopath. He/She will never go on their murderous rampage if they hang from a rope. This is a minor plus for the concept of hanging Non-Believers. It eliminates a potential loose cannon. Although, now that I think about it, odds are the Cultists will take care of the Psycopath for us. They’ll know who he/she is because they’ll know who they tried to night-kill when it failed(activating the Psycopath), so I don’t think normal everyday townies need worry on this front. Killing Non-Believers also takes out the Alchemist, who can be of benefit to the town(blocking a night kill), but will more likely block one of the town’s power roles(investigators or doctors) because there are more of them than Cultists(and the Alchemist only has about a 17-20% chance of blocking a Cultist kill anyway). On balance the Alchemist using his power aids the Cultists more than the townies.
So, some numbers. A general rule of thumb, borne out by mathematical analysis, is that you should have a number of scum roughly equal to the square root of the number of players. So in a game with 30 players, we should have 5-6 scum. More or less skews the probabilities of a scum or town victory significantly. I think we saw this in MII, which(no offense NAF) had too many scum. IIRC the spreadsheet NAF published showed he had 7 scum, which would have been enough for a game with 49 players! And the fact that there were two scum left at the end of the game(even allowing for good play by both Kat and storyteller0910) indicates there was an imbalance. So when the Oracle(et. al.) investigate(s) tonight the odds are they’ll hit non-scum by about 4-1. This is fine. After the first Day/Night cycle we can start making “trusted” lists and this gives our investigative wing their first chance to leave breadcrumbs.
For the first lynch I suggest we all use random.org The numbers show that in a game with no investigators(which the first day basically is) the best strategy is random voting.
Enjoy,
Steven
On Preview: For the love of Nairu people! I knew it took me a long time to compose this post(interpspersed with work), but you’ve got more than full page in between when I started and when I previewed!
I will be the first to admit that this kind of strategic thinking isn’t really my strength, but I can’t see any reason why we shouldn’t employ this strategy. I only see up sides as far as the Oracle and apprentice are concerned.
Anyone else?
I am not getting dragged into a random vote discussion again (See case study M4 crew v ArizonaTeach, 2007 for details), but I will say that I myself was wondering about numbers. That’s probably because it was such a big deal on the Hispaniola. Does the recruitment scenario enter into anyone’s calculations here?
The really big flaw was that it made the detective way too easy to identify right off the bat. ((2) was actually a built-in part of the plan rather than the flaw – the idea was that the trust lists would make it easier to ID the cop / detective’s bread crumbs after the character had been killed.)
I don’t see any problem with sachertorte’s plan as long as we keep it flexible (there could, for example, come a time when we need a sure read on one of the players on the second half of the list, and it would then make sense to have the Oracle investigate this person instead of the Apprentice).
FWIW, I don’t think everyone is going to be agreeing on any one particular point any time soon, if ever.
I agree - since we can always be 100% sure of a dead player’s identity, but rarely of a live one’s.
This may have already been mentioned before, but “random voting” on the first day doesn’t mean everyone is random voting - scum will be sure who NOT to vote for, at least, and the Apprentice and Disciple know the identities of the Priest and Oracle, while the Monks all know each other, correct? Not much help initially, but in the long run I feel it will be (since some of these roles will almost inevitably turn up dead before the game ends), so I second sachertorte’s suggestion that everyone ends up voting someone by the end of the first day.
I was wondering about the powers of the Oracle and the Apprentice - if/when the Oracle finds out the identity of the Apprentice, would it make sense for the Oracle to perform the ritual on the Apprentice until the Apprentice’s accuracy reaches 80%? Or should the Oracle just ignore the Apprentice and concentrate on finding the cultists? And when would it make the most sense for the Priest to protect all of us at the price of being able to protect no one the following night? It seems that early on in the game, a few dead bodies are a necessary evil in helping us gather clues as to who the scum are.
Thoughts? sips coffee