EPTI usually balance games based on how many mislynches and/or how quickly the town can "figure" things out. That means just assuming worst case scenarios for Scum like Detectives getting off a check->living->reveal-protected, Vigis shooting scum Night 1 and Town lynching Scum Day 1. Of course this is unlikely to happen but if you keep iterating over scenarios and find near unwinnable positions for Scum balance may need to change. Likewise, you do the same thing for Town and try to find a medium. As an aside Vigilantes and SKs are among the most difficult roles to balance because they can shoot anyone at anytime unless compulsive. If they end up in the hands of crazy players you might as well forget about predicting outcomes anyway.
Active List of Mafia Games - Page 143
| Forum Index > TL Mafia |
|
Ace
United States16096 Posts
I usually balance games based on how many mislynches and/or how quickly the town can "figure" things out. That means just assuming worst case scenarios for Scum like Detectives getting off a check->living->reveal-protected, Vigis shooting scum Night 1 and Town lynching Scum Day 1. Of course this is unlikely to happen but if you keep iterating over scenarios and find near unwinnable positions for Scum balance may need to change. Likewise, you do the same thing for Town and try to find a medium. As an aside Vigilantes and SKs are among the most difficult roles to balance because they can shoot anyone at anytime unless compulsive. If they end up in the hands of crazy players you might as well forget about predicting outcomes anyway. | ||
|
Acrofales
Spain18344 Posts
On February 21 2013 17:49 Ace wrote: At its core all a detective does is flip alignment. Trackers/Watchers don't, all they tell you is where someone went and depending on if protown visiting night roles are known you can run into "good" problems where they can't reveal information without risk. That's usually good for gameplay. However, I'd say watchers are more powerful (about on par with a DT) due to who they target: it is generally easier to target powerful townies at a risk of getting shot, over guessing not only who ls mafia, but also which member of mafia will carry the KP. I usually balance games based on how many mislynches and/or how quickly the town can "figure" things out. That means just assuming worst case scenarios for Scum like Detectives getting off a check->living->reveal-protected, Vigis shooting scum Night 1 and Town lynching Scum Day 1. Of course this is unlikely to happen but if you keep iterating over scenarios and find near unwinnable positions for Scum balance may need to change. Likewise, you do the same thing for Town and try to find a medium. As an aside Vigilantes and SKs are among the most difficult roles to balance because they can shoot anyone at anytime unless compulsive. If they end up in the hands of crazy players you might as well forget about predicting outcomes anyway. That's great info. I have always worked under the assumption that worst-case for scum is a no-brainer and not really worth puzzling about. I don't have much experience balancing games, but I have always worked from the town perspective: how many mistakes can town make (assuming everything that shoots shoots townies). I guess a good indicator is worst-case for both teams, and if they are the same length and about the same number of disasters/mistakes happened, then it's an indicator it's balanced? | ||
|
iGrok
United States5142 Posts
On February 21 2013 17:11 DoctorHelvetica wrote: I mean more in the sense of exactly what number of DT's/Medics/Vets/etc. work. I try to balance the game so that the town on a perfect night can block all mafia KP, but never more than that. 1 investigative role for every 3 mafia, something like that. I wonder if it would be worth compiling this kind of information for normal games and how they affect the balance between town and mafia. For instance: How does a game with watcher/tracker versus detectives actually affect balance? We might all have subjective ideas about what is better for town but I'm interested in the hard data on that. It wouldn't be too hard to compile that data actually, maybe I'll do it in my spare time despite my fear of numbers. EDIT - Just saw the database, lol nevermind The thing is, proper balance is determjned by the community. What works for mafiascum doesnt work on tl, and what works here probably wont work on a third forum | ||
|
Sn0_Man
Tebellong44238 Posts
On February 21 2013 22:23 Acrofales wrote:I don't have much experience balancing games, but I have always worked from the town perspective: how many mistakes can town make (assuming everything that shoots shoots townies). I'd be interested to hear how that thought process went in relation to last game. PS: In games with a godfather/framer, I'd say that Watcher is measurably weaker than cop since Mafia can Godfather the kill and/or frame a townie with the kill, whereas the chances of framing the DT target or the DT targetting the godfather are probably somewhat lower. Obviously non-0, but somewhat lower. | ||
|
Acrofales
Spain18344 Posts
On February 22 2013 03:19 Sn0_Man wrote: I'd be interested to hear how that thought process went in relation to last game. PS: In games with a godfather/framer, I'd say that Watcher is measurably weaker than cop since Mafia can Godfather the kill and/or frame a townie with the kill, whereas the chances of framing the DT target or the DT targetting the godfather are probably somewhat lower. Obviously non-0, but somewhat lower. Was a 13 player, 3 scum, 1 SK game. SKs are always tricky, but the worst-case scenario left town with 2 mislynches before lylo. All the SK math is somewhere in the obs QT. On D2 the only advantage of lynching the SK over lynching a townie is that you are guaranteed to not have a kingmaker scenario, and you are not dependent on the SK ever shooting scum. You also entered mylo, rather than lylo. However, even if you had lynched a townie and the SK had killed a townie, town still had a chance at winning. So, regardless of what happened, the worst-case was 2 mislynches into lylo/mylo. After talking to Foolishness just now, I am further convinced that this was not a bad balance at all. What went wrong for town was the lack of scumhunting anywhere after N1, not the initial balance, but there's plenty of discussion about that in the postgame. | ||
|
Kurumi
Poland6130 Posts
 | ||
|
Foolishness
United States3044 Posts
On February 21 2013 17:06 Ace wrote: TL had so many good/weird/unique personalities that even a "bad" game could be enjoyable because the player interactions were just too hilarious. As for balance just start out with an idea of how long you want the game to last. If you have it where the town can mislynch 3 times at most before losing then it's not impossible to balance roles around that ideal. This is how I balance my games too. For smaller games, town should mislynch 3 times before they lose. Anymore than that and it's too hard for the mafia. For bigger games you can up the lynch count (talking 30 players or more here). | ||
|
wherebugsgo
Japan10647 Posts
Basically the problem with the watcher is that if you point it at the towniest guy in the game you will almost assuredly catch scum on night 1 (since he's so likely to get shot). Thus it's only really good for bigger games because of how easy it is to use (but even in that case it can still cause problems) It's kind of similar to the problem that listchecks have. Trackers don't really have this problem. | ||
|
wherebugsgo
Japan10647 Posts
[M][T] Smurfville Mafia? :D requirement will be that everyone is a smurf. identities will be revealed postgame. No claiming identities allowed in-game. | ||
|
Promethelax
Canada7089 Posts
On February 22 2013 22:22 wherebugsgo wrote: hey foolishness, mind putting me up for [M][T] Smurfville Mafia? :D you mean LX ii | ||
|
marvellosity
United Kingdom36161 Posts
On February 22 2013 22:22 wherebugsgo wrote: hey foolishness, mind putting me up for [M][T] Smurfville Mafia? :D requirement will be that everyone is a smurf. identities will be revealed postgame. No claiming identities allowed in-game. like a vanilla version of Your Clothes Mafia, I guess. | ||
|
wherebugsgo
Japan10647 Posts
On February 22 2013 22:27 marvellosity wrote: like a vanilla version of Your Clothes Mafia, I guess. wtf is that | ||
|
marvellosity
United Kingdom36161 Posts
Your Clothes, Give them to me Mafia | ||
|
wherebugsgo
Japan10647 Posts
WTF how did I miss that game? That looks like a really cool setup. | ||
|
marvellosity
United Kingdom36161 Posts
| ||
|
BioSC
United States636 Posts
| ||
|
Blazinghand
United States25560 Posts
| ||
|
Oatsmaster
United States16628 Posts
| ||
|
DoctorHelvetica
United States15034 Posts
 I've always wanted to play with a conversion mechanic | ||
|
Oatsmaster
United States16628 Posts
Host a mini, probably easier to balance that. :D | ||
| ||
