As far as number of trials to determine significance. I think we need to run a power analysisfor that, if you want to get technical.
It should be noted I'm a neurobiologist that works in qualitative data, so stats ae the bane of my existance, and I haven't had a real math class since high school.
I think a good idea might be to run these setups that we want run a lot more during marathon days. A rough gauge of balance at best, it might help us make a decision.
(11:26:07 PM) thesheamuffin: I'm counting gurgi because I would probably make out with him if I were drunk enough
On a serious note, I'll try to run at least one flagged setup, but I'd rather mess around with Chosen while it's still popular. (Lord Mith, I'd be curious to see what you were getting at in 141 about abnormal but open setups.) If I play, just keep in mind that any games in which I am scum should not be counted for balance purposes.
Actually...maybe we could hold some kind of pseudomarathon day of our own that keeps a more structured queue and only runs experimental open setups? We get a more concentrated audience, and if it's a speed day that otherwise wouldn't have occurred, everyone's happy.
As I move my vote
Towards your wagon, town is taking note
It fills my head up and gets louder and
Ether wrote:Standard nightless games found by searching for "Nightless" in Little Italy and New York:
Open 6 (4:8; town win) Open 19 (4:8; town win) Open 79 (4:7; I think this was the game Farside was talking about; town win)
So I'm really not getting the need to further handicap the scum.
EDIT: Open 79 was 4:7, not 4:8.
Ether if you read through those game one had scum lynched on day 2, one had scum lynched on day 1, and one even had scum lynched back to back on days 1 & 2.
This really points to very poor performance by the scum in these games. They aren't blending in well and they are throwing their weight around. Hell in a 4:7 game they only need two townies to wagon with them on Day 1 and they still wound up getting one of their own lynched.
Can I just ask for my own interest- is there some way in Excel/other programs to calculate likely win probabilities for various setups extremely easily?
Currently modding Mole Mafia: http://www.mafiascum.net/forum/viewtopic.php?f=23&t=20529
Feel free to PM me to be ready in case I need a replacement.
So at this rate we might finish before a solar apocolypse.
I propose we nominate an Open Game Steering Committee in order to move the Open Game Certification Group in the right direction.
On a more serious note. I don't think we need to dwell on vanilla Nightless this long. Our sample size is obviously small, and I believe it to be rather craptastic as far as representation of a normal distribution. However, the game is clearly not broken, and we are not setting these rules in stone for all eternity. Can we please agree on a set of vanilla nightless games and move on to set ups that might actually involve breaking strategies. If the ratios that get run are clearly not fun they can be altered at any time.
Masons and Monks is a 2 v 2 v 8 I ask that we lump these games together for a separate discussion because I think they all need to have crosskill issues resolved.
I still think those lower ratios are ridiculous and shouldn't go through, but if I'm outvoted, whatever.
What do you mean by "resolved," Shafted? Block the first crosskill on each group, or give the last survivor of a scumgroup nightkill immunity? I think multiballs will always be swingy. I feel kind of out of my depth here; I'd be curious to see some simulations (noting how frequent crosskills and prisoners' dilemmas are).
Speaking of prisoner's dilemmas, is there any way to resolve symmetrical ones without a heavy chance of a town win that didn't involve the town failing to catch scum up to that point?
As I move my vote
Towards your wagon, town is taking note
It fills my head up and gets louder and
Ether wrote:Speaking of prisoner's dilemmas, is there any way to resolve symmetrical ones without a heavy chance of a town win that didn't involve the town failing to catch scum up to that point?
Shootout Rule:
If night starts with three players alive, all of whom are on different teams, a gun battle breaks out which kills everyone (including the poor innocent townie who gets caught in the crossfire). The game is a draw.
#mafia@irc.globalgamers.net
"Xyl was completely berserk" -dramonic
"Xyl's ruthless policy lynching won the game." -Vi
Ether wrote:What do you mean by "resolved," Shafted? Block the first crosskill on each group, or give the last survivor of a scumgroup nightkill immunity? I think multiballs will always be swingy. I feel kind of out of my depth here; I'd be curious to see some simulations (noting how frequent crosskills and prisoners' dilemmas are).
Yeah 2 v 2 v 8 is always swingy, but by resolved I just mean discussed. I don't think the town can win such set ups without at least one crosskill, but more than one and it becomes a somewhat easy win for town. I haven't modeled out how frequently to expect them without any sort of night immunity. But I was hoping to discuss if there were certain simple tweaks that could be made to hit a "preferred" number of crosskills per game.
So basic questions are:
How many crosskills should one expect with no NK immunity of any sort?
What are the EV's given each number of crosskills?
Is there a way to shift the expected number of crosskills to the most balanced EV without totally convoluting the set ups?
for town to win 2:2:8 (daystart) with no crosskills, but the above assumes an almost perfect town--the only thing it did wrong up there was alternate. One mislynch while both scumgroups are still alive causes them to lose majority. Town can spare a mislynch it completely eliminates one source of kills early instead of alternating, or a scumgroup crosskills/fails a crosskill but doesn't get a kill at all that night as a result.
EVs are probably a more accurate reflection of balance in games with more scumgroups.
Ether wrote:Block the first crosskill on each group,
Actually, this could solve at least some prisoner's dilemma situations.
As I move my vote
Towards your wagon, town is taking note
It fills my head up and gets louder and
.