Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Langdon]

Right. Basically, it's random chance but by no means pure chance or equal chance. Stuff you do CAN increase your odds.

When the zombie brak the door, they start attacking random villager until a sufficient number ( equal to your home defence, noramlly) attack you, then kill you.

If this is true, it is in stark contrast to what everyone has been saying on the forums, and much closer to what I've been arguing.

Zombies are distributed as stated above - one zombie chooses random villager, next zombie chooses random villager. Statistically you will get a number of zombies evenly distributed between citizens, but due to the vagaries of the RNG some will get more, some will get less. It doesn't stop at the home defense number - I've seen people with bedrolls get dogpiled by 9 zombies, while the rest of us get 1-2 each.

There are several possibilities:

1) There are enough zombies that make it through the defenses that everybody gets more zombies than their house can defend against. i.e. 200 zombies make it through, there are 4 of you, you get 50 zombies average each. Even if one has a fort (8 defense+4 defense items=12 defense) if you all get 40-50 zombies each, defense doesn't matter, you all die. There is still a vanishingly small chance that fort guy gets 11 zombies while the rest of you get 60 each.

2) Just enough zombies get through the gate that there is an increased chance of survival for those with higher defenses. For example, there are 4 of you in town, and 40 zombies make it through. Someone has a fort (8+4=12 defense) while the rest of you are in hovels (4+4=8 def). If the zombies are evenly distributed, fort guy lives, the rest of you die. There is still a chance that fort guy gets 20 zombies, and the other three gets 3-4 each. Still, the odds are stacked against the other three.

3) The number of zombies that get through the gate is less than the number of citizens in town. In this case you see 1 zombie per citizen, although an unlucky few get 2-3 each. Anything higher than a tent will almost guarantee survival here.

From the point of view of a survivor it looks really random, but it's just the RNG.

Also, each corpse you leave in the town (did not use water to destroy the corpse or drag it out of town) will instakill one citizen come the attack, regardless of defense.

The only thing that assures LMS is if you survive till the next night. If everyone dies, regardless of number of zombies or defenses, then LMS is awarded randomly among the people who "died in their bed" (i.e. you might have more zombies on you, but they were saving you for dessert).

[Zangi]

Also, each corpse you leave in the town (did not use water to destroy the corpse or drag it out of town) will instakill one citizen come the attack, regardless of defense.

Addendum: Saw a non-dragged corpse screw with the water supply.  Lost rations.

[GlyphGryph]

If everyone dies, regardless of number of zombies or defenses, then LMS is awarded randomly among the people who "died in their bed" (i.e. you might have more zombies on you, but they were saving you for dessert).

The problem is that this is not a meaningful statement in regards to what we are discussing (and  most of the rest of what you said was irrelevant).

We know its random. What we want to know is HOW the random outcome is determined. Not all "randoms" are created equal.

If we take turns drawing cards, but I draw 2 on my turn and you draw 1, the first person to draw an Ace of Spades is random. But the odds are certainly not equal.

From the point of view of a survivor it looks really random, but it's just the RNG.

Err... from the point of the view of a survivor it looks random, but it's really just random? Not sure what you are getting at there. :/

[Vactor]

I believe he's saying that everyone who dies on the final night has an equal chance of getting last man standing, regardless of who was the final person to be calculated as being dead.

[forsaken1111]

I believe he's saying that everyone who dies on the final night has an equal chance of getting last man standing, regardless of who was the final person to be calculated as being dead.

And I have to wonder what his source is

[Langdon]

If everyone dies, regardless of number of zombies or defenses, then LMS is awarded randomly among the people who "died in their bed" (i.e. you might have more zombies on you, but they were saving you for dessert).

The problem is that this is not a meaningful statement in regards to what we are discussing (and  most of the rest of what you said was irrelevant).

We know its random. What we want to know is HOW the random outcome is determined. Not all "randoms" are created equal.

If we take turns drawing cards, but I draw 2 on my turn and you draw 1, the first person to draw an Ace of Spades is random. But the odds are certainly not equal.

From the point of view of a survivor it looks really random, but it's just the RNG.

Err... from the point of the view of a survivor it looks random, but it's really just random? Not sure what you are getting at there. :/

If you ALL die, LMS is not determined by how many zombies you had on you or what your defense was. LMS is determined by who was eaten last. This, from my understanding, is randomly selected from all citizens who "die in their bed" on the final night.

My comments are based on posts from German and French players. I understand that the English Die2Nite may very well use a different system, but so far from the posts of people who have gotten LMS (and I myself have LMS from my last town) this system as described by the French and German players seems to be still in place.

Further corroboration would involve a systematic recording of LMS awardees, and a statistical study of LMS selection vs. house defense and number of zombies assigned at death.

[GlyphGryph]

The problem is, all the German and French players I've talked to use the same source you do - "They heard it from someone else", and none of them seem to understand that random, by itself, does not mean everyone has an equal chance.

The developers have said that the Last Man standing is chosen randomly. They have NOT said that things you do can't change the weightings the choices are given - only that it is, ultimately, up the RNG.

The number you roll on 2d6 is random, truly random, 100% random - but some numbers are still more likely to come up than others.

[Mysteriousbluepuppet]

I think it goes like that Zombies are 'awarded' to each player randomly, even on dead players,  and the lms is given to the last to die If you are lucky, the zombies may all well strike the others, and give you the lms, even if you ar in a tent, and they in a fortress. Checking back, on my lms i had 140 zombies on me, and he 220. So i probably got lucky and his defense where overwhelmed before mines

[Langdon]

The developers have said that the Last Man standing is chosen randomly. They have NOT said that things you do can't change the weightings the choices are given - only that it is, ultimately, up the RNG.

This statement is also what the non-English players referred to when I asked them. I believe (and here, I understand if you choose not to believe me) that this statement from the developers "Last Man Standing is chosen randomly" means "all citizens who die in their bed on the final night have equal chances at LMS". I believe that defense does not make a difference (does not change the weighings) in the final roll for LMS.

My belief is based on my interpretation of the developers' statements and from what I have observed on LMS from my own experience and from examining the soul pages of other LMS awardees.  Again, we have no hard data, unless someone runs an analysis on Last Man Standing awardings on a statistically significant number of towns over the course of the beta. Maybe the hoah.org people can help with that.

[Shades]

I'm assuming its not the last one eaten (which would be the easiest implementation for having higher defence at the end of the day) due to the fact it's not always the last one marked as dead who gets it.

However that is not to say there isn't some other random bias.

[Saint]

I have a overly agressive newb in my town, I wrote a strongly worded letter to my town and he's gunna be banished.

[GlyphGryph]

I'm assuming its not the last one eaten (which would be the easiest implementation for having higher defence at the end of the day) due to the fact it's not always the last one marked as dead who gets it.

It's very possible to have the highest defense by a wide margin and still be the first one eaten. It's actually more probable than my winning Hero Mode twice in a row.

And to lose, you don't just have to not be the FIRST one eaten, you have to be the last.

Take the 2d6 example. A player with full upgrades is a 7. There are 11 other players assigned to the other values.
Player 7 is the most likely to win, BUT it is more likely that player 7 does NOT win than it is that he DOES.

Basically, I think that some players are most likely to get LMS, but that doesn't mean that they are actually likely to get it.

And I agree - we need some statistics.

Again, we have no hard data, unless someone runs an analysis on Last Man Standing awardings on a statistically significant number of towns over the course of the beta. Maybe the hoah.org people can help with that.

What do you think would be a large enough number of towns to provide evidence one way or the other?

[Langdon]

And I agree - we need some statistics.

Again, we have no hard data, unless someone runs an analysis on Last Man Standing awardings on a statistically significant number of towns over the course of the beta. Maybe the hoah.org people can help with that.

What do you think would be a large enough number of towns to provide evidence one way or the other?

Depends on the number of towns that have played since start of beta. If we have 1000 towns played since beta, 95% confidence (+/- ~8%) would need about a 100-town sample size.

I've been poking around the soul page display and I see it doesn't record number of zombies or house defense. Ideally there would be some way for a third-party site (hoah.org or someone else) to pull the gazette page upon town death, as the gazette list does have the attack information we need.

Might be easier to ask the devs point blank. :-P

[GlyphGryph]

The devs refuse to say.
They don't like spoiling stuff. ^_^ And they LOVE subtly misleading people (see the "End of Beta") and encouraging player mistakes that are faults of the player's general wisdom.
That is one of the biggest reasons why I inherently doubt the general wisdom of the German/French player base.

[Ozyton]

Poof, back to square one. Let's get started =P

EDIT: First immediate change: the well has more water to start with.