NewbScum Greeting Tell

[Loopdan]

Also "u r a person 2" should receive attribution, as well. That's not me.

[Jingle]

Two things occur: what is the distribution for PR:greet?

2/9 is never really practically viable, numbers wise. To any interested party, the tell starts at 1/4 in a newbie setup because you are necessarily town by wanting to know the alignment of the newbie.

[Loopdan]

Two things occur: what is the distribution for PR:greet?

Interesting question. I didn't track PRs for this.

2/9 is never really practically viable, numbers wise. To any interested party, the tell starts at 1/4 in a newbie setup because you are necessarily town by wanting to know the alignment of the newbie.

A good point that I hadn't thought about.

[Bicephalous Bob]

@Bicephalous Bob - I would like to be cited in your next academic paper as "Prof Hamm T Smiggles III". Thank you.

I don't think I can sell that to the second author, unless that's your legal name.

assuming the sample size is statistically significant.

All of that falls under this. An assumption that the information gathered is relevant. :P

I ran a chi-squared test, which was significant at p < 0.002. I wouldn't take those odds on the correctness of the tell, however. The results are biased because Loopdan was motivated to perform this experiment by looking at a significant part of the small dataset (i.e. he played many of the games), and there are a lot of possible tells he could have chosen based on this. I'd say the tell is still quite a bit more likely to be real than not, though.

[Loopdan]

I don't think I can sell that to the second author, unless that's your legal name.

How about now?

[PenguinPower]

Ouch...expired. Probably should get that updated.

[u r a person 2]

Is it possible to use your dataset to check this tell on a higher sample siZe easily?

[Jingle]

Kinda? It would be man hours alone to backlog newbie games, and presumably someone could fairly easily write a bot to scrape that data without a lot of work. We have the data available, the only obstacle is collecting and organizing it.

[Bicephalous Bob]

just tested the following tells on Normal games:

- greeting count in all posts (limited to hey, hi, hello and howdy)
- whether the first post contains a greeting
- whether the first post contains a greeting that explicitly references everyone. used the following regex: \b(hey|hello|hi|howdy)\s+(guys|folks|all|everyone|everybody)\b
- same as the previous one, but only on posts from 2016 and later, to account for changes in meta

None of these are indicative of alignment. I haven't scraped player join dates so I'm not sure I can test if this is exclusively a tell for new players.

[Bicephalous Bob]

I also tried removing replacements, same result

[u r a person 2]

okay, that's not a precise recreation but that tracks with my experiences outside of newbie games

[Jingle]

God, we're all such nerds.

[mhsmith0]

[Nancy Drew 39]

I think you’re all overlooking the obvious. Greeting the game isn’t a scumtell but a try hard, over-explanatory, detailed post, is more likely to be and both the RC and NMSA examples qualify for this.

Contrast:

“Hey guys, really looking fwd to this game”, to a similar post with excess filler. Also, RC’s example is more likely to come from scum because scum (especially newbie scum) tends to overcompensate, ergo: I will scumhunt with “my fresh eyes”. As for NMSA, it’s the excessive detail that gives it away.

In both cases, there is some form of overcompensation, as if a need to go out of one’s way to let the thread know just how townie, you supposedly are. In contrast, a simple, “excited to play this game”, demonstrates none of that.

Referencing “catching up”, is completely NAI however, mentioning “forgetting about the game” or “not being aware it started” is definitely more suspect. Why? Because once again, it demonstrates a possible need to overcompensate. As in “I’m sorry I haven’t posted but that’s cuz I forgot, not because I have a red role pm”. However, I’ve also seen town do this as well, so not as applicable as the first but I did play a game, where a scum in fact, made this very kind of post.

[chennisden]

It's not the greeting it's the tone that catches newb scum

A lot of people have been learning the wrong lesson from this tell and I'm afraid it's making them learn slower and play worse

[Vi]

surely there's no way I did this

Vote: Query

I agree, he's already suspicious.

vi = 6
6th person on list = avinashv

Accidental OMGUS Vote:

avinashv

Also, two questions while I'm voting for you.
* Where did your username come from?
* You don't mind if I shorten that to "avi" in the future, right?

---

Also, at least I'm not Emacs.

the vote in that second quote seems out of character for me

now

but I'd believe it coming from 2008 me (also avinashv was a scumpartner, which is a different hypothetical tell altogether)
the rest is pretty standard 2008 me

[Nancy Drew 39]

It's not the greeting it's the tone that catches newb scum

A lot of people have been learning the wrong lesson from this tell and I'm afraid it's making them learn slower and play worse

In my experience, either some sort of generally tryhard type of opening or an overcompensating one, is more likely to come from scum than town, especially if it’s excessively long or detailed.

An extremely serious, “I caught scum” kind of post, within the first few posts of the game, is particularly noteworthy. I’ve caught scum twice doing that way too early on, when the evidence for the read is either not at clear or doesn’t involve anything specific to the game they’re in.

The placing of the post is also relevant. Is it during or after RVS? Are they a replace in? Those are both mitigating factors.

A short simple greeting the game sort of post, is both standard for me and also NAI.

[Nancy Drew 39]

surely there's no way I did this

Vote: Query

I agree, he's already suspicious.

vi = 6
6th person on list = avinashv

Accidental OMGUS Vote:

avinashv

Also, two questions while I'm voting for you.
* Where did your username come from?
* You don't mind if I shorten that to "avi" in the future, right?

---

Also, at least I'm not Emacs.

the vote in that second quote seems out of character for me

now

but I'd believe it coming from 2008 me (also avinashv was a scumpartner, which is a different hypothetical tell altogether)
the rest is pretty standard 2008 me

This kind of backs up my point of the scumpost being overly detailed, as contrasted with your simple townie opening. The scumpost also involves things irrelevant to your reason for voting. It’s almost like you’re going out of your way to be respectful of the player you’re voting, which could reveal possible guilt. Also, “accidental OMGUS vote”, sounds really awkward.

[Vi]

This kind of backs up my point of the scumpost being overly detailed, as contrasted with your simple townie opening. The scumpost also involves things irrelevant to your reason for voting. It’s almost like you’re going out of your way to be respectful of the player you’re voting, which could reveal possible guilt. Also, “accidental OMGUS vote”, sounds really awkward.

I mean, in my first game I was literally out for revenge on that player list due to things from offsite; that scum game was like my fifth or something and I was more established and interested in playing with people
and it was an accidental OMGUS because it was the second game post; the first one was avinashv voting their scumpartner like a scummy person

[Ankamius]

Hi guys, let's have a good game yeah?

[chennisden]

Hi I'm Chennis

You can call me obvtown, that works for me.

[Nancy Drew 39]

Hi I'm Chennis

You can call me obvtown, that works for me.

This is hardly a scumtell opening post.

I find, “hi, I’m town” type posts, to be frustratingly NAI.

[Awoo]

You'd be foolish to scumread a player based on this alone

The average D1 scum lynch rate is probably worse than 37%, so I'd say it would make sense to lynch new players based on this alone. In any case, this is the strongest correlation between a linguistic cue and alignment I've ever seen. I'll reference this thread in the next version of my paper. Drop me a message if you want to be cited by another name than Loopdan. I'll also try to replicate this for Normal games sometime in the future.

100% High quality post in 100% high quality thread.

My favourite thing about mafia is when we find actual consistent "reads", like the last thread where we saw how mafia tend to make less posts with more content.

I thought that this was for sure something I did in my newbscum game, but my first post was "test".

[Kublai Khan]

You'd be foolish to scumread a player based on this alone, as a quarter of all Newbie town players do it, too. But it's one tool in finding scum fakeness in RVS. Or at least it was until this thread.

I've found that a large majority of the time scum reveal themselves in the first few pages. If you can scum-hunt from day 1, minute 1 effectively, the game just becomes convincing people that your hunches are right.

[Nancy Drew 39]

You'd be foolish to scumread a player based on this alone, as a quarter of all Newbie town players do it, too. But it's one tool in finding scum fakeness in RVS. Or at least it was until this thread.

I've found that a large majority of the time scum reveal themselves in the first few pages. If you can scum-hunt from day 1, minute 1 effectively, the game just becomes convincing people that your hunches are right.

I was in a game where a scum actually posted that he had rolled scum after a long time of only rolling town and asked for advice. This was his opening post.