Crash Course: Strategic Thinking

Avatar
Blue everyone
Avatar
A quote from Bernard Manning... hmmm...
Avatar
Shoot them anyway.
Avatar
Could you go more in depth please?
Avatar
Yeah, I was wondering that as well. If he's quoting the Northern English racist Comedian. (Look him up on wikipedia). Well I'll let the viewers decide. lol
Avatar
There is only one True Metagame.It is played by two sides: Players and Game-Creators. The two sides alternate moves.On the Players' move, they try to figure out how to have fun playing the Game.On the Game-Creators' move, the GCs make some changes to the Game, to stop the Players having fun.The True Metagame ends only when the last Player abandons the Game.Both sides always lose.
Avatar
This article correctly describes the Prisoner's Dilemma.But is the simple, non-iterated PD the best model to use for EVE? Or is the Iterated Prisoner's Dilemma better?http://en.wikipedia.org/wiki/E...Or maybe sometimes, or for some purposes, the one is better, and other times, or for other purposes, the other is better?
Avatar
Ah, but do they know that we know that they know what we know about their plans?
Avatar
The problem with that is not that it would ruin realism, but that it would destroy player freedom, which is a core tenent of EVE.
Avatar
Some really interesting examples of Game Theory and its application:http://www.youtube.com/watch?v... (an excellent use in a game show environment)http://www.academicearth.org/c... (an open course on Game Theory from Yale)and Iocane powder.
Avatar
I'm eagerly awaiting the CSM lobbying for a forced blue mechanic.
Avatar
And that is where you have to start and define "winning" or "success" in Eve. For some it may be sitting in a nullsec system surrounded by two blue regions and farming to their hearts content all day. For others it may be to have red entities next door and be as close as possible to a potential fight when you turn on the game.Obviously, having all of nullsec blue would be detrimental to the second groups definition of fun.
Avatar
Except in Eve, someone's 'best interests' aren't just a carbon copy of real life - acquire food, don't die. Fights, in spaceships or politics or isk wars, are fun, they're why most of us play the game - they're an essential resource. Alliance morale can depend on gudfites; to crush it crush their fleets without giving them a chance, to piss off invaders blueball them in a station, and when the fights run out, people get bored and things go wrong. A rational actor in Eve doesn't just seek a bluefest because it doesn't just want to survive, and the purpose of its existence comes from having stuff to shoot.
Avatar
Game Theory is very inadequate to explain any phenomena in EvE. Well I dont understand what Game Theory is good for anyway. Its attempt to provide a forecast on which choice people will make based on an inadequate understanding how rationality works. If you want to use a Theory to explain issues in a game like EvE use System Theory instead of Game Theory. Well it is not that evident because System Theory has no "Game" in its name but in return it provides much proper insights in human decision-making. And thats what EvE is all about.
Avatar
You are, of course, correct. Different "victory" conditions are what influence decisions here. I don't think anyone in Null wants ALL of null to be blue.
Avatar
Why do I pay my taxes?I haven't visited an NHS funded doctor or dentist in the last 4 years, I haven't been to hospital in the last 10 years, my house has never burnt down and been saved by the fire brigade. I have seen no material benefit from my government's interventions and wars in other countries and I am no longer in school.I pay them because I believe that I am part of something bigger then myself, something that requires joint effort of people doing something they aren't fond of for a "greater good". The CFC/HBC are having a profound change on Null Sec in terms of how null sec alliances see themselves, if things continue the "elite Pee vee Pee" ~~e-honoure~~ days will be a distant memory, and good riddance I say.Humans always group together in larger and larger social groups to seek protection of their way of life from others with different ways of (which often involves you imposing YOUR way of life on others). EVE just reflects this.However that is a whole different topic :)
Avatar
The Prisoner's Dilemma is used as an introduction to Game Theory as it is easy to understand. Each player only has two options, with 4 possible outcomes. It's easy to see and therefore it's easy to understand why rational actors always pick to snitch on their fellow prisoner.I agree that there are "better" Game Theory models, but I want to try and keep the articles punchy and simple.
Avatar
"Rational actor" doesn't mean "they pick rational decisions", it means they know accurately what outcomes provide which payouts, they know their options, they know their opponent's options and they know what their opponent knows (and on and on).Your payout may be that you start a war and your opposite number's may be that they prevent one. Game theory still applies, it depends on whether or not the game you are playing is a "Zero Sum Game", which essentially means only one player can receive a pay out.
Avatar
Thanks for those!
Avatar
:(
Avatar
"elite Pee vee Pee" ~~e-honoure~~This really hasn't been a dominating thing in nullsec since BoB, which was years ago.
Avatar
omg so bad you erred on the payoff matrix and your analysis is a joke. It's 5/1 20/0!!! so dumb. so.. so bad. But the upshot is that game theory is awesome and if you want to apply it to eve, look at hawks and doves (and soak up the tears)
Avatar
Actually the precise payoff matrix can be variable depending on the specific implementation of the dilemma, but hey, keep up the constructive contribution to the discussion.
Avatar
While you are correct, it is also true that this payoff matrix is not the Prisoner's Dilemma. In particular, the punishment for both players confessing must be greater than that for both players not confessing.The game presented is simply not the Prisoner's Dilemma and analysis that follows is fundamentally mistaken.

One of the most divisive aspects of EVE, something that tends so separate those who love the game from those who hate it. is the "metagame." Metagaming has lead EVE players to some dark places in the name of victory - Third party programs, spreadsheets, and worse. (Though there isn’t much worse than spreadsheets, in this writer’s opinion). Still, it's important to consider how to approach this most misunderstood of subjects. Today, I hope to give you a crash course in strategic thinking, specifically a concept known as “Game Theory," and show you how you can apply this to EVE.

Don’t Forget Your Script!

“It has been said that man is a rational animal. All my life I have been searching for evidence to support this” - Bernard Manning

The first concept I’m going to introduce you to is the “rational actor." For game theory to work, both you and the other “player” must be what is known in game theory as “rational actors." Now, if you were to hazard a guess, you might say that this means the other player is smart, or "gets" the nature of the game. This is not necessarily true: while a rational actor needs to understand the rules of the game and the outcomes they can achieve, this alone is not enough.

A rational actor must know the rules of the game, the outcomes they can achieve, AND they must also know their opponent’s knowledge of the game and the outcomes THEY can achieve. Here’s where it gets confusing, but it's a point worth making: Your opponent must know the game, the outcomes they can achieve, they must know the your knowledge of the game AND they must know that you know they know this.

As you can tell, I could go on forever like that. What it boils down to is that a rational actor is in possession of all the information they need to make an informed decision. Furthermore, rational actors only ever make decisions that result in the best outcome for them.

Without rational actors, there can be no game theory. The purpose of the exercise is to be able to predict actions and choices based on the outcomes an individual can receive. If you have no idea what your opponent knows, or your opponent has no idea what the rules of the game are, how can you ever hope to do anything other then rely on luck to second guess them?

Do Not Pass Go. Do Not Collect £200.

“A game is like a mirror that allows you to look at yourself.” - Robert Kiyosaki

You are in jail.

You have been accused of treason. Specifically, spying against your country in collaboration with your friend, Miss B. You can’t speak to Miss B and you have no contact with her; she is in a different cell. You sit there in the interrogation room all alone until a member of the government’s intelligence agency enters. He offers you the following deal:

- If neither you or your partner confess, they cannot prove you were stealing state secrets, and so will lock you up for breaking and entering. Sentence: 1 year in jail.

- If you snitch on your partner, and she says nothing, you can claim she forced you to go along with it. The government will be grateful and you will get off scott free, while Miss B gets 20 years in jail. However, if she snitches on you and you say nothing, YOU will get the 20 years and she will be released without penalty.

- If you BOTH confess, well then you’re both bang to rights and you’ll both be locked up for 5 years.

What would you do? Your choices look like this:

A) Confess: You will either get 5 years (oh no!) or 0 years (Yay!)

B) Don’t Confess: You will get 20 years (*&$#%!) or 1 year (meh).

The clear choice for a rational actor is that you confess. This is because if you confess and she confesses, you get 5 years - the second worst option on the list. However, if you don’t confess and she does, you get 20 years, which is the worst option. If you stay quiet the very best you can do is get 1 year in jail, whereas by confessing you get a potential of 0 years!

Now, the problem is of course, that Miss B is ALSO a rational actor, and therefore she will also always pick confess. This is called “The Prisoner’s Dilemma” and it worked very well for the KGB when they offered this option to Soviet political criminals. By knowing your opponent will pick the best outcome for themselves (confess) you are forced by default to pick confess to maximise your outcome (while they are thinking the same as you).

The prisoner’s dilemma is used as an introduction to game theory as it is simple, at a much more in depth level you can use complex computer models to formulate games with a vast array of choices to help you predict real life outcomes. 

Applying This To Your Spaceships

Unless you intend to become an expert in both computer programming and game theory, and then complete full, in-depth research into your rival EVE players, actual game theory models will be of no use to you in the game.

What is useful is the mindset of remembering to think about your opponent in order to second guess them. At this point you may be saying “Kitchner, this seems obvious! Of course I need to guess what my opponent knows!”... Alas! If only it were that simple! Part of the spying metagame in EVE is finding out things your opponents are doing. This can be taken to the next level though: Ask your spies to find out what their spies know about what you are doing to help you understand better what decision they are likely to make.

The beautiful thing about game theory is this can continue to spiral up level after level of meta-gaming as you constantly see to find more information about “what they know that we know”  while they find out “what we know that they know”. This is the key to thinking strategically.

Next time you are trying to second guess your opponent, be it on the field of battle or in the heart of a coalition war room, try to think about that exercise. What does my opponent think I will do? How does this affect my decision?

Hailed as both a "posting god" and a terrible poster I am a member of Executive Outcomes giving me access to two coalitions to bombard with good/bad posts. Out of game I have a Politics & International Relations degree and work in Risk Management.