|
On August 28 2014 10:50 Holyflare wrote:Show nested quote +On August 28 2014 10:48 Professor Apathy wrote:On August 28 2014 10:45 Holyflare wrote: But if marv is mafia he's pretty straightforward to read and would be an easy point for town which we might not ever reach because you'd be placing him last. Which is why making orders before everyone has spoken is silly. if we assume that you're confident in reading everyone within 24 hours yes, otherwise no. How can you be so confident on being able to figure everyone out within 24 hours to make it the basis of how you want the groups ordered instead of placing them the way BH and I want (except that BH has this one major flaw in his order which is group E). I am. You also only need to be most sure of the first group in the first 24h anyway. No I don't think so. A group you're really sure about should be in spot 3 or 4 to solidify that read as much as possible. You can put a crappy group that won't improve over time in spot 1.
|
Is liking not you the same as not liking you? I really should just stop talking
|
United Kingdom30774 Posts
What... No? A group I'm really sure about goes first so we get a point. Unless it's got someone good in it in which case it goes 3rd or 2nd so that person has time to solve things.
|
Blazinghand
United States25550 Posts
On August 28 2014 10:54 Amiko wrote: @BH:
I feel like your ordering is strongly based on the idea that marv will lead lynches in a good way / scumhunt well.
But, even under that approach isn't it beneficial to put Cell C before D5? That way at least there's some indication as to the authenticity of marv's reads.
You mean something like DEACB? It's plausible, and I would not be opposed to it. Why don't you run on the DEACB platform, and I'll run on DEABC, and I promise to throw my votes your way if you're ahead of me close to the deadline (and vice versa) so that a DEA B/C order wins?
|
On August 28 2014 10:53 mtamburini wrote:Show nested quote +On August 28 2014 10:48 Not really holyflare wrote: Thats why we go DACBE. C for guaranteed 1 scum lynch (unless marv derps) while A and D goes first. B for more risks and E being the most risky group to lynch into. Why not make that group go first?
Are you even thinking If marv is town, he cannot help find scum in first few groups If marv is scum, he (kinda) gets a free pass for not trying since there is literally nothing to try, and nobody lynches marv d1.
Putting C in first is worst idea ever.
|
On August 28 2014 10:56 Not really holyflare wrote:Show nested quote +On August 28 2014 10:53 mtamburini wrote:On August 28 2014 10:48 Not really holyflare wrote: Thats why we go DACBE. C for guaranteed 1 scum lynch (unless marv derps) while A and D goes first. B for more risks and E being the most risky group to lynch into. Why not make that group go first? Are you even thinking If marv is town, he cannot help find scum in first few groups If marv is scum, he (kinda) gets a free pass for not trying since there is literally nothing to try, and nobody lynches marv d1. Putting C in first is worst idea ever.
But its guarenteed no? So why not take a 1-0 early lead.
|
United Kingdom30774 Posts
Just going to sleep and see who else has posted tomorrow
|
On August 28 2014 10:56 Blazinghand wrote:Show nested quote +On August 28 2014 10:54 Amiko wrote: @BH:
I feel like your ordering is strongly based on the idea that marv will lead lynches in a good way / scumhunt well.
But, even under that approach isn't it beneficial to put Cell C before D5? That way at least there's some indication as to the authenticity of marv's reads. You mean something like DEACB? It's plausible, and I would not be opposed to it. Why don't you run on the DEACB platform, and I'll run on DEABC, and I promise to throw my votes your way if you're ahead of me close to the deadline (and vice versa) so that a DEA B/C order wins?
No. C goes 3rd as that is the most pro town way to do it. C literally have the highest % chance to catch scum. A literally dont have as high of a chance. We don't need a 3 - 0 scum sweep. 3-1 is at least less embarrassing.
|
On August 28 2014 10:57 mtamburini wrote:Show nested quote +On August 28 2014 10:56 Not really holyflare wrote:On August 28 2014 10:53 mtamburini wrote:On August 28 2014 10:48 Not really holyflare wrote: Thats why we go DACBE. C for guaranteed 1 scum lynch (unless marv derps) while A and D goes first. B for more risks and E being the most risky group to lynch into. Why not make that group go first? Are you even thinking If marv is town, he cannot help find scum in first few groups If marv is scum, he (kinda) gets a free pass for not trying since there is literally nothing to try, and nobody lynches marv d1. Putting C in first is worst idea ever. But its guarenteed no? So why not take a 1-0 early lead.
Guaranteed AFTER marv actually posts I have all the reason to think you're scum now for being anti town
|
Blazinghand
United States25550 Posts
On August 28 2014 10:58 Not really holyflare wrote:Show nested quote +On August 28 2014 10:56 Blazinghand wrote:On August 28 2014 10:54 Amiko wrote: @BH:
I feel like your ordering is strongly based on the idea that marv will lead lynches in a good way / scumhunt well.
But, even under that approach isn't it beneficial to put Cell C before D5? That way at least there's some indication as to the authenticity of marv's reads. You mean something like DEACB? It's plausible, and I would not be opposed to it. Why don't you run on the DEACB platform, and I'll run on DEABC, and I promise to throw my votes your way if you're ahead of me close to the deadline (and vice versa) so that a DEA B/C order wins? No. C goes 3rd as that is the most pro town way to do it. C literally have the highest % chance to catch scum. A literally dont have as high of a chance. We don't need a 3 - 0 scum sweep. 3-1 is at least less embarrassing.
3-1 and 3-0 are both the same, which is a loss. I don't give a shit how we lose if we lose, I care about maximizing our chance to win. There's no honor or sportsmanship or bullshit like that in mafia, only winning and losing. The sooner you learn that the sooner you'll be good at this game like me.
|
Is this scum-theorycrafter BH or town-theorycrafter BH? I can't usually tell the difference
|
On August 28 2014 10:57 mtamburini wrote:Show nested quote +On August 28 2014 10:56 Not really holyflare wrote:On August 28 2014 10:53 mtamburini wrote:On August 28 2014 10:48 Not really holyflare wrote: Thats why we go DACBE. C for guaranteed 1 scum lynch (unless marv derps) while A and D goes first. B for more risks and E being the most risky group to lynch into. Why not make that group go first? Are you even thinking If marv is town, he cannot help find scum in first few groups If marv is scum, he (kinda) gets a free pass for not trying since there is literally nothing to try, and nobody lynches marv d1. Putting C in first is worst idea ever. But its guarenteed no? So why not take a 1-0 early lead. there's no difference in wether we make that point on day 1, day 2, day 3 or even day 4 for that matter because if we lose 3 other groups we lose no matter if we score that point early or not. Who cares about scoring early if having good people alive later on gives us a better chance to win, thus resulting in C/B/E all for the last 3 spots and starting with D/A
|
On August 28 2014 10:59 Blazinghand wrote:Show nested quote +On August 28 2014 10:58 Not really holyflare wrote:On August 28 2014 10:56 Blazinghand wrote:On August 28 2014 10:54 Amiko wrote: @BH:
I feel like your ordering is strongly based on the idea that marv will lead lynches in a good way / scumhunt well.
But, even under that approach isn't it beneficial to put Cell C before D5? That way at least there's some indication as to the authenticity of marv's reads. You mean something like DEACB? It's plausible, and I would not be opposed to it. Why don't you run on the DEACB platform, and I'll run on DEABC, and I promise to throw my votes your way if you're ahead of me close to the deadline (and vice versa) so that a DEA B/C order wins? No. C goes 3rd as that is the most pro town way to do it. C literally have the highest % chance to catch scum. A literally dont have as high of a chance. We don't need a 3 - 0 scum sweep. 3-1 is at least less embarrassing. 3-1 and 3-0 are both the same, which is a loss. I don't give a shit how we lose if we lose, I care about maximizing our chance to win. There's no honor or sportsmanship or bullshit like that in mafia, only winning and losing. The sooner you learn that the sooner you'll be good at this game like me.
so why do you want to go 2nd so badly? Or else why not DACBE? Why DEACB? You worried your scumminess will show after 2 days?
|
3rd* you worried you dont have enough time to fight off lynches with 3 days?
|
@BH: You answered the ones you felt would be hardest to read and I pretty much agree with you. But, which group do you think will be the easiest for town to read?
It seems like most people feel it's C. If that's the case, it should be in the first three.
Moreover, if there was one group where I felt 100% certain we'd get mafia, I'd probably want that group to third. That way I know: (1) the group is reached, so we get a point for town; and (2) at least one mafia is known and that information can be evaluated in at least the following day; and (3) the reads of two players in the cell are validated -- the only way we get absolute authenticity of a read is if we lynch the mafia (if we lynch a town, we don't know with certainty which of the two survivors was mafia).
For D vs. E, I don't have a huge preference... as I said earlier, ninjabunnies seems to be more active early and less active later on, so I'd like E to be earlier in the game. But, I don't care all that much if it's D1 vs D2 - I think bunnies dropped off activity D2 in Detention when you were hosting for us so D1 is slightly better in my opinion.
For the DEACB / DEACB platforms, I'll think about it. Not enough impression of A/B yet.
|
How confident that you can read nb/lilwade/smurf? Because from what I see, its the top 3 most risky groups in first 3. Bad.
|
|
Here's another way to look at this which I think supports having C earlier in the order
We need 3 correct lynches to win. It doesn't matter whether those wins are D1, D2, D3 or are D3, D4, D5, or whatever. Thus, the most important thing is to order things such that at least 3 cells have the highest chance of success for lynching mafia.
Let's assume for a second that Cell C is the easiest to read (and I can't say for certain that's true but I'll run with it for this example). If the cell didn't have people with high activity/content, I would typically want that cell to be first. Not because we need to start with a win, but because then we would be most likely to get verifiable information earlier. If we lynch correctly D1, then we know one mafia and two town on four days. If we lynch correctly D3, then we only have that information for two days.
However, in this case people are saying that the easiest cell to read is also the cell with players that are likely to give high contributions and sure, that has weight too. But, I feel like if there is a cell that is the easiest to read, I absolutely wouldn't vote for it to be last.
|
On August 28 2014 11:21 Amiko wrote: Here's another way to look at this which I think supports having C earlier in the order
We need 3 correct lynches to win. It doesn't matter whether those wins are D1, D2, D3 or are D3, D4, D5, or whatever. Thus, the most important thing is to order things such that at least 3 cells have the highest chance of success for lynching mafia.
Let's assume for a second that Cell C is the easiest to read (and I can't say for certain that's true but I'll run with it for this example). If the cell didn't have people with high activity/content, I would typically want that cell to be first. Not because we need to start with a win, but because then we would be most likely to get verifiable information earlier. If we lynch correctly D1, then we know one mafia and two town on four days. If we lynch correctly D3, then we only have that information for two days.
However, in this case people are saying that the easiest cell to read is also the cell with players that are likely to give high contributions and sure, that has weight too. But, I feel like if there is a cell that is the easiest to read, I absolutely wouldn't vote for it to be last. don't forget that the reads of those people get better as time goes on as well. I'd rather take 2 days with confirmed reads that are reasonably stable because they're already 3 cycles old than having 4 days worth of d1 reads no matter who that person is that made the reads.
Slot #3/4 gives us a healthy compromise of everything. We get stable reads because they're not just d1 reads, we get some time with the people still alive and we get some time with confirmed reads afterwards.
And I've said this a bunch already but I said I'm feeling good about reading that group as a whole given a bit of time. Like I said, both Marv and Holy are people who are capable to play like townies as mafias and thus need to be given some extra time to make sure of that in my mind, so I really don't want to make the decision d1/d2 despite calling it easy.
|
Cell order should be random. A townie choosing cell order gets us very little, a scum choosing cell order gets them a lot.
To ensure randomness, we lynch in alphabetical order.
|
|
|
|