In post 649, SirCakez wrote:Can someone ELI5 why Roberts roll claim makes him scum
It's a low chance that he rolls the same role that Thor did. Far more likely he lied about what he rolled and to avoid getting caught outright later on, he stole someone else's roll (Thor's) so it would be a 1v1 at worst. If RMH had claimed before Thor I'd be saying the same thing about Thor rn, in reverse.
Odds of rolling the exact same role seem lower than the odds of him being scum trying to hide what he really rolled (possibly a decent scum PR that one of his buddies or he ended up with)
In post 696, cyrus62 wrote:@dw where are your thoughts this game should we be aiming for scum 1st or cult leader 1st ? for me i think cult leader then scum but i anit sure cult gets a kill .
It's obviously best to aim for cult leader because their recruit is worse for ratio than a night kill is, but hitting cult leader specifically is tough. So just like kill who we think is scummy (like James)
In post 707, ssbm_Kyouko wrote:Dwlee why were you so strongly for a MegA elimination earlier toDay? I didn't really see anything at the time but was more focused on rolls
There wasn't really a good reason I just wanted to make a wagon
out of 1,000,000 simulated sets of 13 rolls, a TOWN role was only duplicated at least once within a set of 13 rolls 87,162 times, or 8.7% of the time. So James maybe you had the formula right but missed a zero somewhere if you did the math by hand. When at least one role was duplicated, at least one of the duplicated roles was a Normal-modifier (1-3 on the third die) role 30,607 times, which is 35.1% of the duplicated town roles, or 3.1% of all of the sets of 13 rolls.
My script treats 1-12 on the first die as the same by mapping those rolls to the string "Town", and treats 1-3 on the third die as the same by mapping it to the word "Normal". If it rolls over 13, it doesnt roll the remaining 2 die as I didn't care about matching results that were not town. It only rolls the first die 13 times = it does not continue to roll until 13 town rolls, it just rolls 13 times and only keeps track of the Town roles in an array to compare to each other later after all 13 rolls are made.
I can post the script (I think) without it proving randomness as long as I don't post the output, for anyone who wants to check my work that knows javascript.
I didn't take into accouny 13 people, I just considered the potential 9 town, which I think is a fair guess.
If there are 9 town, you just do the probability of absolutely 0 sharing and take it away from 1. Comes to about 88% without modifiers, 8% with.
1 - (20/20)(19/20)(18/20)......
1 - (400/400)(399/400)(398/400).......
i really dislike this mech talk i feel we wasted almost 6 days on it and basically we dont know if any one lied about the 1st die or the 2nd let alone the 3dr
In post 712, JamesTheNames wrote:If there are 9 town, you just do the probability of absolutely 0 sharing and take it away from 1. Comes to about 88% without modifiers, 8% with.
1 - (20/20)(19/20)(18/20)......
1 - (400/400)(399/400)(398/400).......
This is how I thought to do it at first, as the inverse of the probability that there were 0 matching town roles. However, with 1-12 alignment die matching, and the third die having 1-3 equate to matching modifier, it is too complex for me to figure out a formula to represent all of it, even if we try to look only at town rolls by fudging the sample size from 13 down to 8 (60% of the d20 equates to town, 60% of 13 is 7.8. Round up to be generous to RMH) The problem with your formula is that when a 1-3 is rolled, it takes more than one possible "match" out of the denominator, and you have to account for that by summing the 2 scenarios, and it just branches an absurd amount because of the 1-3 modifiers. For every time a modifier is rolled, you have to branch the formula into the sum of (role×Any modifier) + (role×Normal modifier), and once the normal modifier is rolled, future matches become more likely. It's a mess - a simulation is the only way to go for this complex of a scenario
She/They
"No ssbm is not grudging me. She's one of my favorites on the website, and i wanna say vice versa." - Transcend
The day senpai noticed me^
In post 696, cyrus62 wrote:@dw where are your thoughts this game should we be aiming for scum 1st or cult leader 1st ? for me i think cult leader then scum but i anit sure cult gets a kill .
It's obviously best to aim for cult leader because their recruit is worse for ratio than a night kill is, but hitting cult leader specifically is tough. So just like kill who we think is scummy (like James)
so with that aside do you think my roll got fudged i mean i thought 7 7 1 was fair and i gave a little edge not much of one but a slight one.
In post 696, cyrus62 wrote:@dw where are your thoughts this game should we be aiming for scum 1st or cult leader 1st ? for me i think cult leader then scum but i anit sure cult gets a kill .
It's obviously best to aim for cult leader because their recruit is worse for ratio than a night kill is, but hitting cult leader specifically is tough. So just like kill who we think is scummy (like James)
so with that aside do you think my roll got fudged i mean i thought 7 7 1 was fair and i gave a little edge not much of one but a slight one.
I don't think there is a huge reason a PGO would need to be fudged but it's possible.
In post 696, cyrus62 wrote:@dw where are your thoughts this game should we be aiming for scum 1st or cult leader 1st ? for me i think cult leader then scum but i anit sure cult gets a kill .
It's obviously best to aim for cult leader because their recruit is worse for ratio than a night kill is, but hitting cult leader specifically is tough. So just like kill who we think is scummy (like James)
so with that aside do you think my roll got fudged i mean i thought 7 7 1 was fair and i gave a little edge not much of one but a slight one.
I don't think there is a huge reason a PGO would need to be fudged but it's possible.
and yes before james acts like im role fishing . pgo is deadlier for town then scum. for example 1 scum vs 1 town . even trade 1 town left alive and 3 or four dead prs.
I feel like James is trying to cover for RMH, and RMH is the scum that actually rolled a strong town role, and is lying about it so that when people claim they are not are not on the list, they can poe to hit that role.
I am not sure if I really see town motivation in James' posting. I do think his Cyrus push is reasonable but I dont like his dwlee one and (confbias alert) he could be looking to wagon dwlee to get a claim and further restrict the poe of the specific role that scum!James and scum!RMH are trying to hunt for.
I think there are only 2 roles (roleblocker or rolecop) on the list that town!James can reasonably have at this point, and those roles both explain his behavior (his assumption that it is likely that the same role was rolled twice). Parity Cops should not target James as town!James could possibly be the Ugly Rolecop.
James, if you really are convinced that RMH rolled the same as Thor, you can back it up with a roleblocker claim and we can wait for someone to corroborate they also have roleblocker.
I think this is 2 scum right here though.
She/They
"No ssbm is not grudging me. She's one of my favorites on the website, and i wanna say vice versa." - Transcend
The day senpai noticed me^
Wouldn't someone with a miller-like modifier claim it early? But regardless, I can see what you're saying but think James is more solid lim but this is partially motivated by feeling bad about Robert always getting limmed
