Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[webadict]

Oh, hey, Webadict.  Who won KYOSN?

Did that not get resolved? I mean, I can look through notes or whatever, but I don't quite know if I'll have the true answer.

Webadict, you still willing to take a look at game ideas?

I'll look at them, but it'll probably be like months from the point you want an answer.

* 4maskwolf begins rolling on the floor laughing

Hehehe oh man I didn't even remember asking that.

Sounds about right. Just got bored and figured I'd check on how the board was doing, saw my name, and figured I'd reply.

[Shakerag]

Oh, hey, Webadict.  Who won KYOSN?

Did that not get resolved? I mean, I can look through notes or whatever, but I don't quite know if I'll have the true answer.

Considering TolyK up and disappeared on us ... yeah, no resolution.

[webadict]

Oh, hey, Webadict.  Who won KYOSN?

Did that not get resolved? I mean, I can look through notes or whatever, but I don't quite know if I'll have the true answer.

Considering TolyK up and disappeared on us ... yeah, no resolution.

Alright, let's check.

[notquitethere]

Do people think the current BM setup is balanced? Seems that scum still win most of the time.

[zombie urist]

I think its fine. There was a time where town won like 4 times in a row. Recently I don't think people have been that active which is why scum have mostly been winninig.

[Jack A T]

I've been thinking about examining the new BM setup games so far (gathering data and all that).  I was going to focus more on when, how, and how often ICs die, but the info I'll be gathering will probably be of use.

I'll try to get the info together soon.

[notquitethere]

I've been thinking about examining the new BM setup games so far (gathering data and all that).  I was going to focus more on when, how, and how often ICs die, but the info I'll be gathering will probably be of use.

I'll try to get the info together soon.

Cool. I look forward to it!

[Jack A T]

Table here (Google Drive).

Points of note:
*ICs are scum much less often than expected with random chance: 11.5% of ICs were scum, as opposed to the 22.2% expected.
*Scum has won 9/13 games (69.2%).
*The average number of replacements is 3.1.  15% of replacements are of ICs (who are, again, 22.2% of the players).
*Scum lynches D1 are a bit below what's expected from random lynching: 15.3% instead of 22.2%.
*In only 1 of the 13 games (one of the 10 with both ICs as town), both ICs died before D3.  The much-feared rapid IC death rush rarely happens.
*In fact, 43.4% of town ICs make it to endgame (live past the last lynch), compared to 26.4% of other townies.  Without the last lynch counted, it's 47.8% to 38.2%, still a substantial difference in the favour of the ICs.

So, town ICs tend to survive more than town newbies; ICs are almost never all dead by D3 (scum kills others, players lynch others...); ICs are replaced almost as often, proportionally, as non-ICs; ICs tend to be town more often than expected, and scum has won almost 70% of BMs under the new setup.

I think the scum lean is less due to the setup than it is due to a few other things: BMs are full of newbies (who have a hard time scumhunting and are hard to scumhunt), some newbies are not very active, ICs are often not very active, and replacements are very common.

[Leafsnail]

That's somewhat better than the old setup at least (where scum won 5 times out of 6).  We could look into switching to another setup though, like Matrix6 which has been looking quite good statistically.

[notquitethere]

Those are interesting results Jack. Did you reach any fresh conclusion about ICs?

[Jack A T]

Those are interesting results Jack. Did you reach any fresh conclusion about ICs?

Hard to determine much beyond what I've said.  The main point: ICs live a lot more than we tend to think they do.

To help figure out why they survive, I decided to add the reasons for each N1 kill to the table: N1 is before any rolecopping has happened.  Basically, there's usually perceived to be a bigger problem (threat or near-clear player) or safer target for the scum than ICs.  Sometimes, specific ICs aren't even seen as major threats to the scum (lynchable, foolable, inactive, or just on the wrong track).  An urge to avoid killing ICs, at least N1, can also be seen in a couple cases.  In the cases where ICs are killed, they are generally seen as major threats.  Nothing too surprising, but certainly a weaker view of IC play than expected (even for N1).

Honestly, we don't have much information on the impact of ICs.  Perhaps a standardized post-game feedback questionnaire would be useful.

[4maskwolf]

Nothing too surprising, but certainly a weaker view of IC play than expected (even for N1).

I have several theories on this, but they mostly involve reasons why our meta is flawed and why we need to move on from the old ways of doing things, since the scum have learned them and can abuse that.  If you all really want, I can post my opinions on that at some point.

Honestly, we don't have much information on the impact of ICs.  Perhaps a standardized post-game feedback questionnaire would be useful.

Feedback is always good.

[Toaster]

Nothing too surprising, but certainly a weaker view of IC play than expected (even for N1).

I have several theories on this, but they mostly involve reasons why our meta is flawed and why we need to move on from the old ways of doing things, since the scum have learned them and can abuse that.  If you all really want, I can post my opinions on that at some point.

I'd like to hear them.

[Persus13]

Nothing too surprising, but certainly a weaker view of IC play than expected (even for N1).

I have several theories on this, but they mostly involve reasons why our meta is flawed and why we need to move on from the old ways of doing things, since the scum have learned them and can abuse that.  If you all really want, I can post my opinions on that at some point.

I would also like to hear this expanded on, with data and stuff if you can provide it.


General question: How important is voting in your first post of the game? I don't use this anymore because I have found it to be completely ineffective in finding scum, but I have been questioned about this a lot in previous games and was wondering what other people thought about that.

[Jack A T]

I have several theories on this, but they mostly involve reasons why our meta is flawed and why we need to move on from the old ways of doing things, since the scum have learned them and can abuse that.  If you all really want, I can post my opinions on that at some point.

Sure.  Could be useful.

General question: How important is voting in your first post of the game? I don't use this anymore because I have found it to be completely ineffective in finding scum, but I have been questioned about this a lot in previous games and was wondering what other people thought about that.

With the current meta, I see the initial randomish votes as having little purpose.  Everyone knows such votes are non-threatening, and everyone knows they'll be gone by the end of the day.  I could see several randomish votes on a single person spooking them a bit, but that almost never happens.  There's rarely much to analyze about specific votes, too, especially with the number of people who make truly random votes.

If the votes were more threatening (and, for that matter, less random), they could probably be more useful.

We should probably move these discussions over to the Mafia Theory thread soon.

Feedback is always good.

I might try setting up a draft questionnaire after my exams.