TWG CXVI: Assassin's Gambit (Game Ovev? GAMA OVAR?!)

[TheZeldaPianist275]

Doesn't that question slightly leak information about who you are? Also, I will not be answering that question.

Also, wanted to comment on this. Poet, the premise of TWG involves lying to each other, so no, A# wasn't giving anything away with that question. Because it's in everyone's best interest to pretend to be a human, you expect someone to claim to be a human either explicitly or implicitly, regardless of whether they actually are.

[BlackDragonSlayer]

Care to elaborate? Do you think non-assassin players would be better at directing the lynch to an assassin by defending themselves than an assassin redirecting the lynch to a non-assassin by defending themselves? Right now I do not see how.

Given my extensive experience in TWG, I'm inclined to say it's more reliable than the alternative you're suggesting. Even if the mechanics of this game are different from a normal game, the lynching process still works fundamentally the same. As I said before, I can understand where your initial thought process originated from, but at the end of the day I feel that if you wouldn't suggest this tactic in a regular TWG, it also wouldn't apply in this game.

Emphasis on "ideal scenario" and "not completely impossible".

Sure, but I don't think it would ever become a possibility in this game.

Firstly, one of the assassins will die, since there is no victory condition with both assassins alive.

Say we get a day phase end state of:

Spoilered to reduce wallpost size

Assassin
Ward
Target

The assassin can only win by convincing the Ward to vote for the target. The ward, knowing he is town but not knowing the roles of the other two, should push for a lynch of either of the other two and so should the target. If the assassin is only trying to convince the Ward to vote for the target, he exposes himself and will get lynched. If the assassin also tries to convince the target to vote for the ward, he risks succeeding and losing anyway.

So, having only three players remaining is to the detriment of the assassin. With more players it becomes easier for the assassin to guard his Ward without arousing suspicion.

I did some calculating what outcome was most likely if we get to a day phase end state where the only target left is the hot target and it cannot be found by the assassin by POE, but concluded that the calculations were going to be too hard for me to figure it out, so I admit I'm unsure if in that case it is better to have many town or few town.

[close]

If the end state is three players. With 10 players and 3 kills per day/night cycle, it's entirely possible the day phase end state could be, for example, two assassins and two wards left, in which case it's a guaranteed human loss no matter who's lynched. It's also possible for there to be a 5+ player MYLO with one assassin and one of their targets left.

Realistically, the faster the game is over, the better chances we have. The longer the game goes on, the more chances one of the assassins has to win. At least, that's how I see things playing out.

Sorta feels like you're either tunnel visioning here, or deliberately throwing out another red herring (either as a distraction or, more favorably, a deliberate test).

I suggest that there should not be discussion about Assassin strategy in the topic, as it can help them form their plans.

As with before, I understand where you're coming from, and I've generally tried to avoid being explicit about the topic thus far as a precaution, at least until we've all come to a consensus about whether it's something worth discussing. At the same time, I feel like it also helps if town as a whole can potentially anticipate assassin strategies—if every possibility is on the table for all parties to see, then it'll likely be easier to suss out the assassin if they follow one of those strategies. If one or both assassins are experienced players (which I think would include you, me, THC, TZP, and Toby), I feel like not talking about it publicly would be wholly disadvantageous to town, since an experienced player would likely be able to devise such strategies on their own regardless.

[davy]

Good to have a third pair of eyes go over my idea. I'd like to hear what everyone else thinks of it as well.

A regular townie is lynched: Humans are slightly harmed, both wolf teams are slightly helped

I'd like to hear your reasoning for this. As I see it, regular townie does not count towards the wolf victory condition and both the pool of potential wolves/wards and target shrinks. Fewer votes will be cast by townies, but I think this matters little since townies can't really coordinate votes anyways (because they don't know which other players are townies and which aren't).

Of course, the profile of the game changes significantly after this, because one assassin team being down strengthens the other one comparatively. So lynching a ward changes the calculation. I think Davy's point holds water though.

Yeah, that's a really cool thing about this game: every death changes the make-up of the remaining player base by a signifigant enough amount that it will require a slight strategy shift from all teams.

[Toby]

a lot of talk

i dont see any benefit to anyone claiming. Only a human hot target knows their role so I guess they can claim their human role if they are pressured too much for a lynch, would at least buy them a phase but I don't see point in claiming for any other reason.

So much speculation happening and I don't see how Davy can try co-ordinate how late game might look given how unpredictable this game looks like it could be for 2 night kills and brutal humans

I'll toss my light vote on TZP as a placeholder hunch

[davy]

Working on a reply to BDS, but seeing how I just got ninja'd, I'll quickly reply to Toby:

So much speculation happening and I don't see how Davy can try co-ordinate how late game might look given how unpredictable this game looks like it could be for 2 night kills and brutal humans

That was an example, not a prediction or something I was trying to co-ordinate. The question under discussion was whether normal towny deaths were more benificial for the assassins or for town, and the example was showing that a situation with few players is detrimental to the assassins

[BlackDragonSlayer]

Swear I'm not a wolf this time.

sus

Kai, are cardflips just the color of the person killed, or do they say the explicit role of the deceased?

Good idea of thinking for this question.

Want to reply to this part of Davy's opening arguments. It is true that only 1 in 3 "normal" human deaths progresses the game toward a wolf victory, but to better evaluate the risk profile of lynches, we should be weighting the different outcomes, not just relying on probability of a certain outcome.

A ward is lynched: Humans are greatly benefitted, one wolf team is greatly harmed, one wolf team is slightly helped
A regular townie is lynched: Humans are slightly harmed, both wolf teams are slightly helped
A cold target is lynched: Humans are moderately harmed, one wolf team is moderately harmed, one wolf team is greatly benefitted

So I think that it's probably true that lynching a ward is more helpful to the humans than lynching a cold target is helpful to the wolves, despite the fact that they have an equal chance of happening. For the finance bros in the crowd, it's the difference between returns and risk-adjusted returns.

Of course, the profile of the game changes significantly after this, because one assassin team being down strengthens the other one comparatively. So lynching a ward changes the calculation. I think Davy's point holds water though.

TZP coming in strong with this analysis. I like it. I disagree with the ward lynching evaluation though; I'd say one wolf team is greatly helped, since the wolf needs to kill the opposing assassin anyway—a human and an assassin being killed gives them one more check off their win condition and two less potential picks for their hot target.

[Toby]

That was an example, not a prediction or something I was trying to co-ordinate. The question under discussion was whether normal towny deaths were more benificial for the assassins or for town, and the example was showing that a situation with few players is detrimental to the assassins

I dont see how it really matters to humans considering all the humans apart from the hot targets just think their normal human anyway. And we don't want the hot targets to die.

[davy]

Given my extensive experience in TWG, I'm inclined to say it's more reliable than the alternative you're suggesting. Even if the mechanics of this game are different from a normal game, the lynching process still works fundamentally the same. As I said before, I can understand where your initial thought process originated from, but at the end of the day I feel that if you wouldn't suggest this tactic in a regular TWG, it also wouldn't apply in this game.

Agree to disagree

Sure, but I don't think it would ever become a possibility in this game.

Makes sense, I'm not having any high hopes of it happening either.

If the end state is three players. With 10 players and 3 kills per day/night cycle, it's entirely possible the day phase end state could be, for example, two assassins and two wards left, in which case it's a guaranteed human loss no matter who's lynched. It's also possible for there to be a 5+ player MYLO with one assassin and one of their targets left.

I mean, I was considering working out different end states, but scrapped that due to me diving into assassin strategy. But to dive into your examples:

• Two assassins two wards end state is possible, but - as every phase cycle there are two kills (a lynch and an opposite assassination) that have four targets that result in an assassin death - I find that quite the unlikely outcome.
• 5 player LYLO is quite a lot more likely with normal townies alive then without, so al the more reason not to worry about lynching them

Realistically, the faster the game is over, the better chances we have. The longer the game goes on, the more chances one of the assassins has to win. At least, that's how I see things playing out.

Maybe I'm taking your first sentence too literal now, but when the game is over (wether fast or slowly) our chances of winning ar 100% or 0%.
The chances of any team winning this game is very dependant on the make-up of the remaining players. The way I see it having a townie be lynched increases the chance of a favorable end game state for town. I'd like to hear everyone else's opinion about that, though.

As with before, I understand where you're coming from, and I've generally tried to avoid being explicit about the topic thus far as a precaution, at least until we've all come to a consensus about whether it's something worth discussing. At the same time, I feel like it also helps if town as a whole can potentially anticipate assassin strategies—if every possibility is on the table for all parties to see, then it'll likely be easier to suss out the assassin if they follow one of those strategies. If one or both assassins are experienced players (which I think would include you, me, THC, TZP, and Toby), I feel like not talking about it publicly would be wholly disadvantageous to town, since an experienced player would likely be able to devise such strategies on their own regardless.

I don't really see what benifit town has from anticipating wolfings, there are no town actions at night that could be used for that benifit. On the other hand, I think it is very important to discuss assassin strategies after they have played out as the assassin can't change their action anymore and town can learn a lot from what the assassins chose to do.

Optimal assassin play is quite complicated this game I feel like. If I were an assassin I would very much welcome any information that town has already thought out for me so I won't have to figure that out myself.

TZP coming in strong with this analysis. I like it. I disagree with the ward lynching evaluation though; I'd say one wolf team is greatly helped, since the wolf needs to kill the opposing assassin anyway—a human and an assassin being killed gives them one more check off their win condition and two less potential picks for their hot target.

I think the benifit for humans is still greater than for the other assassin, though. Humans get the first their two win cons, with the other one still having two targets to go down, whereas the other assassin potentially (or certainly if it happens day 0) has his cold target still remaining alongside the hot target, the latter of which only having themself as a target to go down, but with both a lynch and an assassin to get rid of it.
Typing it out like this, it does seem closer than I initially gave credit for.

[BlackDragonSlayer]

Agree to disagree

¯\_(ツ)_/¯

I mean, I was considering working out different end states, but scrapped that due to me diving into assassin strategy. But to dive into your examples:

• Two assassins two wards end state is possible, but - as every phase cycle there are two kills (a lynch and an opposite assassination) that have four targets that result in an assassin death - I find that quite the unlikely outcome.
• 5 player LYLO is quite a lot more likely with normal townies alive then without, so al the more reason not to worry about lynching them

Saying "eh, it's unlikely" doesn't exactly give me confidence in your plan. Or you.

davy

I'm the type of player to try and take all possibilities into account (or, I should say, give all foreseeable possibilities a fair evaluation before discarding them), no matter how unlikely—because if you simply disregard a possibility as "unlikely," it will completely blindside you if it comes to pass. The fact that you seem to be doing exactly the opposite doesn't sit well with me.

Maybe I'm taking your first sentence too literal now, but when the game is over (wether fast or slowly) our chances of winning ar 100% or 0%.
The chances of any team winning this game is very dependant on the make-up of the remaining players. The way I see it having a townie be lynched increases the chance of a favorable end game state for town. I'd like to hear everyone else's opinion about that, though.

Bold statement is true, but an early win is much more likely to be town-favored than assassin-favored. Town only needs to have two people dead, whereas each assassin needs to have three people dead. The former is much more likely to happen earlier, while the latter gets more likely of being a realistic possibility the longer the game goes on.

I don't really see what benifit town has from anticipating wolfings

Not sure what you mean by this. I never said anything about anticipating wolfings specifically.

On the other hand, I think it is very important to discuss assassin strategies after they have played out as the assassin can't change their action anymore and town can learn a lot from what the assassins chose to do.

I don't think we'll get much of a chance to discuss assassin strategies after they've played out, as if they've played out we likely wouldn't know about them in most circumstances (this could be dependent on the answer to TZP's question).

Optimal assassin play is quite complicated this game I feel like. If I were an assassin I would very much welcome any information that town has already thought out for me so I won't have to figure that out myself.

Subjective. At the end of the day it depends highly on who the assassins are. If all five of the players I mentioned are town, then I 100% agree with you. Given that we can't know that at this point of the game () I'm inclined to agree with holding off on discussing assassin strategies unless most of us feel like it's worth doing.

I think the benifit for humans is still greater than for the other assassin, though. Humans get the first their two win cons, with the other one still having two targets to go down, whereas the other assassin potentially (or certainly if it happens day 0) has his cold target still remaining alongside the hot target, the latter of which only having themself as a target to go down, but with both a lynch and an assassin to get rid of it.
Typing it out like this, it does seem closer than I initially gave credit for.

That makes sense, but at the same time, one assassin being dead basically turns the game into a manhunt with slightly more favorable win conditions for the wolf than a usual manhunt. From the perspective of an experienced TWG player (I specifically recall my experience as third party in Maestro's Relax, I'm a Doctor game for an example of an assassin-esque role in action) I trust the remaining assassin with 2 kills left to win more than I trust humans with 1 kill left to win.

[davy]

I dont see how it really matters to humans considering all the humans apart from the hot targets just think their normal human anyway. And we don't want the hot targets to die.

Did you read the strategy I proposed? Feel like I'd just be repeating myself if I'd attempt to rebut this directly...


Also to clarify: not defending yourself does not mean you can't push for a different lynch in my book, so a player under fire can still try to get a different lynch going.

[davy]

Bold statement is true, but an early win is much more likely to be town-favored than assassin-favored. Town only needs to have two people dead, whereas each assassin needs to have three people dead. The former is much more likely to happen earlier, while the latter gets more likely of being a realistic possibility the longer the game goes on.

Both town and assassin can win night one. Obviously not going to post the requirements for an assassin night 1 win, they can figure that out themselves.

Okay, we are comparing four things, so it's important to know what we are comparing against. Bold is what I think is favored.

Early town:early wolf
Early town:late wolf
Early town:late town
Late town:early wolf
Late town:late wolf
Early wolf:late wolf

Late wins are more likely than early wins, I feel that much should be obvious.
Early town only needs two correct kills whereas early wolf needs three. Favored for early town.
Late wolf needs to keep two players alive and kill one, but has two chances to kill that one. I feel like this is favored for late town, but it's closer than early v early.

So all in all I will cede that getting to the late game improves the assassins chances of winning by more than the increase from town. So humans are slightly hurt from losing a towny, but it only feels very slightly to me. Still think it's worth going for my strategy, but it does seem less worth it now.

Not sure what you mean by this. I never said anything about anticipating wolfings specifically.

Ah, I jumped the gun there. In addition to wolfing strategies there are also lynching strategies. I think those could be fine to discuss during the day phase where they are happening, but I would still approach it with caution.

I don't think we'll get much of a chance to discuss assassin strategies after they've played out, as if they've played out we likely wouldn't know about them in most circumstances (this could be dependent on the answer to TZP's question).

Right, this was assuming card flips show the entire card, not just the color.

That makes sense, but at the same time, one assassin being dead basically turns the game into a manhunt with slightly more favorable win conditions for the wolf than a usual manhunt. From the perspective of an experienced TWG player (I specifically recall my experience as third party in Maestro's Relax, I'm a Doctor game for an example of an assassin-esque role in action) I trust the remaining assassin with 2 kills left to win more than I trust humans with 1 kill left to win.

Not quite, the town win con is considerably more favorable than in a normal man hunt as the one kill needed to be made can be made on two players. This changes the math by a significant enough amount that I trust the humans more than the assassin in that scenario.

[davy]

And with that, I'm going to sleep. See you tomorrow everyone!

[XiaoMigros]

checking in damn a lot happened, hold on

[A# Minor]

 

[XiaoMigros]

this convo is too much to digest rn ill think qbt it tmrw