It's just a little frustration that I've put together the best case I can, articulated it to the best of my ability, and... I get scum-read for it, by the same people that are voting agreeing with my case, it makes no sense

Micro 4 - Joe, agent, trent, GameRager

I don't even have a sure scum read on anybody there

Can we just flip this scum so I can eat my nightkill please

Votus microfish

I don't know about the others, but I myself suspected you for reasons other than your case in Micro and how you presented it...I actually agree with you on suspecting Flocke(and a bit by extension Micro as well) atm.

Also that's part of the game....people will suspect as they will....it's best not to take it too personally and just try to have fun. :)

It happens....I don't have a good read on close to half the people playing atm.

How can you be so sure you'd be the NK? It could be anyone...me, you, someone else....who knows.

Long post (+1,000 words).

Contents

1. Description.
2. Commentary.
3. Suggestions.

(@ZFR, none of the items in bold type that appear in this post are any Votus/Unvotus).

1. Description.

This section is useful for examining the table and the graphics; you may jump to the commentary section if you are in a hurry.


1.1 First matrix

Suspicion \ QuesterMatrix showing the degree of suspicion in wich each subject is held by every participant. Each participant’s allegued view appears as a vertical vector.
Range of values for reads: 1=sure scum. 0=sure innocent. 0.5 = no read or a conflicting read.
Aver. HeatAverage Heat. Average suspicion of the subject as considered by each participant.
Weight HeatWeighted Heat. An indicator that attempts to weight the average heat with the degree of concurrence, synthetising info from both indexes. W_H = Square_root(Heat*Av_Concurr)/num_participants
W. Heat >0.5Weighted Heat 0.5. An indicator that attempts to weight the average heat with the degree of concurrence above 0.5.


1.2 Second matrix

Positive \ QuesterThis matrix shows subjects that are positively considered suspects. Ambiguous reads (0.5) are assigned a value of 0,5. Essentially, values above or below 0.5 are rounded to 1 and 0.
Concurr.Concurrence. Number of participants positively considering the subject as a suspect. Adds nuance to the simple Heat index and also acts as an indicator of votes.
Conc. >0,5Concurrence >0.5 or Hard Concurrence. Number of participants positively considering the subject as a suspect. 0,5 are assigned a value of 0. This helps filtering out noise and may be a better indicator of votes.
Av Concurr.Average Concurrence. Number of participants positively considering the subject as a suspect, averaged for the number of particiants, for comparison to Average Heat and individual reads.
Av Conc>0.5Average Concurrence >0.5. Number of participants positively considering the subject as a suspect, averaged for the number of particiants, for comparison to Average Heat and individual reads.


1.2 Colour coding

Colour codes are almost self-explanatory. For the six vertical variables, the scale goes from green (lowest) to yellow (percentile 50) to red (maximum value). For the 'horizontal' variables, the colour of the lowest point is blue instead of green.


2. Commentary

2.1 Primary and secondary suspects

Average Heat shows five subjects as the ones with more average suspicion. @Micro and @Flocke appear firmly in lead. However, the second matrix adds extra nuances to the situation of the secondary top suspects. Concurrence (people considering the subject as suspicious, including 0.5 values as half a point) is equally high for @yogsloth, @Flocke and @Pooka. However, if we consider Concurrence >0.5 (that is, Concurrence excluding values that are not clearly above 0.5), @Flocke stands out above the rest, with @Pooka as his second. When we consider the synthetic index Weighted Heat, the differences in Heat increase, while, for Weighted Heat >0.5, after Micro and Flocke comes Pooka, with @ConsulCaesar and @yogsloth tied in a fourth position.

@Agent's and @Lift's Heat and Concurrence>0.5 apears low, and it carries on to Weighted Heat. Meaning: the angst about them seems to have dissipated. @ConsulCaesar is drawing less fire and, at this moment, appears surprisingly tied with @yogsloth (a tiny bit less suspicious).

This shows how the complementary indexes may help in adding nuances to the situation of each one.

With the data currently available, lots of padding with 0.5 ('ambiguous read') has been necessary, so the Concurrence>0.5 values are especially useful. Also, they help showing how some subjects like @SPF and @yogsloth (who appear with significant Heat due to a high number of ambiguous reads) really do not have so many participants holding them as primary suspects.


2.2 Potential voting

Concurrence and especially Concurrence >0.5 may be considered as indicators of potential votes. In this regard, @Micro appears cooked, with a Concurrence of 10 and a Concurrence>05 of 9. Which means that 9 participants already regard him as positively suspicious (two more have 'ambiguous reads' of him, which make for the additional Concurrence point).

@Micro's club of fans near contains that of @Flocke (but for @Joe, who gives him only a 0.5), so a little of negotiation might need to happen. On the other hand, @Pooka probably will take fire from @Micro and @Flocke and, with a total Hard Concurrence of 5, anything might happen, since the rest of the crew read him as 'ambiguous', with only @yogsloth allowing a 0.3, making for a Concurrence of 8.5.


2.3 Fun facts

@SPG earns the prize for 'most unscrutable' for flying below everyone's radar. No one gives him anything but an ambiguous read (0.5). Weird!

The most heavy-handed in their estimations appear to be @joppo and @Lift, while the most lenient have been @Carradice and @yogsloth (appart from the old values from RW). The same applies to the number of suspects pointed at.

Most innocently appears @joppo, tied with @agent for lowest Heat but with less Concurrence and zero Concurrence>0.5.


3. Suggestions

@Micro is practically cooked (micro...waved? sorry!) at 9 Hard Concurrence, and should be the center of the discussion. If there is an insufficient number of votes, because of dispersion with the other options, or because of changes of opinion, other promising candidates seem to be @Flocke and @Pooka.

Micro and Flocke share potential voters, while Pooka's fan club is more independent.

All this, before @Flocke and @Micro take the floor for some discussion, that might sway opinions in different directions. Also, people feeling uncomfortable with any of these candidates might want to make a case for alternative options.

Before you mention it: Yes, @RW's opinion does not count for votes. On the other hand, he does not add anything to Hard Concurrence (Conc- >0.5). He will be removed in the next version unless you ask for the opposite.

Hopefully all is as clear as possible. If some of you find this even marginally useful, it will have achieved its purpose. Comments and questions are welcome.

Yeah, actions, not results. Damn no-edit! I hope it was enough to understand what I wanted to say.


While you're at it please also covince my boss to give me a few days off. ;)

Here's my numbers (in case they can be added):

1. SirPrimalform 0.5
2. yogsloth 0.3
3. trentonlf 0.6
4. Lifthrasil 0.7
5. FlockeSchnee 0.6
6. JoeSapphire 0.6
7. Carradice 0.4
8. RWarehall 0.0
9¾. GameRager 0.5
10. ConsulCaesar 0.0
11. Microfish_1 0.8
12. joppo 0.5
13. PookaMustard 0.4
14. agentcarr16 0.2

So in consequence:

Votus Micro

Well, it is 'tomorrow' now. How about you continue?

Meanwhile:
accio FlockeSchnee


Well, not everybody is as callous as I am as scum. Bussing a buddy on D1 is quite uncommon (that's why the WIFOM 'I would never do that as scum' works). It's not unheard of, like yogs pretends it to be, but it isn't common or easy to do either. Most will hesitate to damage their own team on D1 in this way. Usually it isn't necessary either. D1 Town is blind, so a scum has to do a serious blunder to warrant getting bussed.

It is interesting that both yogs and you try to deal in absolutes here. Yogs trying to imply that bussing on D1 never happens. You implying that scum will almost always have at least one buddy on a scum-train. Even on Day 1.
Both absolutes are false and both of you are experienced enough to know that. Which makes me wonder why you try to sell such a tale. Scum will do whatever they deem opportune and it usually involves a lot of WIFOM. So there are no absolutes. Especially not for D1.

As for the 'free night kill' – of course scum didn't get a free night kill in the literal sense. But by no-lynching they got one kill after which we still aren't in a much better situation than we were on D1. They got to kill one of us while we are still in the dark and have no train with a confirmed target to analyze. Everything is still speculation. So by letting it go to no-lynch, they effectively extended D1 into D2.

This is very LAMIST. Why should scum kill you, even if you were Town? At the moment you provide a nice distraction by being unreadable and somewhat scummy. But maybe you're not only scummy but actual scum pretending to expect getting nightkilled.

Why do you rate yogs as towny and agent even as your towniest pick?


So. Micro didn't re-appear yet to explain how an RW-Iso exonerates him. But he has been taken to L-1 in absence. By Consul.
@Consul: are you aware, that you are exposing him to the danger of a quickhammer? You know, the danger that you were so worried about yesterday, when it concerned agent?


By the way, I actually would prefer a FlockeSchnee lynch over a Micro lynch. If Micro is scum, Flocke is most assuredly as well and just didn't dare to bus Yesterday. But there is also the possibility of Flocke being scum alone and letting a Town!Micro train run into no-lynch, because he had been told that it would be bad for Town. And also because Town would keep focusing on Micro today.

Also, when is the deadline? I think it's still quite far off, isn't it?

@Micro: I think it is time that you tell us that 'clearing' by isoing RW you were talking of. If you can. Perhaps even garnished with a claim. You're at L-1 after all.

Doesn't look like my pressure is needed. Micro, where the hell are you?

Yes, but it ends on Monday evening. It might be that some people find it difficult contributing to the discussion during the weekend.

Good post, BTW.

Thanks for the vector, victor!

@Lift will like to know that hard concurrence (=people who regard someone as positively suspicious, and might vote accordingly) for @Flocke has just passed from 6 to 7.

Vote Count

Micro 6 - Joe, agent, trent, GameRager, yogs, Caesar
Flocke 2 - joppo, Lift

Everyone else - 0
Not voting - everyone else

13 Players. Takes 7 to lynch.

Micro is closest at L-1

When you see a vote count, please take a moment to make sure your name is against the correct person.

I've just noticed it's been 48 hours since Micro's last post. Considering some of the stuff he said in that post, it's hard to believe he wouldn't expect questions. Is he just laying low to see if the attention goes away?

@ZFR: Has he said anything about being busy?

Micro has been prodded.

He's online and replied that he's typing now.

I just got on and have very little time
I see I am at L1 in 24 hours, which is frankly, surprising, due to D1's inactivity.

At L1 Shall I claim?
Might as well.

TL;DR summary of the main post that I lost 3+ times includes:
A. the reason Flocke removed his vote from me is potentially explainable as he went
1. "i'm gonna vote just to get a lunch"
2. RW said "What are you thinking?" (para)
3. I said (why? based on my actions you have no real reason--you don't even consider me a main target)
4. He joined an alternate wagon when enough people where online to make the other wagon a reality.

B. RW said he thought I was town in , [url=https://www.gog.com/forum/general/harry_potter_and_gog_mafia_64/post476]476, </span>, [url=https://www.gog.com/forum/general/harry_potter_and_gog_mafia_64/post517]517 565.
B2. He refused--refused, I say--to vote for me--despite being online and able to hammer at the deadline--and must have had a reason. He's too good of a player to not have a reason for allowing D1 NL, and we know he is our only confirmed town. I propose it is because of what he stated--namely that he correctly had me lock-town and would not contribute to a bad lunch.
B3. (,Adventures in, anyone? it was a downright weird book). This waving away suspicions of confirmed townies is bad for town. It is how town lost in at least one of the games I followed, and paying attention to them is how we won in at least one game.
RW suspects Yogs, GR, Caesar and maybe trent. Note 517 again.
B4. 523 is forgotten about. who was neighborised? If no one confirms agent's claim, agent is scum.

C. Yogs's post I mentioned before stinks in part because he lies, he misrepresents, and makes generalizations which he then claims are absolutes. more on this when I have time in 4-5 hours, if D2 is still going on.

D. GR's NAI play looks very off to me, especially how he keeps harping on it being NAI. he did this last game and...was scum. IDR him harping on it soooo much when town.

E. Caesar's D1 immunity has worn off, he looks not good

F. Pooka isn't as scummy to me as he had been yesterday, as he had some decent posts today. He points out that Joe's vote-hopping is highly suspect yet no one else calls him on it.

G. RW was removed from play because
1. He was protecting someone, maybe me.
2. He was a solid lock town; one of the more concise, clear-thinking, detached and logical players through D1. When I tried to do this my first game (57) I was removed from play. You can check the scumchat for proof.
3. He was "safe" as his list of suspects wasn't defined yet IIRC.
4. Random (yeah right).

H. Consider: If I was scum, why in the world would I vote in night chat to remove literally the only player who thought I was town? This would be stupid beyond belief. Literally any other player would have been better from a scum!micro perspective than eliminating a pocketed player. Pooka can vouch for this--he was crowing--rightfully-- in the nightchat for his previous/most recent game as scum (in which I took part) that he'd had a solid vote behind him that would vote as Pooka wanted because I was so deeply pocketed by Pooks in that game.
The same situation applies here. scum don't remove a player who is solidly in their corner. Not on N1, anyway. It is foolhardy, reckless, bad strategy, and downright dumb play. I'd like to think I was too good of a player to be that silly.

I. I think attention ought to be focussed on those who have driven the narative D2 and have been calling for my being scum, deliberately contradicting and ignoring the word/hunch of the only player we know is town.

J. Focus also on those who want to ignore what RW said. This is scumplay.

K. Carradice's chart is too much maths for me atm (though I suspect ZFR adores it, if the math is correct) and I have to be afk for about 30 minutes. Also, I never did see the screenshot of his early-D1 chart. I intend to deliver my version of the rating system which seems to be all the rage here, but for now:

Lock scum:
Yogs, Probably Pooks
Highly likely scum: Pooka,
Leaning scum: Caesar & GR
Leaning Town: Carradice
Leaning Highly town: Lift & Trent

If you (more than 4) want me to claim, I shall.

Unfortunately I really am pressed for time for the next 5 hours.

please respond if you think I ought to claim now or not.

@ZFR VOTE Yogsloth

Why, that's the second biggest vote I've ever seen!

Also I'm severely disappointed that the basis for your claim of being cleared is the faulty logic of thinking a confirmed townie's D1 reads are somehow infallible.

You also claimed in #491 that you think you can prove yourself town. Well, now would probably be the time to do that. You made that claim on D1, long before RW was NK'd so I assume it has nothing to do with your ISO of RW (unless you were already planning to kill him, but I don't necessarily think that).