Team Mafia Results!

[Yosarian2]

I feel as though a lot more people had Hoopla called out as scum in various QTs and in thread, and that her representative (Yos) showed no sign of anyone in their qt suspecting me, but yeah.

(nods) I had no idea you were scum until you suddenly turned and lynched me after calling me town for a while. You played a very solid game. I don't know why the town didn't call you out on it it day 2, or why they let you get away with that ridiculous "Yos was trying to get lynched" defense, but hey.

[SleepyKrew]

I agree that teams should choose which player to put into a game, but not what role/alignment they get. As someone said before, it's all about the environment of the game.

[zoraster]

I wouldn't be opposed to listening to ideas on how I can make it not random yet still assure people can't use the flips of teammates to figure out the alignment of other people.

Technically doesn't clear people's teammates. It's just an assumption that's more often than not going to be true, but that's because any given player is more likely to be town than not. *shrug*

Pardon me?

Getting a scum role PM in one game doesn't make a team less likely to receive one in another game - the probabilities are independent. It's the same thing as assuming that if one player signs up for 4 mini games and gets a scum role PM for one he must be town in the others.

I don't think you read my post very well. I am well aware of independent random assignments. That's why I used the system I did. That said, it does have the risk of having one team with 3 or 4 scum roles, many with none. That's not ideal from a fairness standpoint, so if there were a way to divy it up so that each team got 1 or 2 scum roles, etc. without creating a meta problem, then I'd listen to it. I don't think it exists though.

[SleepyKrew]

Incorporate a no-flip mechanic?

[zoraster]

That does solve the problem (insofar as it really is a "problem"), but it introduces some bigger ones.

[xRECKONERx]

I don't think there's an issue with straight up randomization if you have 15 players, because with the 5-teams setup, there are going to be

at least

2-3 teams that have two scum PMs anyway. I was able to figure out people's alignments pretty early due to the 12 scum PMs -> 13 teams bit. After we figured out Hoopla/mith/Yos2/Ythill had more than 1 scum PM, it became an easy process of clearing teams. At one point we had 8 names on a list and knew that one of them was scum, and after we figured out who that 1 person was, the other 7 were confirmed town due to team makeup.

Adding 3 more scum PMs and 1 more player per team makes that kind of speculation incredibly difficult, and it also decreases the chances of having an 8-way tie due to more variables.

[Katsuki]

I agree that teams should choose which player to put into a game, but not what role/alignment they get. As someone said before, it's all about the environment of the game.

Disagree.

Role and alignment are incredibly important.

[xRECKONERx]

Perhaps the removal of alignment-WIFOM isn't necessary. If that's the case, the process would be something like:

- Team Selection
- Role Distribution
- Game 1 Player Allotment
- Game 2 Player Allotment
- Game 3 Player Allotment
- Games 4 & 5 Player Allotment
- QTs Open
- D1 starts

Also zor: you need to really think of a way to balance out the game lengths.
Psychic Mafia: D1 lasted ~27 hours
Pick Your Power: D1 lasted ~60 hours
White Flag Mafia: D1 lasted ~210 hours
Aggressive Mind Games: D1 lasted ~ 150 hours

There are way too many huge jumps in time. Perhaps you should set standardized deadlines for each game (mean of all four game's D1s is 112 hours; median is 105 hours), or keep 3-week deadlines at the start, but have each game that ends its D1 drop the deadline in the remaining games. For example, after game 1 ends, 72 hours are removed off of the rest of the games... game 2's D1 modscene means the other games lose another 72 hours... something like that, and then just stall night in the other games so that all games start D2 at the same time.

Of course this is probably a bit complex, but just an idea.

[Katsuki]

Giant spread of D1 lengths was more dependent on the playerbases imo.

Length of Psychic Mafia was no surprise considering NINJAS and everything else.
Playerbase of PYP not the type to spend a long time debating as well.
AMG was an example of players that relied more on logic, as seen by the consistantly giant walls.
WF I don't know didn't really look at it. The playerbase seemed like the cautious type though.

[zoraster]

First, all of the games did technically have the same deadline though the bankable deadline led to some differences (not to mention the effect of goofbash)

I have a few ideas on how to try and keep games more in line with each other. The problem is that many of the easier solutions have drawbacks. For example, if I simply said "all day 1s will last 2 weeks exactly" then there's likely to be "dead time" as quick lynching player bases sit around for a long time. That's not ideal.

For what it's worth, I've been playing with some other ideas on the subject such as having a somewhat formula that uses deadline shortening, extending AND lynch requirement shifting based on the average game's day and where an individual game is. For example, if a game 1 was in D3, game 2 and 3 were in D2, and game 4 was in D1, then I might add days to the deadline of Game 1 and increase the number to lynch and do the opposite to Game 4 (decrease deadline, reduce the number needed to lynch).

But that was just something I was considering. I'll return to it more seriously in a few month's time. Part of the deadline system will depend on how the rest of game is designed.

[xRECKONERx]

That formula would work, actually.

Though I don't think you should decrease the # to lynch, since that fucks with balance in a whole new way.

[zoraster]

True, but it's not unavoidable, so long as the formula is simple and easy to understand and people keep an eye on the other games.

[SleepyKrew]

That looks pretty good (don't change the number to lynch, ofc). I see one thing that

might

be a problem, but I think it's just me being unnecessarily critical.

[Hoopla]

I don't think you read my post very well. I am well aware of independent random assignments. That's why I used the system I did. That said, it does have the risk of having one team with 3 or 4 scum roles, many with none. That's not ideal from a fairness standpoint, so if there were a way to divy it up so that each team got 1 or 2 scum roles, etc. without creating a meta problem, then I'd listen to it. I don't think it exists though.

I don't think it's a big deal really. We should be producing games as close to 50/50 as possible, so it should still be four coinflips before skill comes into play regardless of distribution. I actually think it's interesting having some teams with all town to powerhunt together, and other teams with 2 or 3 scum roles to deal with. Diversifies the strategies and way teams interact.

[Mr. Flay]

Wait, why on earth do we (and by we, I mean you) care how dissimilar Day lengths were?

[Hoopla]

Wait, why on earth do we (and by we, I mean you) care how dissimilar Day lengths were?

It's frustrating having one game last less than a week, and one game go for months. That can't really be controlled, though.

[Yosarian2]

I don't think it's a big deal really. We should be producing games as close to 50/50 as possible, so it should still be four coinflips before skill comes into play regardless of distribution.

That's true, but a scum win has a bigger payoff for your team then a town win (since if there is a scum win, only 3 people get the points, while in a town win 9 people get the points). That means that if all games are coinflips, a team with 3 or 4 scum has better odds of winning the whole thing then a team with no scum.

[xRECKONERx]

Wait, why on earth do we (and by we, I mean you) care how dissimilar Day lengths were?

Maybe you should start paying attention more instead of making snarky one-liners.

Day lengths were an issue because one game ended before D1 had even begun in the other games, and since this was TEAM MAFIA and not just a random association of games, other games ending that much earlier than others was frustrating and had a bit of an impact on the game itself.

[Hoopla]

I don't think it's a big deal really. We should be producing games as close to 50/50 as possible, so it should still be four coinflips before skill comes into play regardless of distribution.

That's true, but a scum win has a bigger payoff for your team then a town win (since if there is a scum win, only 3 people get the points, while in a town win 9 people get the points). That means that if all games are coinflips, a team with 3 or 4 scum has better odds of winning the whole thing then a team with no scum.

While a scum win is more valuable than a town win because you're restricting more teams from scoring in that game, a scum loss is equally damaging in the other direction. The payoffs are proportionate - there's just increased importance residing on a scum game. A team with 3 or 4 scum has better odds of losing the whole thing too.

[zoraster]

Still, so long as it's up to judges to pick between 8 different teams, there is at least somewhat of an advantage to getting more scum games than town games. It's just easier to stick out as a good player that way. For one thing, scum have a higher survival rate, even with a vigilante in the mix. I tried to take into consideration that going out early COULD be the best thing for a team (i.e. VT dying N1), but that's hard to control for. For another thing, if I'm evaluating (as I did to a degree) by choosing the best player from each team, just by random chance a scum player has a better shot at being MVP for his side, and thus getting some kudos from the judges.

So even though points wise, you're absolutely right -- getting scum is equal risk and reward -- at least under the way we did it this time, it rewards teams with more scum roles even if the individual games are perfectly balanced.

Let me be clear in case I haven't been recently: random choice is the only way I see to assign roles unless the game was designed wildly differently. That's perfectly fine with me, and it's in keeping with mafiascum as a whole. But it does describe a reason (one of several) that I'm in favor of trying to get away from overall subjective judgments to decide the winner by instituting a point system that takes the end decision away from me.

That's not to say a future point system couldn't involve a measure of subjectivity, but it should be subjectivity that's narrow in focus. For example, +2 points for being the driving force on a successful lynch on a mafia member as town* involves subjectivity, but it's far easier for a judge to determine.

*Please note: this is not an actual suggestion for how points will be structured; it's only an example of the way it could use subjectivity in a narrower way.

[Hoopla]

I think in a game where team cohesiveness was a powerful virtue (just look at SDC's quicktopic), having majority scum roles carries some disadvantages - I think it's a lot harder to contribute to a teammate's game who is scum, than if your team had 3 or 4 town role PM's. Mostly because the skills required in a scum game is not letting people catch onto/suspect you, and a lot of that is based on how you phrase your posts and how you interact with other players. Your team can help in regards to fake scumhunting, but a decent amount of that games chances depends on the player themself. In a town game you're less worried about that, and having good scumhunting analysis in your QT will go a lot further to winning you games than fake-hunting does for a scum game.

It's not a significant disadvantage, because a lot of teams (probably) didn't use the team mechanic effectively/optimally, but in an environment where everyone used it well, it'd help town more than mafia.

[Yosarian2]

While a scum win is more valuable than a town win because you're restricting more teams from scoring in that game, a scum loss is equally damaging in the other direction. The payoffs are proportionate - there's just increased importance residing on a scum game. A team with 3 or 4 scum has better odds of losing the whole thing too.

Well, sure, you've also got a higher chance of getting the lowest score. But assuming that the goal is for your team to win, and not just come out somewhere in the middle, you've got a better shot at actually winning if you have more scum roles.

[Mr. Flay]

While a scum win is more valuable than a town win because you're restricting more teams from scoring in that game, a scum loss is equally damaging in the other direction. The payoffs are proportionate - there's just increased importance residing on a scum game. A team with 3 or 4 scum has better odds of losing the whole thing too.

Well, sure, you've also got a higher chance of getting the lowest score. But assuming that the goal is for your team to win, and not just come out somewhere in the middle, you've got a better shot at actually winning if you have more scum roles.

I agree with Yosarian2... as long as '3rd from last' and 'last' actually have the same 'game effect', your supposed balancing factor is null.

Wait, why on earth do we (and by we, I mean you) care how dissimilar Day lengths were?

Maybe you should start paying attention more instead of making snarky one-liners.

Day lengths were an issue because one game ended before D1 had even begun in the other games, and since this was TEAM MAFIA and not just a random association of games, other games ending that much earlier than others was frustrating and had a bit of an impact on the game itself.

Fine, I'll be more specific. Why the &*%*^$ does it matter if one game ends much earlier than the other? If it's all considered part of the same game-complex, then you're borderline talking about ongoing games to use anything from the 'closed games'. And as already mentioned, nothing said there really "proves" anything about the remaining ongoing game anyway.

[xRECKONERx]

Because it not only messes up the flow of the contest in general, but having an entire game's roles flip at endgame greatly impacts all of the other games still being in D1, due to teammeta and role distribution.

As a hypothetical: let's say three people are revealed as scum in Game 1, and Game 1 ends before Games 2, 3, or 4 can even get out of D1. Team A is scum in Game 1, and Team A is also scum in Games 2 & 3. If any of their team is lynched in Game 2 or 3, it reveals that there are X amount of teams with all town role PMs, where X = 1 + (# of scum roles per team over 1). Like in this iteration, Ythill flipped scum in Psychic Mafia. If Hoopla/mith had flipped scum as well, that would've meant three teams had all-town PMs, which actually allowed me to clear EIGHT people as town just based off of flips/suspicions alone.

Closer days ending and deadlines wouldn't allow this information to be used so openly (or for such a long amount of time).

[iamausername]

wait what

You know Team A has scum in Games 1, 2 and 3. you know Team B has scum in Game 1, you know Team C has scum in Game 1. You know every other team has town in Game 1.

Who are you able to clear with this information and how?