Game Theory for W-space PvP

Avatar
Blobbing is fine. It's a choice. It's blobbing people and then complaining "why don't they fight this unwinnable fight" that's the issue. Burning people out of wormholes because they don't want to fight 1:4 odds is stupid.
Avatar
The problem is a good FC doesn't need to behind numbers in fleet. You often see FC grabbing a large fleet so when they fuck up it is less likely to result in the death of their fleet. A good FC with a good understanding of game mechanics picks a fleet size with his strategy. Keeping the fleet small will also more likely provoke a counter fleet of the same size or a bit more. Unless the hostile FC is incompetent and calls for a 250 man battlecruiser fleet against your 100 frigates.
Avatar
" But being outnumbered because you gave up on blobbing and the other side didn't is worse."Actually, YOU missed the point, Nyv. Your quote above is a false dilemma, there are more than two outcomes when you log in. It's not "outnumber them; equal fight; be outnumbered" (which is already more than two outcomes), there's also the "get blueballed because no one wants to fight your blob" result. then you add in the option to grind structures instead of ship-to-ship fighting, and there's a fifth option.The real flaw in Two Step's argument is that when he says: " In some sense, the blobs will eventually dissolve themselves, as nobody will be willing to fight them and constant evictions get very tiring (and put a lot more load on leadership types)."--people have been saying that about nullsec blobs for a decade, and they've only gotten larger. He's not missing the point, he's just being naive.
Avatar
and, since you are a graduate student in the social sciences, you too should know that when we make predictive models there is a continual balance between parsimony (or lack thereof) and explanatory power. i epistemologically prefer parsimony, so i enjoyed cann's rat choice model and thought it was the appropriate approach for such an audience.ps: the assumption that all players are rational is not a weakness to the theory, but the weakness (which is really his point) is that self-interest is comes in many flavors and is not a universal human trait (which is why cann said "human motivation is often inscrutable.") in other words, what is rational to you may not be rational to me, and thus problematizes a predictive model based on rationality. an example to convey this point is simply the tragedy of the commons -- and it's such an endemic issue with wormholes that i was a little sad that it wasnt mentioned, but its all good.however, i did think the mad comparison was a little off though since it hinges on the perception of second strike capability (aka, retaliatory attack, ) and simply the fear of death (which we're immortal in eve, so who cares).
Avatar
Because of the stress of living in wormholes, I think there's a strong argument to be made that the dynamics will shake out differently.There are population limits to w-space groups because the logistics of coordinating when you don't have reliable travel between systems means you generally want to keep your core pals in one place and things get crowded quickly. The fact that we can't build stations also keep things small because the logistics are harder.W-space doesn't have the force projection that null has. Remember, BLOOD UNION's massive invasion involved 20v15 dreads. That's a big but not epic null battle. But to set it up, BU spent *two months* sneaking capitals into the system.So while there will always be relative blobbing, the scale of w-space empires and battles will always be smaller and more intimate than the epic politics of null.
Avatar
"Hey guys,i heard about that game theory thing, so i just went ahead and tried to relate it to EO, because i thought it would make me look real cool and sophisticated."
Avatar
This is exactly the problem, we had some shitlords last night complaining that we wouldn't fight their T3 armour gang with our half their size T3 PVE group.
Avatar
I live in w-space, so i was interested in the article from the title. I read the first page, but when i got to the bottom and saw two more pages. No thanks.
Avatar
Very enjoyable article Abis. Kudos!
Avatar
Your prinsoners dilemma scenario payoffs are wrong. Confession / Silence should payoff 3 / -3 favouring the dick. Being a dick always pays off more than being nice in this game (not eve, the game theory game... but its interesting how well it translates....).As you identified the Nash equilibrium is to always be a dick, which is why WHs are full of such friendly people welcoming new explorers.
Avatar
You imply that evicted players are permanently barred from living in w-space, as if they were permakilled. When the truth is that intact pvp corps will always regroup.Evictions weed out the hoarders, who only amass assets and quit when they lose their stuff. Good riddance.If you only keep in your wormhole what you really use in fights (plus a few pve ships), you do not have much reason to fear eviction.All that said, there are corps/allys that steamroll other, smaller corps because they 'dared' to kill one of them and then refuse to fight them (=let themselves be slaughtered) at 1:3 or 1:10 odds when they have gathered their blob. Those groups (like Blood Union) are the pest of w-space.
Avatar
You rightly point out that the evolution of w-space to its current form was inevitable because it is the result of mostly rational choices by human actors.Nobody can change the nature of players/humans, so if there is to be a change, it must be in the environment.What if it became much harder to sustain a large group in the same system? Some thoughts, which are not all my actual opinion on what should be done, as they in some cases even hurt my own self-interest ;)1. Remove re-spawn of cap escalations, so no capital escalation waves spawn again the next 3 days, only once. -> Massive reduction in how many people can have a sustained income from one system, encouraging splitting up into more, smaller groups.2. Change POSes so that housing more than a few players in one POS becomes undesirable/impossible (maybe by restricting number of SMAs anchored or something similar), and reduce number of moons in all w-space systems to 2-10, depending on class. So that people literally have no place to live once their group becomes too large.3. Make hole rolling harder or impossible, so people cannot rage-roll to join a blobfest. Drastic change which would need a massive increase in overall connections or it will kill w-space because you are stuck with short, dead chains for a whole day.This would also add pressure to keep corp sizes smaller because you cannot infinitely roll for a new static with new sites to run.None of this can stop people banding together into massive blobs for invasions. But at least a lot of smaller groups that do not/cannot live in the same system together tend to lack cohesion and usually do not last very long.

W-space and Game Theory

It seems like every month there’s a new thread in the Wormholes section of the EVE-O forums bemoaning the state of w-space PvP. Complaints about a lack of decent fights are common, and blobs, NAP-fests, over-reliance on T3 doctrines, laziness, a lack of conflict drivers, and other issues are regularly named as the cause. Inevitably, these threads descend into finger-pointing, accusations of hypocrisy, and flat-out insults. That is to say, they become typical threads on the EVE-O forums and never really solve anything. Below is an attempt to add some additional perspective to the debate about wormhole PvP, using some ideas that I picked up in my real life studies. In my view, blobs, blues, and all the rest are merely symptoms, and the ultimate cause of the situation we see today remains largely unrecognized.

I’ll say up front that I haven’t been in wormholes from the beginning, so I cannot compare what it once was to what it is now. Having only lived in w-space for a little over two years, I lack a frame of reference for the oft-romanticized “good ol' days” when w-space was new and mysterious. What I do have, however, is a solid base of experience from which to draw, and a real life education that included some classes on game theory. For the uninitiated, game theory is the study of strategic decision making, wherever it is found. I’m far from an expert on the subject, but analyzing the state of w-space PvP is considerably less complex than trying to figure out where to set your product’s price based on the current market and the historical actions of your competitors, which is something I’ve also done.

Game theory can be applied to almost any scenario where decisions are made, and the theory assumes that those making the decisions are both self-interested and rational. The first assumption fits EVE players like a glove, but the second one might be a bit of a stretch. This second assumption is actually a weakness of the theory itself, as nobody acts rationally in every situation, and human motivation is often inscrutable. With that caveat aside, applying some of the concepts of game theory to the question of why so many are dissatisfied with the state of w-space PvP is quite illuminating.

Choices, Payoffs, and Equilibrium Outcomes

Before we get into applying the theory to EVE, I’ll explain the core concept of game theory. Essentially, the idea is that rational actors will always make the decision that furthers their self-interest the most, based on the information they have. All things being equal, if a person is presented with a choice between receiving $5 or $10, they will choose the $10 every time. This concept is inanely simple with only one actor, but when other actors are introduced, it gets significantly more complex.

The classic example of game theory is the prisoner’s dilemma. In this scenario, we assume that there are two people who have been arrested in connection with the same crime, and they are being interrogated separately. The police do not have enough evidence to convict them, so the interrogator gives each prisoner a choice: he can confess and go free while his partner receives three years in prison, or he can remain silent and hope that his partner does, too. If both of them confess, however, they each get two years in prison. The following payoff matrix can be used to illustrate the model:

 

payoff matrix

 

Now, let’s think about the situation briefly. Regardless of what his partner does, if one of the prisoners confesses his payoff will be either going free (1) or two years in prison (-2). If he remains silent, his payoff will be either going free (1) or three years in prison (-3). Unless the prisoner has unshakable faith in the silence of his partner, or there are other, selfless motivations at play, the obvious choice is to confess. Confessing has the same best-case scenario (freedom) but a lesser worst-case scenario (2 years instead of 3 years in prison). The decision practically makes itself.

In this case, confessing is the “equilibrium outcome.” This term means that there is a choice that can be made which is better than any other, and that no unilateral change in strategy will produce a superior range of payoffs. The prisoners could both get the best-case payoff by choosing silence together, but they cannot be completely sure that their partner will remain silent and therefore cannot act collaboratively.

This scenario might seem a bit unrealistic at first, but these sorts of calculations actually happen in real life, whether or not the actors are aware. For example, criminal gangs have long recognized the danger of tricky interrogation tactics, and they react by setting the expectation for silence in advance and threatening violent consequences for confessing. This threat, if credible, makes silence the equilibrium outcome, since freedom or a lesser prison sentence means nothing when you’re dead. In response, the police sometimes offer enrollment in a witness protection program to negate the threat of violence and maintain confession as the equilibrium outcome.

So, that's a lot of words, and by now you're probably asking yourself how in hell this relates to w-space PvP. Simply put, the choices and payoffs inherent in w-space (and really EVE in general) create an equilibrium outcome that many players find unsatisfactory. Using game theory we can model the decision-making process and demonstrate how individual choices lead to the current state of PvP in wormholes. With that in mind, let’s look at blobs and blues through the lens of game theory.

Straight outta J115405, I'm a wormhole resident and director in Hard Knocks Inc. When not writing about wormhole PvP and events in the greater w-space community, I enjoy jokes about Kazakhstan and the occasional glass of delicious tears.