@DDD:
This is hilarious, you're so protective of the sanctity of the game that you'll chose sub-optimum strategy just to spite change. And I never claimed this was optimum strategy for anything besides day one where the town has limited information to go off of
Why yes, I am protective of the sanctity of the game. I come to this site to play mafia, not to act as a mere name in a statistical analysis.
That said, your strategy is sub-optimal even for D1, regardless of the amount of times you claim that ~25% (which is not the actual number, though you are ignoring this) is better than an arbitrary number that is directly proportional to the ratio of the skill of the town to the skill of the scum. If we play well and the scum play poorly, our scum catching % goes up, if both sides play well then it remains about the same, and if we play poorly, but the scum plays well, then the % goes down. Your proposal is only optimal in one of those three situations, and I, for one, will not assume a stupid town and a smart scumgroup when my play is a factor in the numbers. I'd rather play my best and go off of actual skill than pseudo-random chance.
You're partially correct, but the lynch doesn't have to be random, just pre-determined and not influenced by the current game itself. Since nothing datadanne does will move my vote (if I can convince people to adopt this plan) then it meets the criteria of pre-determined.
So you are still ignoring the fact that your stake in this game automatically makes any and all decisions you make biased towards advancing the win condition of your faction. The decision on your part to pick data was directly influenced by the role pm you received from the mod telling you what your role is, and the information you gained from that. Your argument does not hold up in this specific point, which is the crux of your entire plan, and falls apart virtually everywhere else as well.
You shouldn't trust me; however as I mentioned true randomness is not necessary for the theory to hold, only that the subject be chosen and that selection unaffected by present game behavior.
see above
As for sussing out my motives, well I guess you'll have to do this thing called "play mafia" which you seemed real gung-ho on before to try and figure out what my possible motives could be, it appears WLC and CS4 have already started this fun endeavor.
So you are admitting to being anti-town? Thanks for clarifying. You should show by your actions that you are town, not expect others to guess at your motives while you do whatever the hell you want. Mafia is a team game, but you don't seem to be here to play it. I suggest you either get this foolish notion that anyone will agree with you out of your head and start trying to find the scum, or accept the fact that your idea hurts the town and helps the scum, which means you deserve to be lynched for it.
NO IT WON'T, read my fucking posts instead of parroting the same tired lines.
The "tired lines" are correct. Your suggestion is foolish. This will remain true no matter how many times you say we are wrong. No one agrees with you, and the reasons for disagreeing with you hold more water than your plan. Argument from repetition is not going to make you right.
The proposal to lynch Datadanne is contingent upon the day continuing as much as normal so that we do have all that same information, but in the end declaring where your vote would've gone and voting for datadanne anyways
Someone mentioned this already (too lazy to look up who), but what you are suggesting here is that we play out a normal D1, determine who is the scummiest and should be lynched, and then leave them alive in favor of your "random" lynch. How on earth is that in any way productive. It is still a waste of a Day, because if said person we "pseudo-string up" is scum, we will have to waste an entire Day not lynching him, and then have to get around to it eventually anyway. Your plan is counter intuitive and fails to make any logical sense.
I agree, if we all just voted for datadanne right now that would be a terrible idea, but that's not what I'm suggesting.
The methods are different but the results are the same. You still advocate lynching an unhelpful player "randomly" over lynching a scummy player for being scummy. Are you actually reading what you are posting?
Please explain to me how my behavior is that of scum and you're not simply moving your vote on to an easy town target.
Did you not read my question or something? Your method benefits the scum while hurting the town. This is inherently scummy, and therefore makes you more likely to be scum. You are a very vocal proponent of your idea, and are not afraid to get your hands dirty in an argument. How does that make you an easy target?
I understand, but it's only by doing that or a random sample will your data be useful in statistical analysis.
Fine. I will go through tomorrow and pick a site area to take every single game from page two and compile a sample set based on that. We will use that as the unbiased set, because it includes an indexing of games, not just nitpicking.
Correct, I don't have hard verification of this, I wish I did. I only can assume that the experience of others has been somewhat similar.
Mine hasn't, obviously. When I'm not scum I notice that the town has a tendency to lynch scum D1 ~30-40% of the time. It is really quite a bit more nuanced than you are arguing it to be. With a crappy player set you are more likely to see townies acting scummy because they don't know better, but in a strong set you can expect good play from everyone that drastically lowers the amount of scummy actions committed by townies.
I didn't include your numbers because you admit that you cherry picked your data to support your opinion. Now I get that you're suggesting the same of me (not true, but maybe this will assuage your fears), but if we look at the games I've played only which is verifiable if you like the number is 0/6, add on VP's verifiable 1/3 and we're at 1/9 or 11%, still not even half as good as a random lynch.
This is a perfect illustration of my point. The games you played in had a 0% scum catching rate on D1, which is rather poor. However, you need to take into account the types of games played, the level of play, the identities of the scum, etc. to determine the actual reason why there was not a successful lynch D1.
On the other hand, if you look at VP's sample in isolation you have a 33% rate, which is a pretty decent bit better than the random 25% rate you want to shoot for. The same variables apply to VP's sample set as well. He said himself that in the game where there was a successful lynch D1 the townies used strong logic to round up VP's buddy.
You are altering the data and the sample sets to make it look like you are right, when simply viewing in isolation changes the perspective completely.
The Lynch: By lynching a random or pre-determined target on day one the projected success rate is better than the rate from the semi-informed guesswork that usually constitutes day one.
This is a totally unverifiable subjective opinion.
Loss of Information: But how do we prevent the loss of information that pushing a pre-determined lynch would invariably create? We do as I proposed, continue the day relatively normally investigating and poking and prodding. We use a device such as Pseudo-Vote: XYZ to indicate voting intent so that voting trends can be monitored as before. Only when a majority of people feel that enough information has been collected that they cast a final pseudo-vote and really vote for the target.
And then we are left D2 with the exact same scummy player still alive and no information about whether or not he was scum.
Failsafe: Even though we've improved the lynch success rate there's still a strong likelihood of lynching a townie of some stripe. Hence, this is where a selection of target occurs, choosing the player least likely to damage the town's chances of winning. Since this is a theme game with bastard roles it's foolish to attempt to outguess the mod on roles, so the criteria should be scumhunting ability and ability to withstand scrutiny so as not to be run down by scum as an easy target.
So because he's a VI we can lynch him without hurting the town if he flips town? Wrong. We will lose one player guaranteed, and may lose more during the Night. That cuts our majority and puts us a Day closer to lylo. Hardly an example of not hurting the town.
The first two pieces are true regardless of who the selected target is, it’s only the fact that I’ve built a relative safeguard into the process to prevent the town from losing it’s more valuable pieces that datadanne is selected.
The fact that you need to build a safeguard into your plan shows that you do not have confidence that you plan will work, and should therefore not be pushing it.
{Preview Edit}
Ok. I missed a decent amount of stuff while typing that up. I'm still gonna leave it in there even though it doesn't quite apply anymore to show my reasoning. There are a few more bits that need attention though.
@VP:
I don't like you backing out of saying it's random because that seemed to be an important part of your argument, and random in this particular game would yield an 8.33% chance of lynching scum--not 25%. I'm curious where you even got that number from.
25% is the number that I would assume is correct as well. Why is it 8.33%.
@DDD:
Surely you've seen players like Datadanne before... empking, zwet, millar? People easily hop on their wagon because their play is consistently anti-town and annoying and no one objects since they are the most anti-town player in the game and afterwards there's nothing for anyone to say because everyone agreed the play was terrible.
As much as I dislike players like that, I prefer not wasting lynches on them. I find that trying to get them involved usefully in the game and then having them get vigged works so much better (and is so much more satisfying). People who lynch VI's for playing like idiots are just too lazy to find a better choice.
No one wants to push the boundaries of strategy
I'm all for pushing the boundaries of strategy, and working against site meta. I do that all the time because of the indecisiveness of most players on this site. However, turning the game into a statistics exercise is not a strategy, but rather an evasion of strategy in favor of luck.
I feel rather uncomfortable at L-1 (if that is indeed where I am) and promise to put together a cognizant defense of myself over the next few hours and to place questions to other people, but I'd appreciate a single unvote so that a drive-by lynching can't happen and hurt the town, especially so early.
Jumpy scum is jumpy now that he realized how close he is to being lynched. That attitude makes me more convinced that you are the best place for my vote. I will not be unvoting.
Someone who made a point of needing information from day one and my plan hurting that should unvote, because a quick lynch on page four would be the epitome of that problem and you'd be a hypocrite to allow me to be quick lynched while talking down my previous plan.
Guilting people into backing off is foolish and scummy. Especially considering that it is common courtesy to allow players at L-1 to speak in their own defense and claim before deciding whether or not to lynch them.
Oh yeah, by the way. L-1. Claim time.
Because, no matter how you dress it up, that's what the world is. A community of idiots doing a series of things until the world explodes and we all die.