M 1029: The webMafia Olympics game 1 SPY VS SPY

[DiplomacyandWarfare]

No kill folks are forgetting the mafia are two different teams whose main win condition is eliminating each other. People are counting them as the same team in their logic, makes no sense. They want to kill each other way more than town because if they do, they get an optional nk and cannot be nked themselves. That means surviving town will likely be able to coast for some time while they duke it out making the standard town vs. mafia counting irrelevant until the other team is on the back foot. Plus HB's logic of saying things like well in a couple nights we could have majority mafia and it'll be game over is kind of like... well yeah but also most of us won't be town anymore... idk, you just gotta play your best from the position you're at currently, and to me that means we kill. 1/4 chance of chopping off the head of one mafia team from the get go is a bit too good to pass up to me.

I buy Dip's thinking on this as kind of genuine, HB's a bit too but less so, but I don't really buy Ghug's take on this at all. If it was baiting people to go along with a bad no kill to see who he could catch, I kind of get it, but he definitely wasn't being real there. Everyone being invested in killing day 1 is kind of a main stay of the genre, if that isn't the case the set up is pretty messy, since we just end up sitting around with thumbs in our bums for a while. Game set up makes sense to kill today for town, maybe slightly less so for scum because while they are likely to live, they have a 1/8 chance of losing immediately, I could see safe boy scum wanting a no kill more than a town would. My feeling is ##vote ghug, since he is the only one I feel is being disingenuous so far.

Everyone being invested in killing D1 is a mainstay of mafia. This isn't mafia. There's no connection information to be gleaned today at all.

We have. 1/5 chance of hitting a spy, without the usual mechanic of mafia having a large sway of who flips. We don’t get that luxury later on

why does everyone care so much about voting info?
KILLING TODAY IS NOT FOR GETTING INFO
KILLING TODAY HAS THE GOAL OF FINDING AND ELIMINATING A MAF LEADER, THEREBY ELIMINATING AN ENTIRE MAF TEAM

[bozotheclown]

Normal 7-2 is loss at the third miskill. This 7-2 we lose on the second miskill unless we specifically kill the leader D2. Two kills of non-leader mafia and one miskill would also be a loss. It's just not a viable path.

The two ways town can win that I can see are to DK the 2 currently mafia D1 and D2, or to get lucky with the mafia randomly trying to recruit each other and then trying to eliminate each other.

Yes. I'm arguing that the former is unlikely and we should thus be putting our eggs in the latter basket.

It may be very unlikely, but we could still take a shot D1. If we are lucky our chances improve D2, and if not, we would still have the opportunity to win by the mafia recruiting each other with one less town.

[DiplomacyandWarfare]

here I was killing day 1 my entire life because you have a chance of killing scum and winning the game. As opposed to no chance at all of getting scum and usually 100% chance of one town dying (or being recruited) each day you do nothing. Info and catching scum by hunting is nice and all, and we can get that by pursuing a kill anyway, but also the win condition isn't pitting mafia against each other, the win condition is killing them. Always has been. Still is in this game. Why isn't that in your list?

It's an implied facet of point two. The only way town wins a normal game is by directing kills. No killing means fewer kills. Usually the debate is "why should we kill if our odds are low?", so that's how it's framed.

I explained in my post that I don't think killing scum is even actually to our benefit, and the other benefits to risking killing a townie is not there.

That is definitely innacurate. Killing a scum today eliminates a team, which removes annoying situations like:
Let's say we nokill, and both maf recruit. That leaves us with 5 2 2. However, at most one maf team can win. That means that at least one of the players who was just recruited from town is going to lose. Both maf teams will further recruit and in the end at most 2-3 players who are currently town end up winning.
On the flip side, eliminating a maf today means that we have both chances to win as town (if we kill the maf leader of the other team tomorrow, when associate reads will work), and to win as maf (which are boosted because there's no second maf team to oppose us)
to make this simpler, in a 3-team setup about a third are likely to win, while in a 2-team setup about a half are likely to win.
Which I am probably explaining in a terrible manner but its kind of self evident that we want to eliminate a maf team today.

[DiplomacyandWarfare]

Reads:
Brain seems sus
Bona's stupidity seems more likely to come from town
I don't think scum Brian pushes a no kill
Good vibes from Haze
That is all

Awkwardly, I apparently share reads with the person I am voting for.

[bozotheclown]

##VOTE brainbomb

Why did your spreadsheet not agree with what you posted before the game started?

[Col7by]

Ok I was hoping y'all would just get this but we're not playing mafia.

Several people have already noted that town is doomed. That's true if we play it like mafia, and it's kind of a feature of the setup. The way town wins this game is by pitting the mafia against each other. Townies can also win this game by being recruited, but my reading of the rules has always been that playing to this outcome runs contrary to playing to one's own win condition (the spectrum of positions on this point kind of undermines the game's whole conceit, which is one of the reasons I'm personally not a fan of recruitment games).

If we kill a townie today, we have less ability to effect our will as the game progresses further. If we kill scum today, there's no counterweight to the other scum team. We go into tomorrow in what's basically a brand new 7v2 mafia game in which we have no PRs, scum has twice as much power as a normal mafia team to reduce the number disparity as long as their leader is alive, and every new recruit is going to lack connections to their teammates. Town is unlikely to win that game. We need the Mafia team to be threats to each other.

We normally kill D1 for two reasons, neither of which applies here:
1. To start getting voting information. Bozo has already pointed out that this doesn't work when the mafia don't have teammates. I'm a little bit shocked Bona didn't notice this.
2. Because our numbers advantage will only get worse over time. This isn't applicable here for the reasons I laid out above.

Counterargument:
There are currently 3 teams
after some time, the recruits will turn the game from 7 1 1 to something like 4 3 2 or 3 3 3
At that point, only a small fraction of the players who are currently town have a chance of winning.
If we eliminate a maf today, we eliminate an entire team. That is the main goal in voting today.
This works for the exact reason that there are only one maf on each team.
Starting d2 and onward, maf will likely control around half combined of the players, which will lead to maf having greater control over who is eliminated. Again, the best option for people who are currently town is to kill one maf leader and try to be recruited into the other's faction.

At 6-2-2 is ever too risky to try to recruit someone at that point and better to do a NK? Like imagine you have a 25% of getting outed as mafia and also a chance of recruiting the same person. Even if it becomes 4-3-3 mafia def at that point want to influence votes to target each other so town can stay afloat

[DiplomacyandWarfare]

R
B
B
I
G
T

Hes either giving a code that hes a frog
Or
Hes decided he supports trump

Not sure which

NO POLITICS

[bozotheclown]

Ok I was hoping y'all would just get this but we're not playing mafia.

Several people have already noted that town is doomed. That's true if we play it like mafia, and it's kind of a feature of the setup. The way town wins this game is by pitting the mafia against each other. Townies can also win this game by being recruited, but my reading of the rules has always been that playing to this outcome runs contrary to playing to one's own win condition (the spectrum of positions on this point kind of undermines the game's whole conceit, which is one of the reasons I'm personally not a fan of recruitment games).

If we kill a townie today, we have less ability to effect our will as the game progresses further. If we kill scum today, there's no counterweight to the other scum team. We go into tomorrow in what's basically a brand new 7v2 mafia game in which we have no PRs, scum has twice as much power as a normal mafia team to reduce the number disparity as long as their leader is alive, and every new recruit is going to lack connections to their teammates. Town is unlikely to win that game. We need the Mafia team to be threats to each other.

We normally kill D1 for two reasons, neither of which applies here:
1. To start getting voting information. Bozo has already pointed out that this doesn't work when the mafia don't have teammates. I'm a little bit shocked Bona didn't notice this.
2. Because our numbers advantage will only get worse over time. This isn't applicable here for the reasons I laid out above.

Counterargument:
There are currently 3 teams
after some time, the recruits will turn the game from 7 1 1 to something like 4 3 2 or 3 3 3
At that point, only a small fraction of the players who are currently town have a chance of winning.
If we eliminate a maf today, we eliminate an entire team. That is the main goal in voting today.
This works for the exact reason that there are only one maf on each team.
Starting d2 and onward, maf will likely control around half combined of the players, which will lead to maf having greater control over who is eliminated. Again, the best option for people who are currently town is to kill one maf leader and try to be recruited into the other's faction.

At 6-2-2 is ever too risky to try to recruit someone at that point and better to do a NK? Like imagine you have a 25% of getting outed as mafia and also a chance of recruiting the same person. Even if it becomes 4-3-3 mafia def at that point want to influence votes to target each other so town can stay afloat

Mafia would not be able to perform a NK if tied at 2 members each.

[DiplomacyandWarfare]

The two ways town can win that I can see are to DK the 2 currently mafia D1 and D2, or to get lucky with the mafia randomly trying to recruit each other and then trying to eliminate each other.

Yes. I'm arguing that the former is unlikely and we should thus be putting our eggs in the latter basket.

It may be very unlikely, but we could still take a shot D1. If we are lucky our chances improve D2, and if not, we would still have the opportunity to win by the mafia recruiting each other with one less town.

yes. Accurate.

Ok I was hoping y'all would just get this but we're not playing mafia.

Several people have already noted that town is doomed. That's true if we play it like mafia, and it's kind of a feature of the setup. The way town wins this game is by pitting the mafia against each other. Townies can also win this game by being recruited, but my reading of the rules has always been that playing to this outcome runs contrary to playing to one's own win condition (the spectrum of positions on this point kind of undermines the game's whole conceit, which is one of the reasons I'm personally not a fan of recruitment games).

If we kill a townie today, we have less ability to effect our will as the game progresses further. If we kill scum today, there's no counterweight to the other scum team. We go into tomorrow in what's basically a brand new 7v2 mafia game in which we have no PRs, scum has twice as much power as a normal mafia team to reduce the number disparity as long as their leader is alive, and every new recruit is going to lack connections to their teammates. Town is unlikely to win that game. We need the Mafia team to be threats to each other.

We normally kill D1 for two reasons, neither of which applies here:
1. To start getting voting information. Bozo has already pointed out that this doesn't work when the mafia don't have teammates. I'm a little bit shocked Bona didn't notice this.
2. Because our numbers advantage will only get worse over time. This isn't applicable here for the reasons I laid out above.

Counterargument:
There are currently 3 teams
after some time, the recruits will turn the game from 7 1 1 to something like 4 3 2 or 3 3 3
At that point, only a small fraction of the players who are currently town have a chance of winning.
If we eliminate a maf today, we eliminate an entire team. That is the main goal in voting today.
This works for the exact reason that there are only one maf on each team.
Starting d2 and onward, maf will likely control around half combined of the players, which will lead to maf having greater control over who is eliminated. Again, the best option for people who are currently town is to kill one maf leader and try to be recruited into the other's faction.

At 6-2-2 is ever too risky to try to recruit someone at that point and better to do a NK? Like imagine you have a 25% of getting outed as mafia and also a chance of recruiting the same person. Even if it becomes 4-3-3 mafia def at that point want to influence votes to target each other so town can stay afloat

night kills only work for a maf faction with more members than the other. (ties mean neither can kill)

##VOTE brainbomb

Why did your spreadsheet not agree with what you posted before the game started?

yea, lets pivot to brain
##VOTE BRAINBOMB

[DiplomacyandWarfare]

forming a readlist here:

Town
dipandwar - understands the mech and is trying to find scum
SpiritoftheRadio, bozo - forming solid reads and not being stupid about mech
null line = HB - it would be funny to miskill HB d1 again
Co7by, yoyoyozo - has been lurking, in the sense that theyve done nothing significant enough for me to remember it
Haze with a Z - has been doing MAJOR lurking
ghug - has been arguing for a no kill today, while town under voting pressure as ghug is should be trying to counterwagon their scumreads.
significant scumread (I would like to vote them today) line
bona - OMGUS read
brain - mostly an OMGUS read, also has been using statistics badly and playing badly in ways brainbomb wouldn't do as town. He's too smart for that.
scum

if I forgot anyone, put them in haze's category

[brainbomb]

Over last ten games, town has eliminated a mafia day 1 4/10 tries
If you count sweet being shot by a gun day 1 its 5/10 day 1 occurences in past 10 games.

Thats actually alot better than it used to be

It's almost always in the context of a successful mafia bus, though, which won't happen here.

I wanted to analyze this further and discuss if this was actually accurate. What I found was rather depressing.

https://docs.google.com/spreadsheets/d/1Ckm21aYaToovAgEYJcINx4D8t3p3tPsr8lvGjJ_jCiE/edit?usp=sharing

Out of 30 games sampled, Mafia was killed day 1 in only 8 of them, and one of the samples is a serial killer, not mafia who was caught day 1. another example includes a multiball game.

I also looked into mini games and found that out of 28 games, mafia were only killed twice on day 1.

In the first study, you notice that 6 of the 9 examples however bussing was not the main factor. Ignoring the serial killer game where bussing isnt a possibility, there was still only 1 mafia driving that wagon, thus town led a successful kill on a serial killer without mafia being the main driver behind preventing a miskill.

There were only 3 of the 9 instances where mafia presence on a wagon that resulted in a day 1 mafia death, was actually directly correlated in gross numbers to mafia influence.

So I think ghug is mistaken in his notion. That said I also think the way ghug has argued his point comes from a town mindset.

In other words, hes wrong about day 1 kills of mafia being caused primarily by bussing, but hes potentially being sincere about his beliefs that a no kill is best play here.

[brainbomb]

##VOTE brainbomb

Why did your spreadsheet not agree with what you posted before the game started?

What are you referring too?

[Col7by]

Counterargument:
There are currently 3 teams
after some time, the recruits will turn the game from 7 1 1 to something like 4 3 2 or 3 3 3
At that point, only a small fraction of the players who are currently town have a chance of winning.
If we eliminate a maf today, we eliminate an entire team. That is the main goal in voting today.
This works for the exact reason that there are only one maf on each team.
Starting d2 and onward, maf will likely control around half combined of the players, which will lead to maf having greater control over who is eliminated. Again, the best option for people who are currently town is to kill one maf leader and try to be recruited into the other's faction.

At 6-2-2 is ever too risky to try to recruit someone at that point and better to do a NK? Like imagine you have a 25% of getting outed as mafia and also a chance of recruiting the same person. Even if it becomes 4-3-3 mafia def at that point want to influence votes to target each other so town can stay afloat

Mafia would not be able to perform a NK if tied at 2 members each.

Ah I’m sure that was listed somewhere nvm on that theory then.

[bozotheclown]

##VOTE brainbomb

Why did your spreadsheet not agree with what you posted before the game started?

What are you referring too?

The win percentages in the spreadsheet you posted in the game do not match those you posted before the game started. It seems like you are trying to influence the mafia recruiting. Is that what you were trying to do?

[brainbomb]

##VOTE brainbomb

Why did your spreadsheet not agree with what you posted before the game started?

What are you referring too?

The win percentages in the spreadsheet you posted in the game do not match those you posted before the game started. It seems like you are trying to influence the mafia recruiting. Is that what you were trying to do?

This is the master data sheet everything came from.
https://docs.google.com/spreadsheets/d/1CV_pkSXvBitSqaZKPloE03peyRd8UIqNJ_6YI_YjhsE/edit?usp=sharing

[bozotheclown]

What are you referring too?

The win percentages in the spreadsheet you posted in the game do not match those you posted before the game started. It seems like you are trying to influence the mafia recruiting. Is that what you were trying to do?

This is the master data sheet everything came from.
https://docs.google.com/spreadsheets/d/1CV_pkSXvBitSqaZKPloE03peyRd8UIqNJ_6YI_YjhsE/edit?usp=sharing

That is impressive, but I don't see the mafia win percentages.

[brainbomb]

Winrate Data comes from the master spreadsheet
https://docs.google.com/spreadsheets/d/161ZAz0j4XUJNZnsJnifBUxhNEIkF-QGOqT2R4CDgx6I/edit?usp=sharing

Look for tabs alltime winrate for mafia stats or mini game winrate stats.

[bozotheclown]

Winrate Data comes from the master spreadsheet
https://docs.google.com/spreadsheets/d/161ZAz0j4XUJNZnsJnifBUxhNEIkF-QGOqT2R4CDgx6I/edit?usp=sharing

Look for tabs alltime winrate for mafia stats or mini game winrate stats.

Thanks, but the information has changed since you first posted the link to the spreadsheet.

[brainbomb]

Yes, it appears that for your entry for example, I took your stats from last 50 games by mistake.

in last 50 games you were
7 mafia out of 11 times, as shown on the last 50 games tab with a winrate of .63
ghugs was listed as only having 5 mafia out of 9 tries with a winrate of .55

screen shot provided shows the tab I grabbed from.

[ghug]

Ok I was hoping y'all would just get this but we're not playing mafia.

Several people have already noted that town is doomed. That's true if we play it like mafia, and it's kind of a feature of the setup. The way town wins this game is by pitting the mafia against each other. Townies can also win this game by being recruited, but my reading of the rules has always been that playing to this outcome runs contrary to playing to one's own win condition (the spectrum of positions on this point kind of undermines the game's whole conceit, which is one of the reasons I'm personally not a fan of recruitment games).

If we kill a townie today, we have less ability to effect our will as the game progresses further. If we kill scum today, there's no counterweight to the other scum team. We go into tomorrow in what's basically a brand new 7v2 mafia game in which we have no PRs, scum has twice as much power as a normal mafia team to reduce the number disparity as long as their leader is alive, and every new recruit is going to lack connections to their teammates. Town is unlikely to win that game. We need the Mafia team to be threats to each other.

We normally kill D1 for two reasons, neither of which applies here:
1. To start getting voting information. Bozo has already pointed out that this doesn't work when the mafia don't have teammates. I'm a little bit shocked Bona didn't notice this.
2. Because our numbers advantage will only get worse over time. This isn't applicable here for the reasons I laid out above.

Counterargument:
There are currently 3 teams
after some time, the recruits will turn the game from 7 1 1 to something like 4 3 2 or 3 3 3
At that point, only a small fraction of the players who are currently town have a chance of winning.
If we eliminate a maf today, we eliminate an entire team. That is the main goal in voting today.
This works for the exact reason that there are only one maf on each team.
Starting d2 and onward, maf will likely control around half combined of the players, which will lead to maf having greater control over who is eliminated. Again, the best option for people who are currently town is to kill one maf leader and try to be recruited into the other's faction.

Yeah fuck that. The rule is to play to your win condition: the team you're currently on. This kind of game doesn't work if it's just groveling for recruitment.