Team Mafia 2015 Winners: Young and Beautiful

[Zachrulez]

Are the captains in?

[Zachrulez]

3. During the side selection phase, each team may place the up to 5 remaning tokens on the games they are in, selecting which side they'd prefer. Each team is automatically entered into each side once, and each token will enter them in for the side of preference one more time.

Took two things from this. Both sides automatically get one town and one scum draw, and the tokens influence the likelihood of the other two draws?

[Zachrulez]

Yeah, that's actually sort of a temporary placeholder for the actual mathematical way I'll do it (since actually having equal numbers of tokens on each side doesn't work out), but the essence is that you get to influence which side your individual players get but don't get to decide it, and it's connected to both stages.

Previous versions had a ton of outright guessing based on team compositions since you got all the role pms at once and decided who was taking what role (so if you're in a game with Mina who thinks she can't play scum, you can be reasonably confident that she would have made her teammates give her town). This way there's a little of that, but really no sure things. That's particularly true if everyone tries to get town in a certain game, for example.

You mentioned in that each side is automatically chosen once. Does that guarantee a town and a scum draw then or am I just not reading that right?

[Zachrulez]

no, absolutely not.

This is the updated token side selection:

Side Selection

If unaltered by tokens, each player in the game has a random number generated between 500 and 600. The lowest X numbers are Mafia (where X is the number of Mafia in the game).
Each token put on Mafia on the game lowers the lower bound by 100. Each token put on Town on the game raises the upper bound by 80.

Numbers are definitely subject to change.

Sounds interesting. I'm still confused as to how the numbers will work, but I imagine that will get more clear as we get closer to the games and if not we can just ask outright about it in out PTs.

I will say it's definitely an improvement that alignment selection is happening after the game choices.

[Zachrulez]

So ummm, I tried to bookmark this topic and it's not showing up in my bookmarks. The option to remove it is below as it should be. Weird.

[Zachrulez]

Apparently you can, they just don't show up in your bookmarks.

[Zachrulez]

Not really thrilled with Matrix for 9p, also that means the game would be semi-open rather than open.

[Zachrulez]

Well if I'm stuck with an open/semi open, I would prefer it to be simple. Matrix seems complex, swingy to the roles and untested.

I ran a 9p normal that had a 1 shot cop and a doctor with just 2 goons as the scum. How far away is something that simple from balance?

[Zachrulez]

I mean our newbies do matrix6. It's really not that complex, it's basically "here are 6 setups it could be."

I think the difficulty with the above is making it an open, and the swing factor (1-shot cop who gets a guilty swings the game toward town significantly, 1-shot cop who gets an innocent/dead guy or is killed before revealing result swings the game pretty heavily toward scum).

1-shot roleblocker against 1 scum can be an even more devastating swing. Scum generally can't even defend that by making the correct kill.

[Zachrulez]

CES's refined version proposes one in 2C.

[Zachrulez]

Twin Trap looks okay.

I don't mind the setup having some layers to it, but I would prefer it's easy to grasp and not something that leads to setup speculation, and I think something like matrix might lead to a game of guess the setup.

[Zachrulez]

Could possibly list all the potential setups and then maybe do it by player vote?

Maybe even do that as a game mechanic where the town votes the setup they want? (Similar to pick your poison/power, ect, whatever.)

[Zachrulez]

i've considered the vote, but don't think having the town vote would be a wise idea.

Because they'd pick the most beneficial setup to them, or that players may go after others for their setup choice in their scumhunting?

[Zachrulez]

Anything that isn't matrix I'm probably okay with.

[Zachrulez]

What's wrong with a simple open setup like JK9 or just JK+6 townies vs 2 goons?
Maybe a sole cop or a JK plus doc? (Field day, isn't it?)

Matrix6 is nice and all, but it's not a simple open setup, something we are really looking for here.

Even doc+cop vs RB seems good

.

If the RB is lynched on day 1 then follow-the-cop would ensue.

That's actually why I went with a 1 shot cop/doc combo vs 2 goons in a 9p normal a while back. Follow the cop is impossible because they can only get one guilty. The only real problem with that is balance, but it seems like it would be close enough to do the job, and if it doesn't a weak town power role could probably bridge the gap.

Chosen is also an interesting setup I've never played in, but I don't think we're going to want both that and white flag in play.

Matrix has a lot of variables and I don't think simply mathematically balancing it if they even are is going to account for the actual win rates off play. I mean I actually prefer 'boring and simple' because it allows me to try to bust the game via day play rather than having a deep setup to think about and speculate on.

Also I'd say that 9p setups in their nature are best off being kept simple. You actually do need around 12-13p to actually make a setup interesting role wise if that's the kind of thing you're looking for. Semi-open setups with many possible power roles might make things a little bit more interesting for those who love the setup having variables, but I don't really think it's a good thing for a smaller player base like 9p.

[Zachrulez]

Of the current suggestions, I'd go for old plain newbie matrix6. Not the revised version, I don't really get why we are going to test new setups here. Old matrix6 has been proven to be balanced enough.

The overall stats might be balanced, but some of the draws aren't necessarily.

[Zachrulez]

Dat 0-4 record.

[Zachrulez]

I hope we have all learned things about busses that apparently don't make sense.

[Zachrulez]

Hey Reckoner: don't worry, I won't ruin the next Team Mafia. Two is enough for me.

You didn't ruin it. Us derping on the night actions contributed greatly to the scum win.

[Zachrulez]

5. Young and Beautiful were the only team that won all four games. I know it was suggested in the dead thread that something other than wins should determine things, but ultimately any team that is perfect in their games should be eligible for a victory.

As someone who didn't participate (didn't know how cancer treatment would affect my time on site) but followed along, I feel fairly confidently neutral in saying this: This is pretty terrible.

It's been shown long ago that win/loss records don't really correlate in any way to talent/ability or contribution. This is especially true when game imbalance and uneven distribution of factions plays a factor. I never won a paragon based on win/loss record, and though I haven't tallied my raw statistics in years, I know that at least as of like 2011 I had more town losses than wins. There's something fundamentally wrong with the scoring when a team that did relatively little in their victories happened to be the only team that went 4-0, because their non-teammates carried them to victory four times. The idea that overall win/loss record should be a major determining factor in Team Mafia is about as backwards as can be.

If there is something "fundamentally wrong" with it, then there is something fundamentally wrong with Team Mafia as a whole. Either we operate under the (perhaps fictional) idea that skill influences wins or we don't. But I'm not having a winner of Team Mafia as a whole be a team that won fewer games than another team. It would be unfair, it would defeat the purpose of Team Mafia, etc. It's one thing to use subjective measures as a tie-breaker. It's another to act as if wins and losses don't matter -- they do.

Ultimately I don't know how much simpler or fairer it can be than this: If you win more mafia games than other teams, you win Team Mafia. Any other standard is unacceptable.

Trying to make the victor of team mafia be determined by something else more primarily than anything other than wins is putting too much value in wins anyway.

People obsess a lot over the outcome of a game, but sometimes things just happen that are out of your control. If those things cause you to lose a mafia game, why should the standard change for team mafia to determine a winner through a subjective evaluation of skill?

[Zachrulez]

Hopefully if there's another one of these and I play again, my team will fare better than 0-4.

[Zachrulez]

Mafia is like that sometimes.

You don't always feel like the people who won are the people who should.

Doesn't mean we should change the fact that win loss record should determine the victor.

[Zachrulez]

That seems like dangerous territory to go in because I already think it's kind iffy to say that teams that won fewer games 'outperformed.'

[Zachrulez]

I mean it's like saying in a hypothetical football game because you performed better than the other team but lost because of some lucky bounces they got that you deserve to win the game instead. The score is what matters in the end.

[Zachrulez]

Frankly I would be pissed if my team swept and a 3-1 team won over mine because they 'outperformed.'

Why even play to win if sweeping might not even matter?

[Zachrulez]

Why even play to win if sweeping might not even matter?

are you kidding

I think it's a fair question, particularly for people who feel like their play style isn't particularly valued by others.

I mean losing to another team that swept due to the scoring mechanics or subjective measures. I can complain but generally accept the result. But 3-1 winning over a sweep?...

[Zachrulez]

I've considered giving penalties to teams who are lynched Day 1, but rejected it as it incentivizes power claims and gives advantages to those who are power roles.

You could have larger teams next time, or have the teams be the same size but each player plays two games or something. It's probably much less likely for things to end in a tie with more games.

Also gives you a chance to come back with a much stronger performance after losing if you have a 2nd game to play afterwards.

Just floating ideas...

[Zachrulez]

It's a fair question when you argue for a system where fewer wins allows a team to beat one that wins more games.

[Zachrulez]

I certainly hope nightless gets nixed if/when the next team mafia is held.

Really didn't even want to draw that game to begin with.

[Zachrulez]

I certainly hope nightless gets nixed if/when the next team mafia is held.

Really didn't even want to draw that game to begin with.

Nightless games are a thing, though, and they're kind of classic in a way. I just don't particularly like them but I guess they have their public, so I wouldn't be opposed to a variant of them in a future Team Mafia edition? In the optimal scenario, I probably wouldn't have had to be involved with it other than give Empire reads (similarly to Regfan on the White Flag).

I was directly involved in a rebels in the palace game very early on in my time on the site and bussing marred that one and allowed scum to repeatedly snatch defeat from the jaws of victory multiple times. (lylo)

It's not just a theory disagreement when I fault Metal for having that mentality, it's that I've seen bussing ruin scum in nightless before. It's exponentially harder to get away with it when you don't have a nightkill to eliminate the people that see through it.

[Zachrulez]

I've considered giving penalties to teams who are lynched Day 1, but rejected it as it incentivizes power claims and gives advantages to those who are power roles.

You could have larger teams next time, or have the teams be the same size but each player plays two games or something. It's probably much less likely for things to end in a tie with more games.

Also gives you a chance to come back with a much stronger performance after losing if you have a 2nd game to play afterwards.

Just floating ideas...

more games is good, but I thnk it's important for all teams to be in all games

It's a balancing act. I'd love to do six man teams in six games, but I worry about making people follow six games. Four seems like a lot as it is. But just Four games was too few. The first team mafia ended with an eight way tie for first with 2 wins a piece. That felt unsatisying.

You could do it so they aren't all running at once.