Game Theory?

Ok i’ve recently just been introduced to this. I was wondering if any one could help me understand better and give me a few examples.

It’s a fun and important area. Got anything more specific you want to know?

Read about the prisoner’s dilemma, or newcomb’s problem. Oscar Mortgenstern (sp) is the man. Read this book: … c_Behavior

Ditto. I’ve read quite a bit about it over the last few years.

Look up with respect to game theory:

hawks and doves.
prisoner’s dilema
iterated/one off games
zero sum / non zero sum
tit for tat/tit for two tats

I’ll be glad to discuss anything.

Ok from the little research i did, and what i’ve learned from my professor, game theory is flawed. Sure there may be instance when i works, but the over all concept is flawed.

I think that statement mostly speaks about your research and your professor.

What else would i go on, the concept is flawed in the reason that it works most effectively when one is alone. Just like in the Prisoner’s Dilehmma, cooperation is key.

Could you elaborate on why you and your prof think it’s flawed…?

Cooperation is the key to success.

That’s not a flaw - it’s an empirical observation

It’s the key to not losing comepletely. Competition yields the best potential reward. Maybe your professor is a hippy?

Hi zero,

I started reading “the Compleat Strategyst” by J. D. Williams, but like most books I have not completed it.

An obvious problem is in assigning values to the payoff matrix. In addition the game needs to be properly discribed.

Examples of failed analysis are the war in Iraq and Von Neuman’s first strike initiative against the Soviet Union. (The latter assumes that 10’s of millions of needlessly destroyed human lives would have been a mistake).

However, I had tried to apply it to a simple business situation, and the payoff matrix was virtually impossible to assess.

Hey Ed3, long time no see. :wink:

Anyway, I think it is distracting to concentrate on the reward/penalty grid aspects of game theory - they are unrealistic because there are many gradients of co-operation/betrayal, not just a black/white divide. A true grid would be near infinite in most ‘real life’ situations, with the many extra factors, chains of relationship and variables involved - the either/or nature of the prisoner situation is too abstract from normative society.

However, game theory remains relevant and useful in discovering under what conditions people can cooperate effectively, which strategies ‘win’ not in terms of pay-off, the pay-offs can be arbitrary, but in competition with other strategies.

The pay-off is unimportant - as long as it is better than the pay-off for a mutual knife in the back, or overall, for a one-sided betrayal.

Game theory predicts, quite rightly, that in order for cooperation to prosper in a society, the people involved must play itterated games - ie. not just a one off, but a radial web of favours and debts among a group of close knit people. And that people must be willing to punish cheaters, even at the temporary expense of their own fortunes.

A tit for two tats is reflected in the bible - the ‘eye for an eye’ of the old testament being replaced by the turned cheek of the new.

Game theory also maps out the arisal (and decline) of societies from small groups to population sizes beyond those of today. And impllies what strategies and checks must be in place to foster and support such at each step.

Stepping back from the nitty-gritty, and seeing the whole picture, is game theory’s forté.

IMO anyway.

u just a cut and dry kinda guy, aintcha smears?

whose head is like a block

jk buddy <3

Exactly. To expect it to be able to predict business outcomes or wars is to commit a categorical error - it’s not a tool for prediction, it’s a descriptive behavioral model which supports a crucial normative lesson about cooperation and success among individuals. Obviously that lesson has important implications for all sorts of human transactions, including business and international conflict, but to expect, say, the prisoner’s dillemma to mirror any given sequence of real-world exchanges is probably like expecting the pythagorean theorem to help you pitch a perfect tent or something.

I think this is why I tend to react to game theory with a big fat “So what?”

well, there’s still an empirically demonstrable normative lesson involved, Faust - not necessarily something that’s wise to ignore

I don’t see a flaw here; this is just a component of gain.
Take a look around and find where cooperation is not part of success in gain?

That is to say, find instances where resistance and isolation is part of success in gain.

Then think of the application…gain.

The application, as said above, is the key for the perspective.
GT is a tool.

You might as well have just said that scopes on rifles are flawed because the overall concept of shooting is flawed.
The scope isn’t flawed at all, as it does it’s job very well.

However, you don’t find value in shooting.
Likewise, you may not find value in focusing on gain in the perspective that GT uses.

However, thousands upon thousands of business models world wide do care very much about the perspective of gain and the various applications of GT in those models.

And now you understand why I don’t post in the straight philo section.

In fact I don’t. I have the same reaction to a lot of Famous Bigtime Philosophy.