[phpBB Debug] PHP Warning: in file [ROOT]/ext/alfredoramos/seometadata/event/listener.php on line 114: Undefined array key 10618911 [phpBB Debug] PHP Warning: in file [ROOT]/ext/alfredoramos/seometadata/event/listener.php on line 114: Trying to access array offset on value of type null Marathon Weekend Open Setup Group - Mafiascum.net
Post
Post #42 (isolation #3) » Thu Jan 24, 2019 8:53 am
Postby Jingle »
When you get a chance tw could you throw a list of setups people have requested into the OP so we can start organizing times for individual setups people want to make sure they catch?
Also, can we add whatever setup wins the challenge thread thing in your sig to the list.
Post
Post #69 (isolation #10) » Fri Jan 25, 2019 10:44 am
Postby Jingle »
^Do you have any setups in mind that could use more testing?
In post 64, the worst wrote:I actually don't hate the idea too much.
Maybe a win is just a redflip on d1 but I feel like the nsg/rc thing is specifically intended to townsided heavily above mechanical ev since they're both Overpowered....
I too would mod/play it
I thought your addition was a they win if scum lynch or accurate guess. I wouldn't like to be town with the win completely out of my hands in that setup.
On the other hand:
Town has a 40% EV (which should be bumped by the Paragon's influence because they want to lynch scum), Mafia is dropped from 60% based on the chance of Paragon guessing correctly, so it shouldn't be too much more than 50%. Paragons have a 25% chance on a scumlynch and a 16.7% chance on a town lynch, which is low but accepted because they're supposed to be good at this and also 3p.
I think it might be more fair to give the Paragons two guesses after a scum lynch. That moves the EVs to town 40%, scum 50%, paragon 30%, which seems about right to me (the Paragons now have a very strong incentive to get people to lynch scum D1).
I also think it might be interesting to not give the Paragons a vote D1, so that town's 40% win chance is entirely in their own hands.
Your numbers have to be off (paragon win increasing would by definition decrease scum win %). I'll look at it in a little bit to see where.
Definitely like the idea of stumping the Paragons, which means there could be any number of them for no balance change. I don't think the second chance is necessary for them (it's supposed to be hard, I think) but it doesn't ruin the setup. Hell, you could even do a guess till you're right sort of thing where winning is completely subjective and the point remains as it is.
Post
Post #74 (isolation #12) » Fri Jan 25, 2019 11:49 am
Postby Jingle »
Looks like your initial numbers are right. In the second group, scum should drop to 40%. Which... Seems pretty fair to me, honestly.
Edit: Paragon should also jump to 43.3%, I forgot to take into account the increase in the Town and Paragon winning rate. Which imo is too high for what's supposed to be a difficult role, but it completely fair if you don't want to make the Paragons have a harder time than everone else.
Post
Post #76 (isolation #13) » Fri Jan 25, 2019 12:30 pm
Postby Jingle »
?
They have twice as many chances to win when town has been lynched (2/6 compared to 1/6) which means scum has 4/6 instead of 5/6 chances to win in that case.
When scum is lynched they have 1/4 + 1/3 (which is ~58%).
Add in the weights (.4 for a scum lynch and .6 for a town lynch) and you get .4 v .4 v.43 Where the overlap of 23% is a joint town/Paragon win.
Post
Post #97 (isolation #18) » Thu Feb 14, 2019 10:54 am
Postby Jingle »
In post 3415, jjh927 wrote:What next for jjh927? Well, I have a dumb open setup idea that I didn't publish in case people mod metaed me for this game;
Spoiler: Currently Unnamed Open Setup Idea
2 mafia, who choose to be 1-shot framers, miller, or godfather. 7 town, who are either 1-shot framers, millers, or godfathers, with distribution depending on what the mafia picked to make 3 of each total in the game.+ 1 town rolecop and 1 town regular cop
Also a couple of ideas that are less effort than this game but still theme games. But I really shouldn't host any of them until I finish my dissertation.
Any guesses on the number of beans in the vegan bean chilli?
JJ wants to run this. Volunteer and pick a time, peeps!
Post
Post #108 (isolation #19) » Sun Feb 17, 2019 8:33 am
Postby Jingle »
I assume signups are open to everyone for as long as the thread remains open. If you have setups you specifically want tried post them (or links to them) here, otherwise head to the marathon queue!
Post
Post #113 (isolation #20) » Mon Feb 18, 2019 8:11 pm
Postby Jingle »
Overall, I think the setups tried went well.
8-ball and Marked for death seemed especially popular.
BoP was probably my favorite, although I may be biased.
Tough on crime (and probably every other setup with multiple vote mechanics) was not well suited to marathons and suffered from bad moderation on my part, but I'd like to see it in the open queue in the future.
Does anyone have thoughts on what setups worked/didn't?
In post 116, the worst wrote:I do think that the 6:9 half mast madness was a success, in several ways.
okay maybe it was impressive that we got that many players together for that long, but I have to point this out for anyone who's considering something similar in the future:
Never ever ever EVER
EVER
run a 15-player nightless marathon game
unless you really want to.
FTFY. And if I hadn't gotten exhausted and basically laid down in the middle of the road on D2 I wouldn't have had any problems with it.
Post
Post #122 (isolation #22) » Sat Feb 23, 2019 6:43 am
Postby Jingle »
In post 116, the worst wrote:My thoughts are that there's no harm in leaving this thread open for next time. If people feel that we were a little underconmunicated, that might be something I consider for next time - I really don't know how popular discord is on MS, but I'd be open to setting up a server for collaboration & numbers-wrangling in the future. If people are keen to push more into that 11-13p sweet spot, I think live external communication is a smart way to do it.
Honestly? I think perhaps changing the group to be non marathon exclusive might help, because to be honest most of the larger setups proposed here aren't well suited to marathons. I think the marathon aspect is super fun, but keeping the thread open year round and using it as a group where prospective players and mods can communicate about what games they'd like to see run would be even better. And besides, a lot of the player list for marathons might also be interested in non marathons, be they blitz or normal.
In post 117, T-Bone wrote:I'd be interested in seeing more too. The Open Queue doesn't have a lot of Mods waiting right now. If you can get a dedicated core of players to play through some opens, we can see some of these set ups in action. I'm very interested in expanding on the approved open set-up list this year. One of the ways we can look at doing that is testing certain set ups and playing them a couple of times.
The numbers are VERY off on that setup, btw. Not sure how I managed it (simplest guess would be my original simulation allowed players to simply lynch the same person over and over again), but it actually had like an 85% town EV as run. 9v4 has ~66%, while 10v3 has ~44%. 8v3 is 52%. 5v2 is 43%.
So:
Scaling, best possible EV at each player number:
5p: 3v2 60% town EV.
6p: 4v2 50% town EV.
7p: 5v2 43% town EV.
8p: 6v2 38% town EV.
9p: 6v3 62% town EV.
10p: 7v3 56% town EV.
11p: 8v3 52% town EV.
12p: 9v3 48% town EV.
13p: 10v3 44% town EV.
Adding a scum and changing to Nightstart doesn't change those numbers at all. 7p, 11p and 13p stand out as the most playable numbers. (And of course the relevant Night start shifts.)
Also, I apologize for not catching my number flub before the game, but it makes the almost scumwin WAY more impressive.
Post
Post #170 (isolation #37) » Fri May 17, 2019 11:16 am
Postby Jingle »
In post 168, themilkcartonkid wrote:That game needs a voting stage at the beginning, I think it was a rough start. Maybe vote on the first forked player and go from there
I'd play it with that change, but disagree that it's necessary. Especially in a marathon, getting people flipped fast isn't going to be too hard to deal with.