• Log InLog In
  • Register
Liquid`
Team Liquid Liquipedia
EST 19:33
CET 01:33
KST 09:33
  • Home
  • Forum
  • Calendar
  • Streams
  • Liquipedia
  • Features
  • Store
  • EPT
  • TL+
  • StarCraft 2
  • Brood War
  • Smash
  • Heroes
  • Counter-Strike
  • Overwatch
  • Liquibet
  • Fantasy StarCraft
  • TLPD
  • StarCraft 2
  • Brood War
  • Blogs
Forum Sidebar
Events/Features
News
Featured News
HomeStory Cup 28 - Info & Preview12Rongyi Cup S3 - Preview & Info3herO wins SC2 All-Star Invitational14SC2 All-Star Invitational: Tournament Preview5RSL Revival - 2025 Season Finals Preview8
Community News
Weekly Cups (Jan 26-Feb 1): herO, Clem, ByuN, Classic win1RSL Season 4 announced for March-April6Weekly Cups (Jan 19-25): Bunny, Trigger, MaxPax win3Weekly Cups (Jan 12-18): herO, MaxPax, Solar win0BSL Season 2025 - Full Overview and Conclusion8
StarCraft 2
General
StarCraft 2 Not at the Esports World Cup 2026 HomeStory Cup 28 - Info & Preview Weekly Cups (Jan 26-Feb 1): herO, Clem, ByuN, Classic win Weekly Cups (Jan 19-25): Bunny, Trigger, MaxPax win Oliveira Would Have Returned If EWC Continued
Tourneys
StarCraft Evolution League (SC Evo Biweekly) RSL Season 4 announced for March-April HomeStory Cup 28 $21,000 Rongyi Cup Season 3 announced (Jan 22-Feb 7) KSL Week 85
Strategy
Simple Questions Simple Answers
Custom Maps
[A] Starcraft Sound Mod
External Content
Mutation # 511 Temple of Rebirth The PondCast: SC2 News & Results Mutation # 510 Safety Violation Mutation # 509 Doomsday Report
Brood War
General
[ASL21] Potential Map Candidates Can someone share very abbreviated BW cliffnotes? Liquipedia.net NEEDS editors for Brood War BGH Auto Balance -> http://bghmmr.eu/ BW General Discussion
Tourneys
Azhi's Colosseum - Season 2 [Megathread] Daily Proleagues Small VOD Thread 2.0 [BSL21] Non-Korean Championship - Starts Jan 10
Strategy
Zealot bombing is no longer popular? Simple Questions, Simple Answers Current Meta Soma's 9 hatch build from ASL Game 2
Other Games
General Games
Battle Aces/David Kim RTS Megathread Nintendo Switch Thread Path of Exile Mobile Legends: Bang Bang Beyond All Reason
Dota 2
Official 'what is Dota anymore' discussion
League of Legends
Heroes of the Storm
Simple Questions, Simple Answers Heroes of the Storm 2.0
Hearthstone
Deck construction bug Heroes of StarCraft mini-set
TL Mafia
Mafia Game Mode Feedback/Ideas Vanilla Mini Mafia
Community
General
US Politics Mega-thread Things Aren’t Peaceful in Palestine European Politico-economics QA Mega-thread The Games Industry And ATVI Canadian Politics Mega-thread
Fan Clubs
The herO Fan Club! The IdrA Fan Club
Media & Entertainment
[Manga] One Piece Anime Discussion Thread
Sports
2024 - 2026 Football Thread
World Cup 2022
Tech Support
Computer Build, Upgrade & Buying Resource Thread
TL Community
The Automated Ban List
Blogs
Play, Watch, Drink: Esports …
TrAiDoS
My 2025 Magic: The Gathering…
DARKING
Life Update and thoughts.
FuDDx
How do archons sleep?
8882
Customize Sidebar...

Website Feedback

Closed Threads



Active: 1605 users

Fruity Mafia - Page 90

Forum Index > TL Mafia
Post a Reply
Prev 1 88 89 90
wherebugsgo
Profile Blog Joined February 2010
Japan10647 Posts
February 20 2013 21:56 GMT
#1781
On February 20 2013 23:08 risk.nuke wrote:
Yes the setup was townfavored, I'd say more then slightly but we can disagree on that, it's not even my issue since it was an experimental setup. As you both say the bigger crook was RNG, evidently town didn't even need some of their power roles. But that balance perspective should had been obvious to the host from the start and the fair thing to do would be to redo the RNG+ Show Spoiler +
I know some hosts do this more then others to ensure the teams are as fair as possible. I also know some hosts as yourself dislike doing it.
and there is a reason for that.

When a game is over, for a player there is nothing as frustrating as to look at a setup and see that it was heavily favored one way or another. More so if you were on the loosing side and put in a lot of effort. And favor doesn't just come from the setup, the players are also a factor. A hosts job is to make sure these frustrating scenarios doesn't happen. Knowing it was imbalanced and writing it of as bad luck because of the RNG isn't acceptable imo. You have a responsibility to your players to try to balance this game, and that should be prioritized before your code of never re-RNG.


How do you figure it was an experimental setup?

The only thing unusual about this setup was the way the town vig role worked. Everything else was a weaker version of normal.

What syllogism said is 100% true and was the basis of my balancing. Foolishness would agree with me on that, as we both talked it over for about an hour or so.

It's funny that you bring up player balance, since that reinforces my point that your team played with the idea that you were behind to begin with. That's a mental issue. You can't qualify the balance of teams simply by who shows up on them. What if there were two vets? Do I split them up? Or do I put them both on the same team? By your logic only splitting them up would keep the game balanced. If there were 3?

The reason I RNG setups is to stay impartial. Anything else, including rerolling, is subject to bias. If you feel you are incapable of winning against certain players, maybe you should improve or change your outlook on the game.
supersoft
Profile Joined April 2010
Germany3729 Posts
February 20 2013 22:15 GMT
#1782
if you don't random the games, they are way too easy to solve... Just saying... I killed Foolish in bureaucrazymafia mainly for balancing reasons and Palmar aswell...
It's so boring if you know exactly okay:
syllo, sandro, Palmar, wbg, Foolish, myself whatever
two of them are scum 100%.
That's just completely boring, like seriously...

Everyone of you can play well if he invests just a little bit more time, separates emotions and reads and tries to understand everyone else. And don't follow any stupid rules :D
my opinion.

And you see what happens, Marv and I went full retardmode against each other... That's because we automatically assume that one of the let's say usually more dedicated players has to be scum! I prefer the way wbg built his setup...
I mean I agree that dayvigs kind of suck, but I have to admit: Vivax was the perfect guy for the job and the only one who can get away with shooting someone like me...
His meta _this game_ would have justified that... Next game the window of opportunity closed... I dare you to shoot me, vivax....
AxleGreaser
Profile Joined November 2012
Australia1154 Posts
February 21 2013 02:09 GMT
#1783
On February 21 2013 06:56 wherebugsgo wrote:

The reason I RNG setups is to stay impartial. Anything else, including rerolling, is subject to bias. If you feel you are incapable of winning against certain players, maybe you should improve or change your outlook on the game.


I agree a lot with players lists being RNG'd....
I agree with the sentiments of the 2 of 4 setup being RNG'd and symmetric so that partial knowledge
http://wiki.mafiascum.net/index.php?title=2of4
Still leaves you with equi-probable information about whether or no the other optional roles exists

very few things are however 100%....

In this game (to the best of my understanding) once Vivax day shot it became very likely one of Vivax or small group of others were scum. Even though day viging your scum buddy might draw some cred, eventually gaming the role setup would have lynched Vivax. *

Bugs.... Given that the players need to be RNG'd so players cant game your choices...
should the roles have been RNG'd too...

(that could lead to very biased set-ups)

Do note that the actual 2of4 set-up has varying amounts of power roles for town hence they cant ALL be "balanced".
That set-ups designers made ungameability of the set-up the highest priority.
Partial knowledge of some roles makes no difference as to how likely the other possible roles are.


Rolling both the weakest player team and the weakest 2of4 setup would be sad. players would be sad

Hosts have lots of choices, I trust they either make them or they don't. If the hosts make good choices there will be good games, if the players make good choices there will be good play,

Me.... I just try to have fun any which way.

RNG is great everything ought be possible but equi-probable is another question.

and at that point I largely :|

I also ask people not to respond unless of course you know you know better that the response is good idea.

+ Show Spoiler +

There exists N different ways to RNG, yet not yield all control. Especially if the control (bias) is exerted before the RNG.
Consider a variable skill level player list and a 2 of 4 setup.
Algorithm 1a:
Sort the players into skill order sort the 2 of 4 setup into balance order.
Roll a player team. If the scum team is stacked, fiddle with the probability of the various 2 of 4 setups. Anything is however still possible ...
Algorithm 1b:
When rolling the player team do it in stages. If the scum team is becoming stacked, fiddle just a bit, to probably not become more so. Anything is however still possible ...
Algorithm 1c:
Fiddle with the probabilities of who gets which role, better town players with roles means they are likely to be better targeted, (consider vig Marv vs Vig Axle that might balance rather lot of player imbalance)
Algorithm 1d:
If you find you still need to mulligan it sometimes, RNG whether you really will mulligan this time... Then if you dont mulligan it check if it turned out as bad as you thought?

Algorithm 2:
Over time apply an algorithm to decide if your fiddles with the RNG are large enough.
A simple way (training algorithm) to determine that is you guess who will win. If you are right, next time fiddle harder, if your not fiddle less. If your setup gets gamed (successfully) fiddle less.

Algorithm 3:
Whatever you do don't tell people exactly how you do whatever it is you do....
Gaming the methods above is still possible. (but less effective)
If you don't tell people what you do and change it sometimes....
then actually playing mafia rather than playing the hosts set-up becomes the better play.

Algorithm 4:
Have someone like me post insanely complex set up procedures, such that players simply give up and
just play mafia against other players not the host. Glad to be of service?

Algorithm 5:
Do what hosts actually do....

Lastly have fun... I am...

+ Show Spoiler +
Reasons the algorithms are hard to game. Who knows exactly how the host actually ranks al the players in this game at this time? Who also knows what PMs the Host has about expected RL issues that will effect players in this game. Who knows how much of what algorithm this host could be bothered with this time....
Thus even guessing at the size of the fiddle/biases added to the RNG is hard.




Axle(OnlySomeTheoriesToAbsurdToBeConsidered)Greaser
wherebugsgo
Profile Blog Joined February 2010
Japan10647 Posts
Last Edited: 2013-02-21 06:32:10
February 21 2013 06:29 GMT
#1784
here's how I roll my setups:

Every player in order of signup is given a number, 1-n where n is the size of the game.

I then take the list of roles and give each role a number 1-n as well. Generally the vanillas are the first, followed by each of the blues, followed by the mafia.

Then I use a random number generator to create a randomized list of n numbers 1-n, without replacement. This list of numbers corresponds to the roles: they are handed out by index of the list to the players.

i.e. player 1 will be given the role in the first position of the list, player 2 will be given the role in the second, and so on and so forth.

AFAIK it's the way most hosts randomize their roles. If I'm not mistaken a lot of them will do pseudo-random setups where they randomize the roles and then handpick a few in order to balance the teams as they see fit.
Blazinghand *
Profile Blog Joined December 2010
United States25557 Posts
Last Edited: 2013-02-21 07:26:39
February 21 2013 07:25 GMT
#1785
The way I roll my setups is pretty much the same. I write out all the roles and all the players then select each one and roll a d20 (or what have you) and assign it to the player with that number. For games with 9 players I use a d10 and reroll if I get a 10 or a player who has already received a role. I haven't ever re-arranged roles based on player skill but if something weird happened like 3 players who are all first-time players are the 3 scum, I'd probably swap one of them for someone who has played before.

Sometimes this ends up with stronger setups for town or scum, and in all my GSL and British Empire games I have yet to roll up the coveted double-BoxeR setup, but I eagerly await that day. Life isn't always fair!

As an aside, 2of4 looks like it could use some BoxeRs to strengthen the weak 3 setups, then it'd be fine.
When you stare into the iCCup, the iCCup stares back.
TL+ Member
Toadesstern
Profile Blog Joined October 2008
Germany16350 Posts
February 21 2013 07:31 GMT
#1786
On February 21 2013 16:25 Blazinghand wrote:
The way I roll my setups is pretty much the same. I write out all the roles and all the players then select each one and roll a d20 (or what have you) and assign it to the player with that number. For games with 9 players I use a d10 and reroll if I get a 10 or a player who has already received a role. I haven't ever re-arranged roles based on player skill but if something weird happened like 3 players who are all first-time players are the 3 scum, I'd probably swap one of them for someone who has played before.

Sometimes this ends up with stronger setups for town or scum, and in all my GSL and British Empire games I have yet to roll up the coveted double-BoxeR setup, but I eagerly await that day. Life isn't always fair!

As an aside, 2of4 looks like it could use some BoxeRs to strengthen the weak 3 setups, then it'd be fine.

completly rerolling works better than swapping people in those scenarios imo. I'd really advise against picking people yourself as a host.

That being said I don't think it was heavily town favored like risk said but it was town favored somewhat, pretty much like my C9++ ended up being for similar reasons.
Bugs you told me I can't make two single-shot vigis because that confirms 2 townies instead of just 1 and I should have rather taken a single vig with 2 shots. You gundealer is the same thing. Artanis was a decent lynch option but after he gave the gun to syllo he confirmed himself and syllo. It's the same thing.
<Elem> >toad in charge of judging lewdness <Elem> how bad can it be <Elem> also wew, that is actually p lewd.
HiroPro
Profile Joined March 2012
United States2624 Posts
February 21 2013 07:34 GMT
#1787
There's a difference though. They were only "confirmed" because he shot a mafia. c9++ vigs are "confirmed" no matter what they shoot since mafia don't have any vigs.
wherebugsgo
Profile Blog Joined February 2010
Japan10647 Posts
Last Edited: 2013-02-21 08:25:57
February 21 2013 08:12 GMT
#1788
On February 21 2013 16:31 Toadesstern wrote:
Show nested quote +
On February 21 2013 16:25 Blazinghand wrote:
The way I roll my setups is pretty much the same. I write out all the roles and all the players then select each one and roll a d20 (or what have you) and assign it to the player with that number. For games with 9 players I use a d10 and reroll if I get a 10 or a player who has already received a role. I haven't ever re-arranged roles based on player skill but if something weird happened like 3 players who are all first-time players are the 3 scum, I'd probably swap one of them for someone who has played before.

Sometimes this ends up with stronger setups for town or scum, and in all my GSL and British Empire games I have yet to roll up the coveted double-BoxeR setup, but I eagerly await that day. Life isn't always fair!

As an aside, 2of4 looks like it could use some BoxeRs to strengthen the weak 3 setups, then it'd be fine.

completly rerolling works better than swapping people in those scenarios imo. I'd really advise against picking people yourself as a host.

That being said I don't think it was heavily town favored like risk said but it was town favored somewhat, pretty much like my C9++ ended up being for similar reasons.
Bugs you told me I can't make two single-shot vigis because that confirms 2 townies instead of just 1 and I should have rather taken a single vig with 2 shots. You gundealer is the same thing. Artanis was a decent lynch option but after he gave the gun to syllo he confirmed himself and syllo. It's the same thing.


1. your game was smaller
2. the gun dealer doesn't confirm two people at all, it's only true if he shoots a mafia (as HiroPro said)

If the guy with the gun shoots a townie you have no idea what the alignment is of either player until you kill one of them. I actually considered giving scum such a role, but with a different flavor (I'm not going to expound on this any further since I might use the role in a future game)

e:

also, there are a couple disadvantages to the gun dealer that do not exist for the vig

1. there is a predisposition for the guy to give a gun to one of a couple players in the game. In this game I would have been shocked if anyone other than marv/sandro/syllo had been given the gun.

2. there is likewise a predisposition for these players to be:
a. roleblocked by scum
b. shot by scum
c. protected by the JK (and thus roleblocked) this affects the pass too.
It's also very likely that one of them might be scum too.

3. In addition to all of these things it's still possible to roleblock the gun giver, just as one would directly roleblock the vig at night.

Obviously the advantage is that if the pass succeeds, the shot is more likely to succeed too. However, there are plenty of ways for this to go completely wrong and for the role to be completely nullified. You can RB two different people and stop the same role as scum. You can also shoot the receiver and the role can never be used. This is unlike a vigi, which can be used n1, so a shot isn't necessarily guaranteed to stop the action. If there are two vigis you can only roleblock one of them.
Acrofales
Profile Joined August 2010
Spain18207 Posts
February 21 2013 15:05 GMT
#1789
On February 21 2013 16:25 Blazinghand wrote:
The way I roll my setups is pretty much the same. I write out all the roles and all the players then select each one and roll a d20 (or what have you) and assign it to the player with that number. For games with 9 players I use a d10 and reroll if I get a 10 or a player who has already received a role. I haven't ever re-arranged roles based on player skill but if something weird happened like 3 players who are all first-time players are the 3 scum, I'd probably swap one of them for someone who has played before.

Sometimes this ends up with stronger setups for town or scum, and in all my GSL and British Empire games I have yet to roll up the coveted double-BoxeR setup, but I eagerly await that day. Life isn't always fair!

As an aside, 2of4 looks like it could use some BoxeRs to strengthen the weak 3 setups, then it'd be fine.

This sounds incredibly tedious, lol.

Here is how you can randomize without having to roll dice (or even use random.org):

1. Make an excel (or google docs) spreadsheet.
2. Place roles in column B of sheet 2 in whatever order you like
3. Place players on sheet 1 in whatever order you like (such as signup order)
4. Use http://www.jerrydallal.com/random/random_permutation.htm with the number of roles you have
5. Copypaste the outcome into column A of sheet 2
6. Go to Data -> Sort sheet by column A
7. Copypaste your randomized role list next to the players in sheet 1
8. PROFIT

Dandel Ion
Profile Joined November 2010
Austria17960 Posts
February 21 2013 15:23 GMT
#1790
Rolling dice is half the fun of hosting.

Just wait until I get green light on my RNG Mafia setup.
Just you wait.
A backwards poet writes inverse.
Artanis[Xp]
Profile Blog Joined December 2005
Netherlands12970 Posts
Last Edited: 2013-02-21 18:04:12
February 21 2013 18:03 GMT
#1791
On February 22 2013 00:23 Dandel Ion wrote:
Rolling dice is half the fun of hosting.

Just wait until I get green light on my RNG Mafia setup.
Just you wait.

Oh god, I have so many ideas for an RNG Mafia. You could have shots hit 80% of the time and livestream it as you open the RNG to decide the fate at the deadline. You could make it even more exciting if you announce who would or wouldn't be hit beforehand.
Dandel Ion
Profile Joined November 2010
Austria17960 Posts
February 21 2013 18:36 GMT
#1792
On February 22 2013 03:03 Artanis[Xp] wrote:
Show nested quote +
On February 22 2013 00:23 Dandel Ion wrote:
Rolling dice is half the fun of hosting.

Just wait until I get green light on my RNG Mafia setup.
Just you wait.

Oh god, I have so many ideas for an RNG Mafia. You could have shots hit 80% of the time and livestream it as you open the RNG to decide the fate at the deadline. You could make it even more exciting if you announce who would or wouldn't be hit beforehand.

Ooooh, an RNG soulmate.

We should combine our powers :D
A backwards poet writes inverse.
Prev 1 88 89 90
Please log in or register to reply.
Live Events Refresh
Next event in 27m
[ Submit Event ]
Live Streams
Refresh
StarCraft 2
PiGStarcraft355
Nathanias 140
ProTech135
Livibee 103
CosmosSc2 52
StarCraft: Brood War
Artosis 653
Shuttle 100
Hyuk 46
NaDa 21
Dota 2
monkeys_forever195
League of Legends
C9.Mang0305
Counter-Strike
taco 2
Super Smash Bros
AZ_Axe149
PPMD47
Other Games
summit1g8978
tarik_tv4109
Day[9].tv712
shahzam602
ViBE157
ToD146
Maynarde129
Mew2King58
Liquid`Ken9
JuggernautJason2
Organizations
Other Games
gamesdonequick909
StarCraft 2
Blizzard YouTube
StarCraft: Brood War
BSLTrovo
sctven
[ Show 17 non-featured ]
StarCraft 2
• Hupsaiya 70
• davetesta23
• Kozan
• LaughNgamezSOOP
• sooper7s
• AfreecaTV YouTube
• intothetv
• Migwel
• IndyKCrew
StarCraft: Brood War
• RayReign 20
• STPLYoutube
• ZZZeroYoutube
• BSLYoutube
Dota 2
• masondota21150
Other Games
• imaqtpie1632
• Day9tv712
• Shiphtur249
Upcoming Events
PiGosaur Cup
27m
WardiTV Invitational
11h 27m
Replay Cast
23h 27m
The PondCast
1d 9h
WardiTV Invitational
1d 11h
Replay Cast
1d 23h
RongYI Cup
3 days
herO vs Maru
uThermal 2v2 Circuit
4 days
Replay Cast
5 days
Wardi Open
5 days
[ Show More ]
Monday Night Weeklies
5 days
Sparkling Tuna Cup
6 days
Liquipedia Results

Completed

Proleague 2026-02-02
HSC XXVIII
Underdog Cup #3

Ongoing

CSL 2025 WINTER (S19)
KCM Race Survival 2026 Season 1
Acropolis #4 - TS4
Rongyi Cup S3
Nations Cup 2026
IEM Kraków 2026
BLAST Bounty Winter 2026
BLAST Bounty Winter Qual
eXTREMESLAND 2025
SL Budapest Major 2025
ESL Impact League Season 8

Upcoming

Escore Tournament S1: W7
Escore Tournament S1: W8
Acropolis #4
IPSL Spring 2026
HSC XXIX
uThermal 2v2 2026 Main Event
Bellum Gens Elite Stara Zagora 2026
RSL Revival: Season 4
LiuLi Cup: 2025 Grand Finals
IEM Rio 2026
PGL Bucharest 2026
Stake Ranked Episode 1
BLAST Open Spring 2026
ESL Pro League Season 23
ESL Pro League Season 23
PGL Cluj-Napoca 2026
TLPD

1. ByuN
2. TY
3. Dark
4. Solar
5. Stats
6. Nerchio
7. sOs
8. soO
9. INnoVation
10. Elazer
1. Rain
2. Flash
3. EffOrt
4. Last
5. Bisu
6. Soulkey
7. Mini
8. Sharp
Sidebar Settings...

Advertising | Privacy Policy | Terms Of Use | Contact Us

Original banner artwork: Jim Warren
The contents of this webpage are copyright © 2026 TLnet. All Rights Reserved.