Gameshow Mafia (Reroll): Game Over
Forum rules
- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
-
-
Ankamius Survivor
- Survivor
- Survivor
- Posts: 21802
- Joined: May 9, 2011
- Location: Target Locked. Initiating Combat.
the problem isIn post 7795, chennisden wrote:there were no consensus TRs except like
NF and Sky later
the third consensus was, well, scum so.
at that point in the game, a lot of the general reads were that the players outside of the coalition were the predominant scumreads
my entire thought process was that since we know we're wrong, we need to start figuring out which of these scumreads are actually town so we can start figuring out WHICH townreads are wrong
it's a lot easier to fix reads when you start flipping people who aren't the alignment you thought they were rather than flipping people who were correct reads- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
- callforjudgement
-
callforjudgement Microprocessor
- callforjudgement
- Microprocessor
- Microprocessor
- Posts: 3972
- Joined: September 1, 2011
You gain a Night in that scenario. (And weren't you complaining about the lack of scum-directed kills?)In post 7765, chennisden wrote:This is what the town does.
"oh chemist is town
time to kill creature!"
how is that arena ever a misplay?
Re: coalition discussion, I have a suspicion that the first coalition may actually have been negative utility the way it played out. Coalitions that large give very little information (e.g. Titus's plan of "let's split the playerlist so that there's 2 scum in 1 half and 1 scum in the other half" gave no useful information all game, but even a more focused plan where the second coalition was almost the same as the first wouldn't have helped much). However, town tried to use the information for much more than it was actually worth, something that probably hurt town more than the actual information from the coalition helped.scum· scam · seam · team · term · tern · torn ·town- callforjudgement
-
callforjudgement Microprocessor
- callforjudgement
- Microprocessor
- Microprocessor
- Posts: 3972
- Joined: September 1, 2011
- RadiantCowbells
-
RadiantCowbells Smooth Criminal
- RadiantCowbells
- Smooth Criminal
- Smooth Criminal
- Posts: 70566
- Joined: February 24, 2013
- Contact:
I'm irritated I never got to mislynch ankamius and Eddie cane
I was looking forward to those
Yes, the first coalition was actively anti-town because we abused it to make it so. That's good scum play. Even when it cornered creature deepwolf we still had a deeper wolf etc2019 stats: Town WR 76.7%, overall WR 81.667%, 1 scum defeat involving a major mod error in lylo vs 8 scum wins.- RadiantCowbells
-
RadiantCowbells Smooth Criminal
- RadiantCowbells
- Smooth Criminal
- Smooth Criminal
- Posts: 70566
- Joined: February 24, 2013
- Contact:
Btw
I chose to reveal isolate and kill Titus as opposed to pocketing her and making her lose lylo for town because I didn't really want to make her even more of the reason that town lost.2019 stats: Town WR 76.7%, overall WR 81.667%, 1 scum defeat involving a major mod error in lylo vs 8 scum wins.- Something_Smart
-
Something_Smart HeSomewhat_Balanced
- Something_Smart
He- Somewhat_Balanced
- Somewhat_Balanced
- Posts: 22616
- Joined: November 17, 2015
- Pronoun: He
- Location: Rochester, New York
The thing is, this ignores the fact that not all players are equal. When scum can kill off the strongest townies N1-2-3, the remaining town players are by definition below average, even if the town was of average strength to start.In post 7764, chennisden wrote:Vanilla mountanious is NOT scum sided 4 scum of 17 players. There are papers on arxiv which prove that mountanious assuming lynches and nightkill every night is fair when around sqrt n of all players are scum. You guys just expect town to fail whenever possible which isn't really true and besides shouldn't be compensated for.
Also, I'd be really surprised if the EV for 13:4 mountainous was above, like, 40%. You can add significant power to 40% and have it still be balanced.Shame on a martyr claiming friends
From either perspective of &- RadiantCowbells
-
RadiantCowbells Smooth Criminal
- RadiantCowbells
- Smooth Criminal
- Smooth Criminal
- Posts: 70566
- Joined: February 24, 2013
- Contact:
-
-
Ankamius Survivor
- Survivor
- Survivor
- Posts: 21802
- Joined: May 9, 2011
- Location: Target Locked. Initiating Combat.
Towns really shouldn't be relying on mechanics to win the game when there is any possibility of scum having someone who can competently manipulate setups like this, it's almost always better to use mechanics to supplement day play when you have the ability to and I've seen scum win through mechanical play far more often than town unless the setup itself is hilariously broken in a way where scum just can't deal with it at all.
that's the primary example that I think town would've been far better off this game, I really think the coalition -> veto -> coalition play that ended up happening was a horrific mess that basically threw away town's chance of correcting their reads.- RadiantCowbells
-
RadiantCowbells Smooth Criminal
- RadiantCowbells
- Smooth Criminal
- Smooth Criminal
- Posts: 70566
- Joined: February 24, 2013
- Contact:
But we couldn't in this gameIn post 7807, Something_Smart wrote:The thing is, this ignores the fact that not all players are equal. When scum can kill off the strongest townies N1-2-3, the remaining town players are by definition below average, even if the town was of average strength to start.
We had to burn 2 nightkills on clears and we had 2 more nightkills so I needed to be the third towniest player to win while dodging double coalition and 1 of 4 cornering me
It took a very precisely laid out path to endgame2019 stats: Town WR 76.7%, overall WR 81.667%, 1 scum defeat involving a major mod error in lylo vs 8 scum wins.- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
- RadiantCowbells
-
RadiantCowbells Smooth Criminal
- RadiantCowbells
- Smooth Criminal
- Smooth Criminal
- Posts: 70566
- Joined: February 24, 2013
- Contact:
- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
but the setup is hilariously brokenIn post 7809, Ankamius wrote:Towns really shouldn't be relying on mechanics to win the game when there is any possibility of scum having someone who can competently manipulate setups like this, it's almost always better to use mechanics to supplement day play when you have the ability to and I've seen scum win through mechanical play far more often than town unless the setup itself is hilariously broken in a way where scum just can't deal with it at all.
that's the primary example that I think town would've been far better off this game, I really think the coalition -> veto -> coalition play that ended up happening was a horrific mess that basically threw away town's chance of correcting their reads.- RadiantCowbells
-
RadiantCowbells Smooth Criminal
- RadiantCowbells
- Smooth Criminal
- Smooth Criminal
- Posts: 70566
- Joined: February 24, 2013
- Contact:
Just remember that in this game
Only two nightkills sliced off of the top of the town pile
12 mechanics were used to remove people from the bottom of the town pile and the 13th/14th got cleared.
That's how you should look at balance btw. Look at mechanics that remove or clear townread players versus scumread ones.2019 stats: Town WR 76.7%, overall WR 81.667%, 1 scum defeat involving a major mod error in lylo vs 8 scum wins.- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
- Ether
-
Ether Lyrical Rampage
- Ether
- Lyrical Rampage
- Lyrical Rampage
- Posts: 4790
- Joined: July 24, 2006
- Location: New Jersey
- Contact:
I'm very confused how you got this? I have a simulator that puts it at like...around 20%. The EV Project doesn't list games that big, but the mountainous setups it does list also have EVs that are wildly lower than what you're getting.In post 7813, chennisden wrote:13:4 is veeeeery close to 50 percent. its over EV wise.
the recursion is UGLYAs I move my vote
Towards your wagon, town is taking note
It fills my head up and gets louder andLOUDER- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
- Something_Smart
-
Something_Smart HeSomewhat_Balanced
- Something_Smart
He- Somewhat_Balanced
- Somewhat_Balanced
- Posts: 22616
- Joined: November 17, 2015
- Pronoun: He
- Location: Rochester, New York
No I know, I'm just telling chennis why I think the powers should be better on average than Lynch.In post 7810, RadiantCowbells wrote:
But we couldn't in this gameIn post 7807, Something_Smart wrote:The thing is, this ignores the fact that not all players are equal. When scum can kill off the strongest townies N1-2-3, the remaining town players are by definition below average, even if the town was of average strength to start.
We had to burn 2 nightkills on clears and we had 2 more nightkills so I needed to be the third towniest player to win while dodging double coalition and 1 of 4 cornering me
It took a very precisely laid out path to endgame
Also my quick simulation, assuming random lynches, puts town's EV for 13:4 mountainous at 15%, so I'm curious what assumptions those papers are making to conclude that it's balanced.Shame on a martyr claiming friends
From either perspective of &- chennisden
-
chennisden Macho Pichu
- chennisden
- Macho Pichu
- Macho Pichu
- Posts: 8907
- Joined: February 11, 2019
- Location: sheltered in place
theres an arxiv paper ill link if i findIn post 7819, Ether wrote:
I'm very confused how you got this? I have a simulator that puts it at like...around 20%. The EV Project doesn't list games that big, but the mountainous setups it does list also have EVs that are wildly lower than what you're getting.In post 7813, chennisden wrote:13:4 is veeeeery close to 50 percent. its over EV wise.
the recursion is UGLY- RadiantCowbells
-
RadiantCowbells Smooth Criminal
- RadiantCowbells
- Smooth Criminal
- Smooth Criminal
- Posts: 70566
- Joined: February 24, 2013
- Contact:
I can't easily calculate EV for this game because it's really hard to calculate
But I really really don't think scum wins this a lot
If anyone disagrees they're welcome to do a guilty child rerun of the game2019 stats: Town WR 76.7%, overall WR 81.667%, 1 scum defeat involving a major mod error in lylo vs 8 scum wins.- Something_Smart
-
Something_Smart HeSomewhat_Balanced
- Something_Smart
He- Somewhat_Balanced
- Somewhat_Balanced
- Posts: 22616
- Joined: November 17, 2015
- Pronoun: He
- Location: Rochester, New York
I mean, I don't disagree with you RC. Veto and Reveal/Rank combo were the killers in my eyes.
How do you think the setup can be improved to be more balanced? That's what the postgame discussion should focus on, rather than pointing fingers and feeding egos.Shame on a martyr claiming friends
From either perspective of & - Something_Smart
Copyright © MafiaScum. All rights reserved.
- RadiantCowbells
- chennisden
- Something_Smart
- chennisden
- Ether
- chennisden
- chennisden
- chennisden
- RadiantCowbells
- chennisden
- chennisden
- RadiantCowbells
- chennisden
- RadiantCowbells
- RadiantCowbells
- Something_Smart
- RadiantCowbells
- RadiantCowbells
- callforjudgement
- callforjudgement
- chennisden
- chennisden