The Bandit Dilemma: An Inner Look at the Economics and Psychology of Banditry and the Inevitable Collapse of Day Z

[Chiefmon 93]

I see forum members constantly posting about penalties against bandits. I notice a very wide schism between the Pro-PvP and the Anti-PvP. This post is an ATTEMPT at bridging the schism by explaining the underlying psychology and logic behind being a bandit and by showing the basic flaw in balancing bandits with survivors. NOTE: I posted this thread to both the survivor and bandit boards to get a better idea on others' ideas. As far as what kind of bandits, I'm referring to those who seek survival, not thrill killers. IF YOU ARE A BANDIT WHO KILLS FOR ANY OTHER REASON THAN SEEKING SURVIVAL, THIS THREAD DOES NOT CONCERN YOU.

The biggest perceived problem with the Bandit issue, at least from what the other posters have complained about, is the disproportionate ratio of Bandits:Survivors. While I do not claim to know the exact ratio of bandit to survivors, the way the game is constructed makes it so that the bandit population would rise if the theory I am setting forth is true. I call it 'The Bandit Dilemma'. Here is Survivor Bob. He is scavenging and surviving. He sees a survivor across the street in Cherno. What can he do? He has 3 basic choices: Evasion, Diplomacy, and Elimination.

1. Evasion: Simply avoid the other player. This is often not feasible since they will usually see you first or at the same time. Risk^{Relative chance of loss compared to gain} :None Gain:None Loss:None As you can see, there is no risk in a successful evasion, but cannot be relied on, leading to our intrepid survivor to the other two choices. SUCCESS YIELDS NOTHING TO BOTH SIDES
   
2. Diplomacy: Try and talk to the other and forge an alliance. This is, for many, the best outcome. Both players gain a better chance of survival, but at the risk of death on their first encounter. They could gain a companion, or they could be shot on sight. Risk:Med-High Gain:High Loss: Increasing relative to life length The loss in this one is unusual. If Bob is a new player with just a bandage and a flashlight, death is not a big loss for him. If He has a helicopter waiting outside of town, a gillie suit, and a GPS, death would be a huge loss. So, the more progress He makes in this game, the higher the risk with diplomacy. SUCCESS YIELDS GAINS TO BOTH SIDES
   
3. Elimination: Attempt to kill the other player as expediently as possible. IF I fire first, I have a low chance of death. A n00b with only a Makarov can kill an experienced player with a rifle if they can get a good burst off before the other player has time to respond. This decision will prevent the other player from harming Bob. Also, the other player could have better loot on him. Risk: Low-Med Gain: Variable based on location Loss: Increasing relative to life length Bob open firing on the other player could yield excellent loot, especially if they are further inland where the better equipped survivors loom. This is not discounting the chance of a super player on the coast or a n00b on the NW Airfield, but is instead relying on statistical tendencies. The CHANCE of death is lower than risking diplomacy even though the loss is equally severe. If they BOTH choose elimination, one or both will die. SUCCESS LEADS TO GAINS FOR ONE, DEATH TO THE OTHER
   

So, what should Bob do? Anyone who has taken a Economics class has surely heard of the prisoner's dilemma. It's the same as the Bandit's Dilemma. 2 prisoners are called on to testify against each other. If they both stay silent, they both get 1 year in prison. If they both talk, they both get 5 years in prison. If one talks and the other doesn't, the silent one gets 20 years and the talker goes free. Both are naturally inclined to talk to avoid the 20 year sentence, but they could both get the 1 year sentence if they work together. The problem is, much like with the Bandit's Dilemma, they have no idea what the other will choose.

Early game, the logical choice is to work together. Even if you die, you lost only an hour of play. Late game, everyone will logically choose to kill each other. If you die, you lost 5 hours of play. I myself adopted a shoot-on-sight policy when I went to the NW Airfield because I didn't want to lose all my stuff. (Thankfully it was abandoned, so I did not have to kill anyone.) For the good of both, they should work together. Late game, an extra sniper providing cover fire is a great advantage. But what if he shoots you instead? He would get all your stuff and you'd respawn kicking yourself for trying to be diplomatic.

Here's the flaw in balance: The more items a player gets, the more reason they have to kill anyone who gets near them. The less they have, the better reason to be diplomatic. So we are stuck in a world where, through natural selection, Bandits are well armed and the diplomatic are either constantly killed or are forced to take up banditry as a form of self defense. If EVERYONE stopped being bandits, everyone would prosper. But now we are moving toward a exclusively PvP world. When the ONLY way to become fully armed is through banditry and shoot-on-sight tactics, the bandit population will only grow.

THE ABOVE IMAGE WAS NOT MADE BY ME. I FOUND IT WHILE SEARCHING THE FORUMS. IT DOES NOT NECESSARILY REFLECT THE TRUTH, BUT I FOUND IT TO BE AN AMUSING VISUAL AID. IT IS NOT SCIENTIFIC IN NATURE.

Back to survivor Bob:

B=Bob S= Other survivor D=Diplomacy E=Extermination

BD/SD- Both call out friendly and salute. They team up and exchange supplies. The other survivor is low on food but has tons of soda, Bob has tons of cooked meat, but is low on water. They both benefit and continue on, covering each other and scavenging. Working together, they eventually get a car repaired and set off on a grand adventure.

BD/SE- Bob waves and is promptly shot in the face. Bob dies.

BE/SD- The other survivor calls out friendly and Bob empties a M1911 into his chest. The other survivor dies and Bob picks his corpse clean of supplies, making out with more food and water than he could consume. He comes across a repairable car later, but doesn't have the time or inventory space to repair it.

BE/SE- They open fire on each other. Bob kills the other survivor and survives with 1000 blood and a broken leg. He doesn't last long against the zombies that heard the gunshots.

That is the Bandit's Dilemma in a nutshell. This does not take into account the jackasses who kill for fun. Seriously, those guys are asshats. WHO KILLS A SURVIVOR 5 SECONDS AFTER HE SPAWNS!? YOU HAD NO REASON TO WASTE A SNIPER ROUND ON ME!* Also, for those interested, just gonna put this here:

Yeah.

I AM NOT ARGUING FOR THE OMISSION OF BANDITRY FROM THE GAME. I am simply suggesting that the current situation is not preferable. This is also meant as an analytic piece.

The problem is that, IF the current trend in increasing lifespan holds true, banditry will get worse and the game will become almost exclusively PvP.

People like you need to shut up.

Just the title alone ... "An Inner Look at the Economics and Psychology of Banditry". Are you kidding me? It's a fucking game.

To which I reply, do we not act logically in a game as we do in real life to reach our goals? I am omitting the morality from the equation completely and am putting a variable on the value of the individual's life. Why do you kill? If it's for the thrill, then this does not apply. This is all in context of seeking survival.

UPDATE; After doing some continued thinking, I realize that the above problem could lead to the collapse of Day Z as we know it. If we think about the Bandit vs. Survivor issue in terms of Natural Selection, there is a massive problem.

Imagine, if you will, an island in the middle of the ocean. On this island are always EXACTLY 100 bunnies. The bunnies are born either docile or violent. The violence is a relatively common mutation that occurs around 25% of the time. Every time a bunny dies, their descendant replaces them instantly. So, the violent bunnies will seek to kill ALL other bunnies, docile or not and doing so will help them survive against other violent bunnies. If we start out with only 10% of the bunny population being violent, it's not that bad. Some bunnies die, but the natural order is kept in place.

Over time, the dead bunnies are replaced by their descendants. If the violent bunnies have a natural advantage, then they will live while they kill their brethren. So we have slots being taken up by the veteran violent bunnies, with no room for the new docile baby bunnies who will be slaughtered upon spawn birth. Eventually, when the ONLY way to reliably survive is by being violent, none of the bunnies will be docile. Those that are will keep dying until their descendants become violent. The happy bunny island is now a massive bloodbath.

When players see only bandits flourishing, they too will become bandits, if only out of self defense. Right now, it's not so bad, because there is a steady stream of new players flowing in, keeping the balance. I've noticed that few veteran players are pacifists. When the flow stops, we will be stuck with servers full of people who all want to kill each other for resources or out of fear of being killed. Once again, I am not arguing for the omission of Banditry. It just needs to be fixed, fast. If EVERYONE is PERCEIVED as a bandit, everyone becomes a bandit to defend themselves. There is too much potential to have the situation snowball and cause the PvP in the game to explode. It will no longer be PvE with some PvP, it will just be a massive free-for-all in a sandbox... with some zombies.

For those of you who think that would be awesome, it'd suck for the bandits too. You will no longer have easy prey. Every survivor you come across will fight back. It's a classic example of too many predators, not enough prey.

What should we do? I have no idea. I wish I could claim to have the perfect solution, but I can't come up with a single answer that would please everyone.

Also, if anyone has any statistics from the past concerning the bandit concentration, PLEASE TELL ME. It would be great if I could graph this out and figure out an approximate logarithmic formula to go with it.

Found some: http://dayzmod.com/forum/index.php?/topic/32425-week-of-stats-show-interesting-pvp-trends/page__hl__vipeax

I was very surprised by the statistics. It is definitely worth having a look at. I do feel that the influx of new players would contribute to an atypical sample, but it is still solid evidence against my theory.

Found some more courtesy of Vanchelon.

Been keeping track of DayZ stats on a weekly basis since 25th of June (back when we had ~300k unique).

The Survivor/Bandit ratio has remained fairly even, with a decrease coming with the huge wave of new players last week and a further decrease this week.

From the 25th of June to 22nd of July, the ratio has been:

18.15%

18.78%

18.44%

17.91%

17.68%

This may appear to be an encouraging trend, but I must say it is not. IF we assume my Bandit's Dilemma theory to be true, then we can assume that MOST of the new players coming in would not be bandits at first. This throws off the math of this experiment MASSIVELY. The theory includes that the better a player gets at surviving, the bigger the logic behind banditry. I wish I could see a chart of players who have played longer than a week in terms of banditry rates. The fact that the banditry rate DIDN'T completely plummet when the population of DayZ doubled due to incoming new players is worrying. In my experience, THE VAST MAJORITY of new players start off the game trusting others and few jump directly into banditry. This means that last week's n00bs are turning into today's bandits. When the incoming pool of new players dries up we are looking at a classic example of an Economic Bubble.

This is my favorite part. "The data and trend do not support my claim, but they will! You just wait and see! So lets ignore the data and instead I'll just throw out some random speculations and baseless predictions but since my post has some charts they will seem more valid."

Why don't we just wait until the data exists and do an actual analysis? Since none of the data or trends seem to support any of the claims you're making, don't you think it's a bit premature to start drawing ridiculous conclusions about the "inevitable collapse of the game?"

The recent influx of players isn't that massively imbalanced compared to where it was a month or two ago. The game went from 75k to 150k and 150k to 350k with just as much fervor and that was well long enough ago for these people to have fallen victim to your inevitable-banditry theory, yet banditry and murder rates continue to decline.

What I am saying is that it DOES support my claim. This is me clarifying how the data relates to my theory because, at first glance, it appears as though it is contrary to my theory. Please note that this whole theory is based on the following postulate: players start off inexperienced and don't immediately jump into banditry. THIS COULD BE WRONG AND I ACCEPT THAT.

Secondly, to wait for the problem to appear is to be ignoring the purpose of prediction. To do what you are saying would be like, and this is hyperbolic comparison, to demand that we wait for water levels to submerge america before we decide whether or not the sea levels are rising. I am working with the data I have to draw my conclusions: game theory, logic, basic psychology, banditry rates over time, population rates over time.

Thirdly, the current banditry rates, as of 9:34 PM CST is at 17.28%. This is APPROXIMATELY 1 in 6. Do you think that 1 in 6 FIRST TIME PLAYERS kill another player in that life? I personally doubt that.

THE FOLLOWING IS SIMPLY AN EXERCISE IN MATHEMATICS:

If I had a good estimate of the banditry rates of first time players, then I would have much more evidence for my theory, but for the purposes of this exercise, let's ASSUME that it is closer to 1 in 12. Now let's assume that first time players make up 10% of the current population. The math leaves us at a basic experienced bandit rate of 17.59%. THIS IS NOT A BIG DIFFERENCE. However, the variables I am giving are hypothetical. My point is this, we are dealing with a system that has evolving variables. The way most people would view the statistics are as flat rates, but with a growing base value it leads to deceptively** low rates.

Another hyperbolic example:

There is a virus that takes 1 week to gestate before symptoms appear. It appears in a town. The town has a population of 300 people. 3 people, or 1 percent have symptoms appear every day. Suddenly the town's population doubles. Now there are 600 people but the rate is still 3 people a day. You would claim that the virus's infectiousness was going down. I would claim that it would catch up in 2 weeks. THIS IS THE POINT I'M MAKING.

**Before any English-Major smartass points it out, yes, I know that 'deceptively' is a meaningless word. However, it is contextually meaningful in this situation.

*This is meant as ironic humor by introducing a burst of emotion into an otherwise ambivalent post. If you didn't catch the humor, I'm sorry.

Edited July 26, 2012 by Chiefmon

[malphos101 71]

The biggest problem with the Bandit issue is the disproportionate ratio of Bandits:Survivors. Right now I would have it pegged at 6:4.

Anecdotal evidence used as the base for a pseudo-analytical look at a perceived problem? Seems legit.

EDIT:

The problem is that, IF the current trend in increasing lifespan holds true, banditry will get worse and the game will become almost exclusively PvP.

That's exactly what rocket wants in this game. He doesn't want a pve game. He wants everyone who isnt part of your group to be considered part of the giant FFA pvp game. If you choose to try and be diplomatic then you have a chance of making your group stronger, but there should never be any guarantees. The one thing I do agree with is that there is little reason to ever want to team up because survival is just too easy. For the solution to that there is another great thread that I think everyone should read before they make yet another "FIX BANDITS PVP" thread.

http://dayzmod.com/forum/index.php?/topic/20112-the-only-realistic-way-to-prevent-deathmatching-make-dayz-a-living-hell/

Edited July 25, 2012 by malphos101

[Chiefmon 93]

Anecdotal evidence used as the base for a pseudo-analytical look at a perceived problem? Seems legit.

You are right, I shall omit the anecdote from this post. Please save the sarcasm for someone who deserves it.

[Foogals (DayZ) 61]

Good points, while I do not believe it nessecary to remove banditry at all (Honestly, it makes the scares and excitement.) I do believe in it's current status it can be summed up by asking someone "Do you want to be the wolf or the sheep?". Plus I've been reading over all the threads highlighting how good people are at killing other people who had no intention or method of harming them. And then laughing, Laughing over the corpse of an honorable human being who actually tried to make their time in this game just a little bit better through teamwork and kindness which they willingly tried to share. Honestly, i'm worried. Truly scared at the way some of these people brag at how good they are at and how much they enjoy making other people miserable. Lot of hidden psychopaths playing this game I'd wager.

Edited July 21, 2012 by Foogals

[DarkRaven123 602]

There is already a "fix" to the "evil bandits."

Get some friends.

[A Classy Emu 24]

Rational/logical discussion on the general boards? This isn't the boards I used to know. However prepare for hardcore flaming because people are stuck in their mindset and won't give any ground whatsoever for any reason at any time. Because, as we know, everyone is either a QQ carebear or bloodthirsty savage who likes to ruin people's fun :rolleyes:

[gizm0 96]

you have my beans

[Riiquiem 35]

Why do you whine about bandits Oo Just shoot back and learn to how play and avoid them if you have to.

Seriously people grow some nuts.

And all that psychological bullsh't... it's a game!

[Mrksjke 60]

Hey, OP!

That was a pretty good post to read, hope you won't mind me translating it to Russian and posting it in our general discussions section?

Here's a link.

Cheers! May the Beans be with you.

[gizm0 96]

Why do you whine about bandits Oo Just shoot back and learn to how play and avoid them if you have to.

Seriously people grow some nuts.

And all that psychological bullsh't... it's a game!

wow, just going straight to flaming.

[noobfun 87]

yay game theory :P

i usually go with the home owner and the burglar myself, its more apt

"that the current situation is not preferable." says who? we the players made it that way, becasue we the players are real people and it became a way of saving our hardwork/get free toys, its become habitualised so now some people dont even care and just grab a lee enfield and sit on the roof at churno saying i r sinpaz *pow pow*

And all that psychological bullsh't... it's a game!

with a person playing it, and people have psychology

[Daddy'o (DayZ) 146]

The only reason pvp is so abundant in this mod is because it's the only threat and goal.

As I've said in 100 other threads, surviving doesn't take anything. You can do it easy by yourself solo. There's no thrills do be had from that after a very short while.

Only excitement from surviving you ever get is from hostile players.

And then when you, alone or a group, have everything, rarest guns and items and what ever, whacha gonna do? Nothing left to do but find people to kill and get thrills from pvp.

This mod is nothing without it and so you find it everywhere.

Edited July 21, 2012 by Daddy'o

[phorest 54]

People like you need to shut up.

Just the title alone ... "An Inner Look at the Economics and Psychology of Banditry". Are you kidding me? It's a fucking game.

[bonx 8]

Interesting reflections but you go to far. It is a video game (and a popular one, so full of jackasses), people does not care at all about any kind of logic or reasoning. They shoot other players like a pong send back the ball. End of the story.

[perestain 33]

People like you need to shut up.

Just the title alone ... "An Inner Look at the Economics and Psychology of Banditry". Are you kidding me? It's a fucking game.

crawl back under the rock where you came from please

[phorest 54]

crawl back under the rock where you came from please

Shut your face baddie.

[01da 218]

crawl back under the rock where you came from please

Mr serious pants

[perestain 33]

Interesting reflections but you go to far. It is a video game (and a popular one, so full of jackasses), people does not care at all about any kind of logic or reasoning. They shoot other players like a pong send back the ball. End of the story.

Just because some players are too simple minded to grasp them doesn't mean game theory doesn't apply to games.

Prisoners dilemma is a simple example from game theory, not originally from economics. Economics is just an application for it.

You don't need to understand underlying mechanics to play, but it helps with educated decision making. Thats why some people are "mysteriously" more succesful than others in a lot of scenarios like this.

Edited July 21, 2012 by perestain

[Catalogue 39]

Take My Beans for a good post, but i openly believe banditry too be a good option towards surviving in this world.

Hypothetically you prove a point, but if were to pull some rough statistics up the chances of BE/SE would be much higher than BD/SD, along with the fact that if the other survivor often chooses to be Diplomatic but always or more often encounters players that choose to eliminate, they will be inclined to change there ways for survival, slowing ruling out the "friendly" gene as such.

Beans for thought.

Edited July 21, 2012 by Catalogue

[spammy@live.de 19]

Interesting reflections but you go to far. It is a video game (and a popular one, so full of jackasses), people does not care at all about any kind of logic or reasoning. They shoot other players like a pong send back the ball. End of the story.

This. The prisoner's dilemma is missing the point here.

Every smart survivor/bandit will have his back covered by a sniper, and diplomacy becomes the best option for survival. The end.

This brings us to the real problem: If the bandits are "jackasses" who just want to shoot stuff and don't care if they survive or not, the whole issue becomes moot. They will open fire anyway.

[Ankhenaten 111]

http://en.wikipedia.org/wiki/Template:Game_theory Edited July 21, 2012 by Ankhenaten

[01da 218]

Take My Beans for a good post, but i openly believe banditry too be a good option towards surviving in this world.

Hypothetically you prove a point, but if were to pull some rough statistics up the chances of BE/SE would be much higher than BD/SD, along with the fact that if the other survivor often chooses to be Diplomatic but always or more often encounters players that choose to eliminate, they will be inclined to change there ways for survival, slowing ruling out the "friendly" gene as such.

Beans for thought.

You just had to color your text?

[Ankhenaten 111]

You just had to color your text?

Makes it easier to find a post, until everyone does it . :)

[Catalogue 39]

This. The prisoner's dilemma is missing the point here.

Every smart survivor/bandit will have his back covered by a sniper, and diplomacy becomes the best option for survival. The end.

This brings us to the real problem: If the bandits are "jackasses" who just want to shoot stuff and don't care if they survive or not, the whole issue becomes moot. They will open fire anyway.

I seem to be missing the point of this post, What is it?

that the prisoners dilemma doesn't apply correctly to the situation?

EDIT:

Makes it easier to find a post, until everyone does it . :)

Thnx :P i like the colour Cyan and have ever since it was my room paint when i was 8 :)

Edited July 21, 2012 by Catalogue

[01da 218]

I seem to be missing the point of this post, What is it?

that the prisoners dilemma doesn't apply correctly to the situation?

EDIT:

Thnx :P i like the colour Cyan and have ever since it was my room paint when i was 8 :)

IMA SPESHUL BUTTERFLY IMA COLOR MAI TEXT