See also: http://en.wikipedia.org/wiki/Price_of_anarchy. A way of analyzing the difference between the "optimal" outcome and the "rational" one. Common in the computer science literature, slowly infiltrating the economics literature I think.

I think economics got there first; that looks like http://en.wikipedia.org/wiki/The_Nature_of_the_Firm : companies form, even in an efficient market, because they can potentially globally optimize groups of individuals into more efficient interactions than the individuals themselves could in any market.

The other side of it is that such optimization doesn't scale and a sufficiently large firm will be unable to find practical solution at all.

Cool, seems interesting and related, but also very different and more specific. The point of price of anarchy is to take any game and quantify the "loss" due to selfishness or lack of cooperation.

1) Someone posts something about game theory

2) Someone comments that the scenario is unrealistic or contrived, then makes clever points bringing in nuances of real world behavior/situations.

3) [Optional] Original poster responds explaining why the situation reductively captures some underlying mechanism or why the commenter is mistaken

4) Somebody posts how this means communism is the solution

Actually 4) should probably be 1).

I see your communism and raise you self-driving cars.

it's interesting that if you replace the cars with electrons and change the graph to its electronic analogs, the electrons would do the super-rational thing and go along separate paths.

The same would happen with cars if the drivers didn't care of getting to their destination faster, just enjoying the ride for the ride's sake.

> Asking these questions is admitting to the wishful thinking that competitive optimization should lead to an optimum. Such is not the case, competitive optimization leads to an equilibrium and not to an optimum.

Well, that explains a lot about capitalism and social inequality.

> "Altruism, evolutionary driven cooperative behaviors, and mystical concepts like super-rationality are put forward as missing features in game theory."

Is Superrationality considered "mystical"?

Very interesting read. I wonder if there are any concreate examples of this situation happening in the real world?

In Seattle there are many people who commute across Lake Washington to get to work, where you have the 520 bridge, the I-90 bridge, or you can just drive around the lake. After reading this article I wonder if it would be effective to divide everyone who makes this commute into two groups. On even numbered days group A can use the bridge while group B must drive around the lake. On odd numbered days they would switch. Could such a scenario effectively reduce the overall time spent commuting for all drivers?

Highway 101 and 280 on the Peninsula in the Bay Area? :-)

It was a great read, I haven't seen a lot of work that talks about the extended Prisoner's Dilemma situations.

Most 'free markets'.

The very same thing happens around the world in many political elections; because people think that voting for a candidate perceived as unpopular is wasting their vote; so some candidates remain unpopular not for their ideas but for their perceived unpopularity.

People when are free to make decisions individually usually assume everyone else doesn't have group mentality so they themselves decide not to have one, making it a self-fulfilling prophecy.