A few other points to consider:
- He could be town and lying to attract a night-kill which would protect the real cop.
- There's no other role which would be more likely so implying 1/11 is a small chance for the claim to be true is misleading.
- It would have been better if scum did not know he was the security guard.
- If d3mon is not scum, you're letting some other scum use his role.
Based on the points you have chosen to respond to, I think you missed the point of my dividing my analysis into two separate sections: one that assumes d3mon is scum, and one that assumes d3mon is town. Some of your responses show up in my 'if he's town' section.
Either way, I'll respond to them here:
@1: While possible, that would require him be left alive today to be their bait at night. Per my post, leaving him alive in any form with his behavior suggesting scum THIS strongly is extremely risky.
@2: Yes every role is equally likely - however having THIS specific role when people were saying he'd need a major one to survive reduces the field. My point could have been paraphrased as saying: "It's very convenient for him to have this important role when everyone was saying he would have to have a major role to be potentially worth keeping alive"
@3: I agree it would have been better if he never roleclaimed with this role. My point here is that once he STARTED his claim and was facing suspicion, he could/should have claimed SOONER instead of us having to drag it out of him. The manner of his eventual claim has done him no favors. Also note that under my "If he's town" list, I addressed your statement; that if he actually IS security guard, it would explain some of his reluctance to roleclaim. Though I still have difficulty understanding why he would then choose death over claiming if he had THIS important of a role.
@4: Generally speaking, any non-scum lynch on day 1 (including no-lynch) lets some scum use their role. My earlier post specifically dealt with evaluating scenarios for testing d3mon, and had to acknowledge in pros/cons that if he IS scum - possibly scum with a role - letting him get to night 1 for the test would mean letting him use that role.
If the prospect of 3 dead townies in a row in so important that it deserves emphasis, consider:
- Day 1 - Wrongly lynch our security guard
- Night 1 - Night-kill by scum
- Day 2 - Another wrong lynch
- Night 2 - Night-kill by scum
The day listings were again evaluating what could happen if we no-lynched to test d3mon on day 2. The scenarios all assumed a no-lynch on day 1 in order for the test to occur, and thusly started their counting with nightkill 1. Your list here is a valid scenario if d3mon is town AND we lynch him today - but it also assumes that we get the day 2 lynch wrong. I listed scenarios with us failing the day 2 lynch as possible casualties of testing d3mon, because d3mon would be picking our pool of targets while under suspicion - either via the guard role or via scum picking people. I included false positives (townies indicated by guard as being possible scum because they moved) as a scenario because the security guard role from last game produced false positives: EM and Kevin showed up moving and ccabal SHOULD have shown up but didnt due to mod error. The security guard role is likely to see as many town as scum move, if not more: 3:1 if you want to base it off of what CD should have seen night 1. It cannot be ignored as a scenario. If we lynch a false positive while testing d3mon, we'd be likely to then lynch d3mon for 'proving' wrong - and if d3mon was actually town, that would be ANOTHER town lynch and probably game over. The point is that if a townie is lynched in response to letting d3mon pick the day 2 lynch candidates, we will still be stuck on day 3 debating if he is scum or not - is he really a guard with real results and the townie was a false positive, or is he scum that just tricked us and is going to give us another bad target?
You have a consistent preference for no-lynch, and that's an understandable position on day 1 - there is indeed no solid information to be had. Day 1 of last game we didnt really have anyone stand out as likely scum based on their actions; this time we do. You have repeated that that he has 3/11 odds of being scum. That is the literal truth, but it assumes no other information - the raw numbers do not take his behavior into account. If you have four closed boxes containing three dogs and a cat and one of the boxes meows - that box is still 1/4 likely to contain the cat, but which would you pick?
-Preston