Back to Epicmafia

Comp Setup Catalogue [Discussion]

over 10 years

Comp Setup Catalogue

Original Thread


Thank you, Connor, for your contributions to the language used in the OP and formatting. Thank you, itg123, for making visual represntations of the setups in 2 sheets.

This thread is to be used for discussion of competitive setups, both currently recognised in the catalogue and new potential creations from the community, appropriate for competitive play in both the Training and Comp Lobbies. This is a resource for recent trophy winners to discuss or question the validity of their desired setups for use in the next competition, as well as the place for any user to add, challenge or question any changes to the catalogue in its current state.

If you see anything previously comped or with a vast number of plays that has missed, please point it out. This is a pretty massive undertaking so any community support is greatly appreciated and encouraged. Likewise, if something is added and shouldn’t be – please let us know and we’ll take it down. Currently, I’m not adding any setups that are old and with a few plays unless specifically requested by the community.

--

There are the flexible guidelines of attributes commonly seen in competitively viable setups, and it certainly isn’t law, however it encompasses almost everything used currently in competitive play.

  • 7-10 Players
  • Town:Mafia win ratio does not exceed 42:57 or 55:45 (after ~200 plays)
  • Has no 3rd parties unless the third party wins with either town/maf exclusively. Ex. Traitor wins with Mafia, Amnesiac wins with either town or mafia in Suppa Magical Grey Scales
  • Cannot utilize roles or role combinations that can be deemed gamebreaking. (Doc joints on DYLG, War on Terror, etc)

Other Rules related to setup selection

  • A setup can only be comped once per three rounds. So if Jan 2.0 is comped Round 1, it can not be comped again until Round 4.

  • This includes similar variants of said setups. So if someone comps SDS w/ML no Whisper Round 1, they cannot comp the version with a Miller and one less blue the following round. Either setup can be selected for Round 4, but not both.

  • Setups with similar variants will be allowed/disallowed at mod discretion.

  • Any setup which becomes recognized as poor for comp will be taken down at moderator discretion.

--

It is recommended and encouraged that all new setups to be comped are linked here so we can review and discuss their viability. If a setup is at one point, deemed unviable due to a lack of balance in point spread, if the balance is restored via playing in red hearts and the meta on it changes, the setup can be re-submitted to be listed as comp viable. But if it slips far past the accepted marks, then it will be taken down after community and moderator deliberation.

As new setups arrive, this is the perfect place to discuss their balance as they are play tested, and it would be assumed that a setup over the course of theory discussion and playtesting of about 200 plays, it's balance and viability can be determined and added as necessary. It is recommended and encouraged that all new setups to be comped are linked here so we can review and discuss their viability. If a setup is at one point, deemed unviable due to a lack of balance in point spread, if the balance is restored via playing in red hearts and the meta on it changes, the setup can be re-submitted to be listed as comp viable."

As this thread is new, this is a pretty massive undertaking so any community support is greatly appreciated and encouraged. Likewise, if something is added and shouldn't be – please let us know and we’ll take it down."

itg123's Visual Version ||| Popular ||| New/Other |||

over 10 years

Anthony says

fiddler and silencer both should be banned from any competitive setup


Make a petition and see if it gains traction. I'd sign it.

Fiddler/Silencer, although they do add a certain strategic element, it goes against a lot of the conventions: A role which disallows people from participating in a game seems sort of counter-intuitive.

Most of the setups using either of them could be replaced with a traitor or a nilla and probably not be too functionally different. BAC, YYC are the only prominent setups I can think of right now that feature either.
over 10 years
fiddler and silencer both should be banned from any competitive setup
over 10 years

Giga says

That bug with the points has been happening for 9 or 10 months lol


Is he aware of it at least or no?

Itg, the idea is that there shouldn't be a ban list - it's much easier to make a list of what makes a good competitive setup as opposed to what doesn't. This adds for more flexibility too. As Pranay said, there are millions of possibilities. If someone thinks they can make a competitive Cthulu setup, then go for it.

I think those versions of A&D are very similar to others previously made, the Sheriff one is exactly the same as Nero's only with a miller if I recall correctly.
over 10 years
why are we trying to avoid a banlist? troll roles like Cthulu and OP Roles like Santa were clearly made for Sandbox and not Comp.

With regards to A&D, I made two versions of hopefully more townsided setups.

https://epicmafia.com/setup/1017015
https://epicmafia.com/setup/1017001
over 10 years
that's a great idea, itg, except it's the exact same thing we're trying to avoid
over 10 years
I suggest making a few guidelines for prospective setup makers, a list of roles that would be banned competitively, and prizes for people who can devise new setups utilising the underused roles.
over 10 years

Steven says

Chances of being town: 6/9
Expected number of hearts as town: 33
Chances of winning as town: 41%
Expected number of wins as town: 14
Points as town: 71
Expected number of points as town: 994

Chances of being mafia: 3/9
Expected number of hearts as mafia: 17
Chances of winning as mafia: 58%
Expected number of wins as mafia: 10
Points as mafia: 51
Expected number of points as town: 510

Total expected number of points by playing A&D: 1506

------------------

Chances of winning: 50%
Expected number of hearts won: 25
Points for winning: 60

Total expected number of points playing a balanced setup: 1500

-----

That's actually interesting, lucid's formula was probably made so that mafia/town would be balanced no matter what under a 2:1 town to mafia ratio/the ratio of the setup (the 6 point difference is likely due to rounding done by me). I always thought it was meant to be balanced over a 1:1 town to mafia ratio, meaning mafiasided setups would give you a higher yield in points in the long run (because you are town more often, so you get the extra points more often). However apparently that's not true for the average person. Unless I messed up with the math.

I initially did these calculations to show why severely skewed setups are unfair but I suppose they are not, point wise at least. Of course, they are still unfair because they lead to easy wins for one side, and if you're particularly above average at being the disadvantaged side.


Do your calculations with a setup considered wildly unbalanced, moreso than A&D. You should find it comes out about the same.
over 10 years
That bug with the points has been happening for 9 or 10 months lol
over 10 years
you're right my bad
over 10 years

Happeh says

It also favours those are lucky enough to roll the favoured side more often.


It actually doesn't. At least not unless you are able to increase your win% for that side compared to the average as well (but you can also do this for balanced setups regardless). Still, even if you roll town all 50 games it's not going to help your score at all unless you do substantially better than everyone else as town. If you dont believe me just do the math
deletedover 10 years
you can mail him; he rarely frequents the forums.
over 10 years
Gotcha, would you mind emailing lucid about it or posting it in the bug report forum? I would but I'm about to step out for a bit. Thanks Nero. If not I can do it later.
deletedover 10 years
it's always the case so it's almost definitely a bug. you only really need 33 wins for 2K now due to this.
over 10 years

Nero says

on a related note, there's been a bug lately where players get 1 more point that what's displayed on the setup page. so one town + scum win on a&d gets you 124 points instead of the 122 displayed.


Yeah, I've been noticing this too. I assumed it was because the setup page updates faster so it shows what the amount should be, whilst the setup mechanically lags behind? I could be wrong. I'd write to lucid about it and see what he says or if he knows about it.
deletedover 10 years
on a related note, there's been a bug lately where players get 1 more point that what's displayed on the setup page. so one town + scum win on a&d gets you 124 points instead of the 122 displayed.
over 10 years

Kyle says

if a&d is to be removed, then frontier justice needs to go too. it's just as bad if not worse.


Why is it bad in your opinion, Kyle?
deletedover 10 years
if a&d is to be removed, then frontier justice needs to go too. it's just as bad if not worse.
over 10 years
It also favours those are lucky enough to roll the favoured side more often.
over 10 years

Edark says

its time for people to stop posting

over 10 years
its time for people to stop posting
over 10 years
Chances of being town: 6/9
Expected number of hearts as town: 33
Chances of winning as town: 41%
Expected number of wins as town: 14
Points as town: 71
Expected number of points as town: 994

Chances of being mafia: 3/9
Expected number of hearts as mafia: 17
Chances of winning as mafia: 58%
Expected number of wins as mafia: 10
Points as mafia: 51
Expected number of points as town: 510

Total expected number of points by playing A&D: 1506

------------------

Chances of winning: 50%
Expected number of hearts won: 25
Points for winning: 60

Total expected number of points playing a balanced setup: 1500

-----

That's actually interesting, lucid's formula was probably made so that mafia/town would be balanced no matter what under a 2:1 town to mafia ratio/the ratio of the setup (the 6 point difference is likely due to rounding done by me). I always thought it was meant to be balanced over a 1:1 town to mafia ratio, meaning mafiasided setups would give you a higher yield in points in the long run (because you are town more often, so you get the extra points more often). However apparently that's not true for the average person. Unless I messed up with the math.

I initially did these calculations to show why severely skewed setups are unfair but I suppose they are not, point wise at least. Of course, they are still unfair because they lead to easy wins for one side, and if you're particularly above average at being the disadvantaged side.
over 10 years
Also, I would wager to say itg321 has included some undesirable setups in his "Visual Version"
over 10 years
The autonomy of an unbalanced setup is removed from the players on the side least likely to win through sheer difficulty. This is especially true in town-sided setups because more often than not it is impossible for mafia to make town make a mistake. That's why the townsided ceiling is lower.
deletedover 10 years
people need to stop caring so much about setups. you're all turning into brakes.
over 10 years

Steven says


Happeh says

The list is a resource of competitive setups, which A&D doesn't fit the criteria of. That doesn't stop it from getting comped.


I think if a setup doesn't fit the criteria of "being competitive" it shouldn't be allowed to be comped.


Well as of right now, that's not how it works, and if it were to ever change, good. I don't think it should define the resource list.

Steve says

Everyone's definition of a competitive setup is slightly different


"Uncompetitive game aspects (or strategies) are those that take away autonomy (control of the game's events), take it out of the hand's of player's decisions-- and do so to a degree that can be considered uncompetitive.

This can be luck-based, but doesn't have to be. Most uncompetitive strategies that are banned usually have a high appeal to luck.

While there is always luck involved in Mafia, the problem is the degree to which control is taken away from the player. Removal of autonomy is the key to an competitive or uncompetitive ruling."


Steven says

though so we should go by the official (the moderator's) definition of a competitively acceptable setup (anything that is allowed in comp)


I would argue against this as they allowed the 2gs setup to get comped this round.

Steven says

in supposedly objective threads such as this one, whether or not you disagree with their decision on a setup (and I am a firm believer of 70 being the max a setup should give for town or mafia on 2 party setups, but like I said before that discussion seems to be a different one than this thread's purpose?)


What I'm proposing follows the definition above.