Newbie Data

[Elmo]

lol @ Town doing 2% better than random in F11-D and 1% worse than random in C9-D.

Nice job, imo. I'd be interested to see how cop deaths correlate with win rates.

[Elmo]

Oh, wow, I had only looked at the C9 lynch numbers. That's... surprising.

o_O in what respect?

I'd be interested to see how cop deaths correlate with win rates.

You mean Cop deaths after D1? It already drops Town chances to 7.7% or so... not sure how much more you'd learn, though what Town chances are after a N1 Cop death could be interesting.

Where do you get that number from? Think it'd settle some arguments over how much power roles influence the game.

Here are some 95% binomial confidence intervals for F11 as given by this:

• [A] 2 Mafia Goons, 1 Sane Cop, 6 Townies
  
• 2 Mafia Goons, 1 Sane Doctor, 6 Townies
  
• [C] 1 Mafia Goon, 1 Mafia Roleblocker, 1 Sane Cop, 1 Sane Doctor, 5 Townies
  
• [D] 1 Mafia Goon, 1 Mafia Roleblocker, 7 Townies

[Elmo]

o_0 Are people just not reading the first post? F11 set, Day One lynch data: of the 13 games where the Cop is the first lynch, Mafia win the game 92.31% of the time.

Oh, I misread your post, I thought you were saying they won that % when the cop died afterwards. :V

In general it's not likely to be similar since there's a big difference between the cop being lynched D1 and dying N1.

We're somehow winning an average number of those, as Elmo pointed out in post 17, but it's probably through blind luck.

Dunno what you mean by average. I'd say the win rate is a long way from the site average, but (engage the tautological drive!) average from what I'd expect from newbies. I think newbies are special in that it's actually harder for scum to get newbies to lynch worse than random, so I'm not surprised to see something like that... not too sure about that.

I really don't see a particularly good reason to remove F11-D. If you take that out, then it only changes the overall Mafia winrate by about 1.5%.

I mean there are a number of things to be said about balance, some stemming from the fact we don't have a practical definition of it. Shruggo.

[Elmo]

Counterpoint: Mr. Flay's position seems to be that without power roles Towns are successful at finding scum more through blind luck than hunting. (Is that right?)

idk. My position is that the data strongly suggests Town lynches marginally better than random in a F11 game without power roles. Given scum can fakeclaim, scum have to worry about docs when NKing, scum have to worry about being investigated, etc, I don't really draw any conclusions about pure vanilla games from it.

In the general sense I would not be shocked if the balance effect of scumhunting was seriously overrated by most people, tho.

[Elmo]

I wonder if perhaps something in the debate process is hurting our success rates.

There are things in the meta that are absolutely poisonous but meh that should be fairly obvious.

For a while, I toyed with the idea of running what would amount to a semi-hidden prediction market for mafia games. Prolly wouldn't work out tho.

[Elmo]

Well, there's a combination of factors.

1. Unless they have been specifically trained, most people are bad at providing an accurate probability (i.e. 95% likely) even when they have a clear view of the actual likelihood of an event.
2. Almost no-one keeps a written, objective track of their beliefs as the game progresses; memory biases are well known.
3. The vast majority of players perform little or no post-game analysis ("lessons learned").
4. There is a significant amount of variance in Mafia which complicates any kind of analysis; the correct long-term decision can result in a local failure, and it's rare that someone plays enough games
5. It is extremely difficult to assign blame (as in causative effect) between someone acting suspicious and someone suspecting them. People who were objectively wrong to suspect someone have a pronounced tenancy to place 100% of the blame on them.

off the top of my head. What I mean is that when we do something like balance a setup, we are required to put a number on the benefit people can gain from scumhunting in that setup, and I feel this number is, in general, poorly justified at best; you may remember the statistics that towns were getting crushed in Large Games a while ago.

That said, although the majority of players probably overestimate themselves, I'd be pretty confidant most competent players do better than random over the long-term, and some players very significantly so.