|
On April 08 2013 00:58 Oatsmaster wrote:Show nested quote +On April 08 2013 00:56 AxleGreaser wrote:On April 08 2013 00:33 Oatsmaster wrote:Why is WoS scummier now, rather than yesterday? Something mustve made you vote him right? What is IT? based on previous posting I am hoping he can be in the thread soon. The questions are a ways back in the thread now. I have arranged my sleep to allow me some overlap.I do want to be sure he sees the questions and realises they are current. This post ensures he knows I will be around. This doesnt answer the question. Why did you not vote him before you went off(to sleep I imagine) Why did you vote for him now?
I have arranged my overlap quite bit... I may have nanna naps, but I am around.
|
On April 08 2013 01:20 WaveofShadow wrote: I'll be around now, I was around earlier today but I was out for a couple hours and apparently the 'active' times in this game are quite unlike many of the games I've been in so far. I think it's the ratio of UK/Aussie/Europe/whatever.
Axle I think I remember reading some threat where you're going to vote me if I don't respond to you, I don't give a fuck about that but I'll re-read whatever it was you posted and see if I can actually decipher something you've said for once.
Marv before I left last time I had asked you why exactly you and Ace seemed to think I was lying about my claim and you never answered. I think it was you as well who called my early posting 'derpy' and I'd like to know what about it you find as such.
Oh and one more thing Marv, what are your thoughts on the so-called 'Kenpachi rule?'
You appear to have misremembered who said what.
I asked questions where I said I did.
|
I have seen this post.
On April 08 2013 04:05 Dandel Ion wrote: If no majority is reached by 02:00 GMT (+00:00) (~7 hours from now), we'll end the day in a no-lynch.
I am comfortable with my vote where it is on WoS.
I will be here for bit then nap then come back.
|
On April 08 2013 07:17 marvellosity wrote:Show nested quote +On April 08 2013 07:17 AxleGreaser wrote:I have seen this post. On April 08 2013 04:05 Dandel Ion wrote: If no majority is reached by 02:00 GMT (+00:00) (~7 hours from now), we'll end the day in a no-lynch. I am comfortable with my vote where it is on WoS. I will be here for bit then nap then come back. You'd rather no-lynch than kill Ace? That's what you're saying, right?
No. That is not what I said. That I will be back means I can change my vote. If it is required. I think I can see it may look like I said what that, but to be clear i did not say that.
No Lynch today is very very bad idea.
ACE is TBMK very hard to read. I have not seen stuff that would make me want to lynch him today. I would much prefer to be able to see what he says about things that happen tomorrow.
|
On April 08 2013 07:46 sciberbia wrote: @marv Can you stop arguing with DP and get back to me about Axle. I'm looking at his filter from Noir where he flipped town and he is discussing a bunch of different things with a bunch of different people, and seems to be trying to lead town in the right direction. I see none of that this game. I think he's more likely scum than Ace and it seems there is enough interest in lynching him that we can get a legitimate wagon going.
##Vote AxleGreaser
Palmar, can you stop ignoring everything I say to you and tell me why you think AxleGreaser is town?
The number of people I discuss things with would depend on the number of questions I had about what people had done. The tendency not to interfere when another person is applying pressure also means, if someone else in the thread is asking the right questions i can listen to there answers rather than just chime in with me too questions.
How deeply id you think about why the difference if there is one might exist? Did you ask if there could be a towny reason for the difference? Did you consider the options above?
|
|
+ Show Spoiler +On April 11 2013 05:25 XXX wrote: I don't really know what to make of it: - Oats questioned a lot of other people than WoS at that time. - I was not around until last moments. I was suspicious of WoS earlier though. - prpl voted for WoS cos busy and marv/Palmar voting for him, later said he only then understood Tunkeg's case fully. - Axle/Tunkeg were MIA. AKA Axle asleep in a different TZ. On April 11 2013 08:23 XXX wrote: Because i was at LAN-party with shitty internet and about 24h/weekend drunk like a fish. AKA XXX was MIA?
What I find funniest(funny peculiar) about these games is how the same or similar observations get re-interpreted entirely depending on the current posters current beliefs. Although the example is of rayn, I ripped the name out as its only an example. While I know I was scum, and I did get caught, some part of the 'scumminess' of my play... wasn't.
Oh and ditto on the thanks to the hosts. LOl still coming. (edit: got distracted by shiny things.. .the time has past.)
|
On April 11 2013 14:56 Oatsmaster wrote: Also when Axle is playing, I ROLL SCUM. EVERY FUCKING TIME. Thanks Axle. <3
Anytime, no charge.
|
On April 11 2013 21:08 marvellosity wrote: Mafia were never going to win this game, there was just too large a difference in abilities/thread impact.
This might sound like I'm bashing on the skill of the mafia but it's not how it's meant to come across.
None of the people on the mafia-team have a history of large thread impact in a game and that's just a fact.
This game was always going to be a town win.
If you're happy for a game to be one-sided for the sake of some silly RNG concept, then fine, but you'll get games where mafia basically can't win like this one.
Choices (that exist on a continuous spectrum) 1) "fair" "just" games: in which if your team has the best chance of winning if it plays the best (maybe because it has an unbalanced share of the (historically)better players who by definition (historically)usually play 'better' more often.)
2) Challenging games where, you if play better/worse than your usual average that makes your team have a better/worse than average chance of winning.
Both have value, it is a choice of what you want out of the game. Basically its scratch VS handicap racing or interesting racing. Usually, in other competitive settings, with widely spread fields unless its handicapped people get bored, and thus arse around, or simply find something else not boring.
For me as it stands, I kind of don't need 1), I already independently evaluate my play and game and effort. BTW: So far I am at played 4 won zero... go team me. My personal estimate of how I have done is based more on the effort I put in. If I tried hard and failed anyway then that merely means learning to play was harder than I first thought. While play to win is the rule, it is not why I played any game I was in. So yeah this game was hard and uphill, but it gave us three the space to be great, if we could be. Turned out we were not. (Wasn't really a big surprise to me, although just how and why it went wrong was.) Next time I am in a scum game, and its that up hill my goal will to not get lynched D1.
One other problem with 1) is that it may promote this kind of response. When scum team thinks it has a below average set of players... why bother. Just wait until next time when you get a good team. This would be bad place for the game to wind up.
A problem with 2) is, once some part of the scum team flips... the other part gets tied into a narrower pool. If you are the scum in that pool, then it seems really unfair, remembering that being in an all 'not good'/'bad'/ scum team from one also feels unfair when its you.
Question: (aka I have NFI never having done it.) Can the role distributions make up for it? In semi open or closed setups, there are I expect a range of what is considered near enough to be balanced possibilities, but some are on the town/scum favored end of the spectrum. (I think not all 2of4 setups are equal, ie they are orderable: from most to least town favored.)
By not rolling both the players and the setup independently, you could include some counter balance by biasing the probabilities of each setup. This would still mean once mass claim occurs you can maybe guess probably whether the scum team has more or less stronger than average players. (even then town may not know how loaded or bare the scum powers cupboard is.)
Another option, is to do true RNG, but who says the PDF has to be flat.... There exist variety of algorithms that can for instance by assigning players to teams one at time, make adjustments such that once one team starts becoming strong or weak, the odds on which teams gets the strong players after that is no longer quite even... By pushing the RNG back that far, analysis of who is scum and who is town cant know for sure there (Must/Must not) be another Vet on the team.
|
On April 11 2013 22:10 wherebugsgo wrote:Show nested quote +On April 11 2013 21:49 marvellosity wrote:On April 11 2013 21:36 wherebugsgo wrote:On April 11 2013 21:33 GMarshal wrote:On April 11 2013 21:28 marvellosity wrote:On April 11 2013 21:26 wherebugsgo wrote: You say it's not the player's fault for not having an ability when they are "known" not to have it.
What the fuck? How can you know a person's ability in this game? There are no reliable indicators of skill level in mafia when they change even between games.
Not even winrates are that useful because the sample sizes are too small, and the games are team-based, not individual based.
You can't tell who's a good player and who's not so good? Seriously? Are you unable to assess players' strengths depending on the numerous games you've seen from them? Are you that unaware? When I first started playing my first two scum games I lurked, I lurked like there was no tomorrow, easiest scum read ever. Cue game #3 as mafia, where I got elected pardoner and generally misled town until drama exploded. You can tell who's good and who's bad, but you can't predict breakout performances, and discriminating against players because they "aren't good" based off a sample size of 3-5 games seems like an awful idea. Plus, reading "bad" players is part of playing mafia. this times 1 million. What happens when two new players come along and you balance their team of 3 players by putting a vet like Ace on that team to balance it, but those two new players turn out to be really good at scum? What happens when you have a game with 6 vets, but only one of them is actually good at scum? Do you condemn that guy to playing on scum? What happens when every host thinks the same way you do? That guy will end up playing nothing but scum. You're taking a way simplistic view when it should be obvious that it's not one I advocate. Can you give me a single example of the Ace + 2 newbs playing amazing scumgames happening? I bet you can't. There are monumentally few players who take to mafia like a fish to water. Since I've been playing, I can think of me, Acro, probably DP too - and that's about it. Tunkeg has never played mafia before and Axle has played mafia once. The funny thing about this game is that Oats *did* play much better than he played as mafia before, so he got caught properly day 3 instead of day 1. Teams shouldn't be handpicked with massive scrutiny and eye to detail, but if you put 3 mafia players with 3 scumgames between them and no previously indicated ability as mafia together, then you get a game like this. That's just how it is. my first scum game. gg bitch e: joking aside, there are lots of problems with hand balance. GM brought up good points and you aren't doing anything to address any of them. Your assertions are baseless until you can come up with a good argument outside of "these guys aren't good at mafia therefore they had no chance to win". (hint: it's not a good argument) There have been plenty of games with average scum teams like this one that turned into mafia wins. If you can predict the outcome of the game reliably with nothing more than the names of the players on each side, then cool, but I don't think anyone on this forum can do that. Most important is the fact that people don't improve unless they change their attitudes about balance. I think it's fine to call out questionable role balance because there have been some setups that are just bad, but blaming a loss on player distribution is basically like saying you gave up when you saw your team. It's both insulting to your fellow teammates and to the community at large.
> If you can predict the outcome of the game reliably with nothing more than the names of the players on each side, > then cool, but I don't think anyone on this forum can do that. While no one can do that very accurately
With what degree of accuracy can it be done? At some degree of accuracy better than randomly guessing which team will win, it is likely to be possible.
I don't have any problem with the game I just played... but i am pretty sure if we could somehow replay it 100 times there is almost zero chance scum would have won half of them. That such a team could be the scum is I believe quite wise, does it have to be exactly equally as likely as some other more likely to win combinations?
To be clear I think I had some real benefits, by playing in scum team with a less than average chance of winning, bu also one where we could not just sit back (sheep) and let the scum expert make the calls. While there were real benefits, is it wise that it happen as often as it would by chance. One good question is WHY did Oats play better? There is an entire spectrum between, every scum team gets this fraction of the vets, and true Flat RNG
I think I have said all I usefully can as, (IMO), I don't comparatively know jack about mafia, never hosted a mafia game (have hosted other social interaction games, where feeling like fate wasn't determined pregame, was highly correlated with player enjoyment, note that cuts both ways, why work hard to be a good player if the host is still going to handicap you into losing 50%, ...etc etc ) and I couldn't balance a setup with out rather lot of research (and maybe not at all).
However, FYI: I have 'some' back ground in things such as 'better than randomly guessing', often more formally described as ROC, (Receiver_operating_characteristic), if after discussing it someone wants technical help... I have skills in translating intent into algorithm.
|
On April 11 2013 23:16 GMarshal wrote: I think you guys arguing for balanced teams underestimate two things 1.) How volatile games of mafia are, take nomination where mafia had the game in the bag and managed to inexplicably blow it. Look at how quickly people begin doubting "confirmed" townies, etc. Look at how marv played in personality (sorry marv). having a vet on a team is no guarantee of anything.
2.) Breakout games, I haven't seen many of late, but there are occasions where a previously unpromising player absolutely rocks a game, usually this happens when they are forced to step up, e.g. town lacks focus and direction (The absurd medic/leader in one of the new games, got nominated for best newbie for that play. DrH in his first game where he became a focal town leader. Myself in the pardoner game. Jackal in XXXIII.), or they are placed in a team with no leadership and are forced to step up.
I personally think its totally impossible to predict accurately who will win from a playerlist. This game could have easily turned into the town infighting and lynching itself had day 1 gone differently, or had one member of the scumteam done an absurdly ballsy play, etc.
Anyway, blaming the "balance" of rnged rolls seems like a waste of time, its better to focus on lessons learned so that next time that a "weak" mafia team rolls out they know what to do. So, what could the mafia team have done better to establish thread presence? How could they have avoided the disastrous day 1 hammer (if at all)? What was the proper form of damage control post fact?
To be crystal clear. I am not arguing for "balanced teams"
I am saying there is something between pure flat RNG, and handcrafted guessable teams.
The thing i am saying will not be reasonably gameable by real players.
> I personally think its totally impossible to predict accurately who will win from a playerlist. This game could have easily > turned into the town infighting and lynching itself had day 1 gone differently, or had one member of the scumteam done > an absurdly ballsy play, etc. You are perfectly correct, it is "impossible to predict accurately who will win from a playerlist."
I am not sure that it being 'impossible to be accurate', means it is anything like impossible to be able to guess considerably better than average(randomly guessing) , or to be able to choose which combination of teams ought be less likely to be generated as they are more likely to lead to a ROFL stomp. The algorithm to then roll such teams is no longer, simple enough to be easily done with a dice pen and paper, its also not prohibitively difficult.
> This game could have easily turned into... I suspect that while it could have easily done this, the odds of such things happening was not near 50% > 2.) Breakout games, I haven't seen many of late, but there are occasions So i guess break out things don't happen all that easily... Do note however, break out things may start happening never if hosts start balancing games such that scum teams never think OMG were fucked... Lets try something extraordinary. Lets try to be great !!
So yeah i am both with you and against you, scum team like the one I was just in need to be possible, but does it have to be exactly as likely as some of the other possible teams?.
Syllogism said: I'm personally not going to join a game if I know for a fact the host balances teams. I don't like it when people work out who is and is not scum due to nothing that player did at all. (eg Host balance and other flips.) That doesn't feel like you beat the scum player at all. WORSE. Outcomes will tend to be scum stomps or town stomps because if any scum players flip, now not only the inter-player interactions but the game balance inter actions solve it.
However, if the host only makes it some unknown amount more or less likely that a team would have all strong or weak players, and the hosts actual estimate of how strong all the players are is unknown and some of that imbalance may be addressed by role distribution in scum or town, and.... Then using host balance speculation as part of any lynch would become much more dubious.
Fundamentally how much do I think hosts should tweak teams and power roles to get balance.
1) The goal is not that everyone has 50% chance of winning. better players ought win more often.
2) If the players have gamed he host and are successfully guessing teams based on balance... Time to change the algorithm.
|
On April 12 2013 00:04 wherebugsgo wrote:Show nested quote +On April 11 2013 23:53 AxleGreaser wrote: 2) If the players have gamed he host and are successfully guessing teams based on balance... Time to change the algorithm.
so what do you propose this game would have been a fair scumteam?
That would be the rub, while I know something about the algorithms to do such things and how to implement and create them, I think I lack the expertise to do one critical step (accurately).
So I will walk through version alpha zero, of an algorithm. If it does not in fact work I will just keep making up crap until I find one that does.
Also note I substituted aqua for me as its less wifomy to do it for others.
Step 1. Split players into two skill groups (doesnt matter where)
Ace Palmar marvellosity iamperfection WaveofShadow prplhz DarthPunk Oatsmaster Tunkeg Aquanim raynpelikoneet sciberbia
step 2 Repeat step one until you look at every group and say meh all the same AFAIK.
Group 3 Ace Palmar marvellosity Group 2 prplhz iamperfection DarthPunk sciberbia raynpelikoneet
Group 1 WaveofShadow Oatsmaster Tunkeg Aquanim
Pls note If I put you in the wrong bucket... booo hoo... It really doesn't change much. If I did put you in the wrong bucket all it means is I am bad at mafia. Also I have some order in my head for those buckets and some doubt about where the best divisions are.
At this point players within each group is not ordered. The categories are based on average of town and scum play according to my best guess. If you can split it into more or less groups, then fine, it wont matter. Minorish tweaks of what comes next will work with that too. So far I am probably not too far off what other people think on the groupings and even if I was it wont matter much. Indeed me injecting my personal perspective if i am the host makes it that much harder to guess.
here is where expertise I don't have, but the people who host games have better than I do.
Does the team prplhz, iamperfection, DarthPunk got an even money chance of winning? How about Ace, Iamperfection, WaveOfShadow/Aquanim If you suddenly find yourself saying Ace, Iamperfection, Oats is lot better then perhaps your original groupings are wrong.
Anyway if the green two are balanced, then i suspect you will find the following will workish. Give each player their group number. Adding up the players group number in the scum team measures its strength.
Then one trivial algorithm would be to first generate the scum teams exactly randomly. Score the team (add up its group numbers.) The two green ones add up to 6. If its is 6 keep it. If it is 5 or 7 keep it 83% of the time ( roll a D6, reject team on 6 ) If it is 4 or 8 keep it 66% of the time ( roll a D6, reject team on 5,6) If it is 3 or 9 keep it 50% of the time ( roll a D6, reject team on 4,5,6) If it is <=2 or >=10 get your cohost to do it or wait until you sober up.
if you rejected the scum team go back and roll another random one.
When you look at that algorithm and groupings and Go yeah Thats Ok... BUT you really must do that before you roll, you also ought check the boundary conditions thus.
Just how BAD is Scumteam(Ace Palmar marvellosity) or (Oatsmaster Tunkeg Aquanim) is 50% less likely than random enough... or would just how epic it might be compensate?
Note what you also might do is based on, your scum team total rating change the setup by?
Edit: Note this post will probably get edited for clarity. and becuase its alpha.... Caveats: if the algorithm gives you the wrong answer dont be afraid to just mess with it. Group 3 might be given the score 3.5 or 2.5. Hell I can imagine giving player -1, if they were positively likely to get flipped after having first fingered all their scum buddies. Host ranking of players etal would all of course be shredded. If I was host Id ask the cohosts to rank them as well and mix their ranking result with mine even if I thought they were wrong.
Caveats: in practice I would try to get more than 2 example teams that looked 50-50 and check they or most of them scored 6.
Caveat: The above algorithm, as it stands, is probably a bad idea, as it changes the probability that anyone player rolls scum. By re-rolling games that are group3 and group 1 heavy, the odds of group 3 and 1 rolling scum in any team go down. While that probably is an issue the size of the difference is minimised because, there naturally is not a lot of combinations that are group 3 or 1 heavy. While that is still some issue, it would i believe be repairable by fiddling at the margins with some scum teams frequency a little bit. For instance reducing the frequency of teams 222 would be able to fix most of the problem, spreading that out into the 221 and 223 (decreased) (311, 331 increased) ... would compensate for this and still leave a bias against the unbalanced skill teams being rolled. While I will probably work that out, as its interesting, posting the results in an intelligible form, does not seem worth it.
|
On April 12 2013 01:02 raynpelikoneet wrote:Show nested quote +On April 12 2013 00:56 DarthPunk wrote:On April 12 2013 00:52 raynpelikoneet wrote: DP have you ever been lynched in a game? Yeah. Twice. The first time was my first newbie. The second time it was marvs fault but Bugs yelled at him a lot and pm'd me apologising for letting me get lynched. So that made me feel better. Yeah then your record is very impressive. What i mean is even if you got a record of 100-0 it doesn't really mean anything if you are lynched in 80% of the games. And then you are clearly doing something wrong. After all this is a game where not only what you say matters, but who are the people analyzing what you are saying and do they believe you. If nobody believes you it doesn't matter how right you are and you are doing something wrong. :D
A question: If there was player with a 100-0 win rate as town and scum. Got lynched 100 times D1.... But somehow had the uncanny ability to get the scum to out themselves by mislynching him, and town to fall in WIFOM hole when they lynched him D1 as scum.
While probably not plausible, just as the 100-0 isn't. Wouldn't that be spectacular play?
|
On April 12 2013 01:37 Oatsmaster wrote:Show nested quote +On April 12 2013 01:36 marvellosity wrote:On April 12 2013 01:29 wherebugsgo wrote: I disagree still.
There is nothing that indicates a scumteam of sciberbia/DP/whoever would necessarily have been better than what we had.
You know why? Cause people like Axle, who play very differently dependent on alignment, are unlikely to get lynched as town.
It's all just conjecture, saying that you would have been able to balance the game better on that basis without considering those 3 vets. The argument is based on seeing the results, but there is no empirical evidence that it actually holds in practice (also there's no evidence that the reason the scumteam lost was because of player imbalance either!)
And to top it all off there's no measure of skill that says the scumteam you proposed is better than the one that was in this game, or that the change would have kept town level the same.
E: this was @ DP and whomever else was talking about the teams I'm kinda tempted, in a completely arbitrary manner, to do some kind of ELO rating for players with both a town and a mafia ELO, with results in games averaged across the team's ELO. Just for curiosity of how it looks. On a speculative note, you could say that you'd expect games with a higher proportion of players with large disparity between town and mafia ELOs to have more positive town results, regardless of 'objective' rating. For example a game of 12 iamperfections would more often result in a town win than a game of 12 wbgs. 12 Iamps. Hilarious
I thought it was very bright Idea.
|
On April 12 2013 01:58 syllogism wrote: While you are doing arbitrary math, can you factor in the advantage town gets from knowing that teams aren't RNGed
If you can tell how the players can guess, how the host groups and scores the players. How the players, know what scope the host had to rebalance unbalanced teams by tweaks to the setup/roles which can vary from game to game. How the players deal with the fact that whenever they do that and get it right regularly, then the host does it less, until they give up.
I would suggest hosts ought not say whats the maximum probability to reject a team. I think you would find if you actually examined every possible 5,6,7 scoring team in the algorithm I suggested, that fairly often even with two flipped players, there is still very little actual information about who the third would be.
I am not sure, I could also try and factor in the advantage town gets when some scum players give up when they see their team mates. Well not so much give up but don't push as much other real life commitments out of the way to play as they would have if they had seriously viable team. Knowing that whenever you roll one of those hard games.. you were quite unlucky but that this is your chance.. to either shine if your the noob, become star by dragging a bunch of misfits over the line if you are the better, but not yet stellar player.
The math BTW was not arbitrary...
Also please note, I think i may have at this time a _personal_ preference for playing in RNG games. I will have to wait until I get ROFL stomped a statistically interesting number of times to be sure. Also at the moment I learn so much each time I play, winning/losing is meh. I do know however from having hosted RL games that were social games, (not mafia, but still social), what it took to keep my players happy. Reducing ROFL stomp frequency smells like it. I also hope that if I become better that I also wont get any satisfaction from beating an easy scum team.
(edit: not mafia == not any epic guessing game / social == played by people who largely use the game rules as an excuse, which game not specified as that would be misleading as most people dont play that game that way. )
|
On April 12 2013 02:00 WaveofShadow wrote:Show nested quote +On April 12 2013 01:55 marvellosity wrote: Just as an arbitrary example for this game, if we take 1500 ELO to be 'starting' ELO with 2500 being totally exceptional and 500 being very poor, you might see something like the following (apologise in advance for over/under estimating players, more for shits and giggles)
1. iamperfection; town ELO 2000 mafia ELO 1000 2. WaveofShadow; town ELO 1500 mafia ELO 1500 3. Ace; town ELO 2100 mafia ELO 2500 4. prplhz; town ELO 1600 mafia ELO 1200 5. DarthPunk; town ELO 2000 mafia ELO 2000 6. Oatsmaster; town ELO 1400 mafia ELO 1300 7. Tunkeg; town ELO 1500 mafia ELO 1500 8. Palmar; town ELO 2400 mafia ELO 2000 9. marvellosity; town ELO 2100 mafia ELO 2400 10. Axlegreaser; town ELO 1300 mafia ELO 1300 11. raynpelinkeet; town ELO 1600 mafia ELO 1600 12. sciberbia; town ELO 2100 mafia ELO 1700
Average strength of mafia team: 4100/3 = 1366.6 Average strength of town team: 17400/9 = 1933.3
Best theoretical mafia team: 6900/3 = 2300 Worst theoretical town team: 15000/9 = 1666.6
Average town ELO: 21600/12 = 1800 Average mafia ELO: 20000/12 = 1666.6 I'm not sure what the point of doing something like this is with no real mathematical way of estimating skill.
Its a schtick for me to play with.... it will take little while though.. its 3am. Gdnight.
|
On April 12 2013 09:46 Dandel Ion wrote: What the actual fuck are you guys doing to my perfectly fine game thread?
Apparently playing a different game.
|
|
|
|