Micro 1003: Divide and Conquer - Game Over!

[Bingle]

Fairly confident in Vandertown aorn, btw.

Why? Not sold on the "not thinking through mech" argument, you can just post what you thought when you first saw the setup

From interactions: I doubt skitter puts the majority of her teams fate in their hands and I doubt that you go for the early softbus right here when it's a potentially game ending strategy.

[Bingle]

I REALLY don't like the idea of leaving the big pool lynch up to the small pool, I'll fight hard against that one.

Not quite what I meant when I said it was step one. I think we should have the three of us decide who we would elim in the next few days and then use that to inform the rest of D1.

[PookyTheMagicalBear]

but we're guaranteed at least a scum by flipping thru 3p, no?

going for 6p today means we *might* get a win today but otherwise we end up in a similar space tomorrow with a large hood and an unresolved 3p that if i'm counting right we won't be able to flip through, which by definition means mitigating the chances of netting one scum, if i'm counting this right

it kinda reminds me of coalition of whether we ought to be trying to win (getting the coalition day1) or whether we ought to try to mitigate the losses if we don't get the coalition day1 by getting a coalition that will at least be easier to sort through up failing, as those things aren't always necessarily the same thing.

but i think it's usually better here to net the guaranteed scum and then try to find the other scum in the 6p because while it's not as ~flashy~ it'll help our overall chances of winning

if we go thru the small pool we have

(1/3) We hit scum right away - we are playing 1-6 Mountainous with 1 IC rest of the way. (50% WR)

(1/3) We hit scum on 2nd try - we are playing 1-4 Mountainous rest of the way. (40% WR)

(1/3) We hit scum on last try - we are playing 1-2 Endgame (33% WR)

Overall WR -> 41.1%

if we go thru the big pool we have:

(1/6) we hit scum on first try -> game over

(1/4) we hit scum on second try -> we have 2 shots to elim 1 scum in three people (2/3) victory

We miss both times -> we have a (1/2) shot of hitting scum, then a (1/3) endgame if we hit correctly.

Overall WR -> 50%

[Bingle]

but i kinda feel infinity has entered the game assuming that we're going to flipping in 3p first and is setting the stage for why that ought to be me

It had occurred to me.

[Infinity 324]

I have a permanent grudge against anyone I have correctly scumread in the past and failed to eliminate

Skitt get in here and fite me

VOTE: skitt

Ydrasse you're free for now

ok let's fite
- from your pov why is it me and not bingle
- do you understand how from *my* pov this looks to me like you entered the game trying to force a 1v1 with me
- aside note: if infinity is scum and taking this tactic i expect her partner to be less confident / the partner doesn't think they would be able to do the 3p (no offense to whoever that may be if that is indeed the case)

-If jingle is as strong as scum as you, it could be either one
-Sure, but I expect you to be TRing me once you read up (I think)

[Bingle]

Pooky's math is off. EV for little pool elims until scum with optimal nightkills is ~40%, EV for big pool elims until scum with optimal nightkills is ~44%. More importantly, actually hitting scum in the big pool leaves us in a much better place to find the remaining scum from a subjective standpoint, and the only way we lose by looking there first is if the scum is literally the towniest player in that pool.

[Bingle]

they're probably not together from skitter agreeing that it's a thing.

Eh. skitter trying to deny that it's a thing doesn't work. Too many people in the lobby know better.

[Menalque]

Mena, you have significant partner equity with Infinity. Thoughts?

Uh, why

[Vanderscamp]

but we're guaranteed at least a scum by flipping thru 3p, no?

going for 6p today means we *might* get a win today but otherwise we end up in a similar space tomorrow with a large hood and an unresolved 3p that if i'm counting right we won't be able to flip through, which by definition means mitigating the chances of netting one scum, if i'm counting this right

it kinda reminds me of coalition of whether we ought to be trying to win (getting the coalition day1) or whether we ought to try to mitigate the losses if we don't get the coalition day1 by getting a coalition that will at least be easier to sort through up failing, as those things aren't always necessarily the same thing.

but i think it's usually better here to net the guaranteed scum and then try to find the other scum in the 6p because while it's not as ~flashy~ it'll help our overall chances of winning

if we go thru the small pool we have

(1/3) We hit scum right away - we are playing 1-6 Mountainous with 1 IC rest of the way. (50% WR)

(1/3) We hit scum on 2nd try - we are playing 1-4 Mountainous rest of the way. (40% WR)

(1/3) We hit scum on last try - we are playing 1-2 Endgame (33% WR)

Overall WR -> 41.1%

if we go thru the big pool we have:

(1/6) we hit scum on first try -> game over

(1/4) we hit scum on second try -> we have 2 shots to elim 1 scum in three people (2/3) victory

We miss both times -> we have a (1/2) shot of hitting scum, then a (1/3) endgame if we hit correctly.

Overall WR -> 50%

Your math is wrong.
The top half is correct, but the chance of hitting scum on the second try is not 1/4, it's 5/6 x 1/4.

I don't know how you got 50% but the equation is
(1/6)x1
+ (5/6 x 1/4) x 2/3
+ (5/6 x 3/4) x 1/2 x 1/3

which is 40.97%.

Going for the big pool first is marginally worse but probably doesn't outweigh wanting to lynch whatever our strongest read is.

[Vanderscamp]

I have a permanent grudge against anyone I have correctly scumread in the past and failed to eliminate

Skitt get in here and fite me

VOTE: skitt

Ydrasse you're free for now

ok let's fite
- from your pov why is it me and not bingle
- do you understand how from *my* pov this looks to me like you entered the game trying to force a 1v1 with me
- aside note: if infinity is scum and taking this tactic i expect her partner to be less confident / the partner doesn't think they would be able to do the 3p (no offense to whoever that may be if that is indeed the case)

-If jingle is as strong as scum as you, it could be either one
-Sure, but I expect you to be TRing me once you read up (I think)

Why do you not think it can be skitt as scum who decides to scum read you?

[Vanderscamp]

they're probably not together from skitter agreeing that it's a thing.

Eh. skitter trying to deny that it's a thing doesn't work. Too many people in the lobby know better.

Ok, good to know

[Vanderscamp]

Pooky's math is off. EV for little pool elims until scum with optimal nightkills is ~40%, EV for big pool elims until scum with optimal nightkills is ~44%. More importantly, actually hitting scum in the big pool leaves us in a much better place to find the remaining scum from a subjective standpoint, and the only way we lose by looking there first is if the scum is literally the towniest player in that pool.

I think this is wrong too, walk me through how you got these numbers if you care to argue it

[Menalque]

Who cares, math is boring

[Menalque]

6p first is obviously correct I don’t care what the math says

[Vanderscamp]

but we're guaranteed at least a scum by flipping thru 3p, no?

going for 6p today means we *might* get a win today but otherwise we end up in a similar space tomorrow with a large hood and an unresolved 3p that if i'm counting right we won't be able to flip through, which by definition means mitigating the chances of netting one scum, if i'm counting this right

it kinda reminds me of coalition of whether we ought to be trying to win (getting the coalition day1) or whether we ought to try to mitigate the losses if we don't get the coalition day1 by getting a coalition that will at least be easier to sort through up failing, as those things aren't always necessarily the same thing.

but i think it's usually better here to net the guaranteed scum and then try to find the other scum in the 6p because while it's not as ~flashy~ it'll help our overall chances of winning

if we go thru the small pool we have

(1/3) We hit scum right away - we are playing 1-6 Mountainous with 1 IC rest of the way. (50% WR)

(1/3) We hit scum on 2nd try - we are playing 1-4 Mountainous rest of the way. (40% WR)

(1/3) We hit scum on last try - we are playing 1-2 Endgame (33% WR)

Overall WR -> 41.1%

if we go thru the big pool we have:

(1/6) we hit scum on first try -> game over

(1/4) we hit scum on second try -> we have 2 shots to elim 1 scum in three people (2/3) victory

We miss both times -> we have a (1/2) shot of hitting scum, then a (1/3) endgame if we hit correctly.

Overall WR -> 50%

Your math is wrong.
The top half is correct, but the chance of hitting scum on the second try is not 1/4, it's 5/6 x 1/4.

I don't know how you got 50% but the equation is
(1/6)x1
+ (5/6 x 1/4) x 2/3
+ (5/6 x 3/4) x 1/2 x 1/3

which is 40.97%.

Going for the big pool first is marginally worse but probably doesn't outweigh wanting to lynch whatever our strongest read is.

I will say that I think in practice the math for this is kind of misleading (even though if I'm not wrong, which I'm pretty sure I'm not, it comes to the exact same conclusion that I've been advocating) because it assumes we're lynching randomly.
I'm pretty sure a fairly big part of the reason why lynching into the small pool first comes out ahead is because from a random kills POV, the fact that we force scum to NK into the big pool before solving there reduces our chance of miskilling in the big pool when we eventually do get there, but in practice I don't think this is as helpful since when scum just NK the towniest player in the big pool, we were not going to kill them anyway, so math assuming random lynching assumes this is a bigger deal than it will turn out to be in practice.

On the flip side of this, the math argument also doesn't care if a scum gets eliminated at any point in the game or the reads we would get from that.

[Infinity 324]

Still TR u tho

U can’t take that away from me

And was correct too

You started of SRing me, then you TRed me, then you were trusting skitt. Somehow I feel like you would've been SRing me if skitt wasn't there but I guess we'll never know

You can still sheep skitt's read on me this game, we can't be partners

i think you're conflating jk++ with pyp x|y i|m, no ? isn't pyp the one where mena tried to flip you like six times but i wouldn't let him. or was that something else

either way i'm not really townreading you here so

I just reread jk9++

Pooky: we should lim in the 6p hood
Infinity: ehh idk
Pooky: here are all the advantages of limming in the 6p hood
Infinity: on the other hand, here are all the advantages of limming in the 3p hood
Everyone: wow infinity is so against limming in the 6p hood

i'm also not sure this actually, like, happened ?

It was tongue-in-cheek, I don't actually think there was much of a misrep it's just hard to argue for a neutral position and be interpreted correctly

[Menalque]

Infinity u said something earlier that made very little sense

[Menalque]

You can still sheep skitt's read on me this game, we can't be partners

This was kind of a bizarre thing to say given that you were advocating scum!skitt within your hood at the time

[Menalque]

Yah I think the problem with putting myself in with skitt would be the paranoia, esp in the 3p

ngl i was kinda thinking you were being +town for you until you said this
but this *does* make me paranoid ^

btw skitt i don’t understand why this makes u paranoid

[skitter30]

Skitt can read me well I think but she's gonna say I'm town

i kinda feel like you're expecting me (and wanting me) to townread you here while also pushign me as the likeliest in 3p to be scum?
and like you want me to townread you to get mena off your back

which doesn't entirely make sense as a coherent pov to me here

[Infinity 324]

Why do you not think it can be skitt as scum who decides to scum read you?

I'd think that it would be too hard for scum!her to justify, but maybe she can anyway. I'm not sure. I feel like I'm already kind of out of my scumrange though, and will get more so as the game goes on.

[Infinity 324]

You can still sheep skitt's read on me this game, we can't be partners

This was kind of a bizarre thing to say given that you were advocating scum!skitt within your hood at the time

See my last post, same with skitt who asked p much the same thing

[Bingle]

I think this is wrong too, walk me through how you got these numbers if you care to argue it

Exhaustive analysis of the cases within the parameters. All of the possible townwins for littlepoolfirst, for instance, are

TTSS (.083)
TSS (.066)
TSTS (.07)
SS (.0556)
STS (.0556)
STTS (.074)

Which sums to 40.4%.

I assume you will take my word at this point, but analyzing setups is kinda my jam.

[Ydrasse]

3p group: doing all this

6p group: music party

[Menalque]

Oi bingle explain why I have equity with infinity