THE PRISONERS DILEMMA AND THE MINI-MAX STRATEGY By Charles Warner The Pris ners Dile!!

a The Prisoners Dilemma is perhaps the best-known strategic game, as suggested by Dixit and Nalebuff,i and it illustrates how cooperation is often the best strategy. The Prisoners Dilemma also is useful for demonstrating how to use two ery useful decision tools, a decision tree and a payoff matrix, and how to employ a mini-max strategy. !uppose that in "ussia during the !talin era, a conductor of an orchestra was tra eling by train and was reading the score of the music he was to conduct at his next engagement. Two #$% policemen watched him reading and, thinking that the musical notations were some secret code, arrested him as a spy. The conductor protested that it was only Tchaiko skys &iolin 'oncerto, but with no success. (n the second day of his imprisonment, an interrogator walked up to the conductor and said confidently, )*ou had better tell us e erything you know. +e ha e caught your friend Tchaiko sky, and he is already talking., The #$% had, in fact, picked up a man whose only offense was that he was named Tchaiko sky, and they were sub-ecting him to the same kind of intense interrogation. .f the two innocents withstand this treatment and confess nothing, they will both get off with a relati ely mild three-year sentence /the standard punishment at that time for doing nothing0. (n the other hand, if the conductor makes a false confession and implicates Tchaiko sky while Tchaiko sky holds out, the conductor gets a reduced sentence of one year and Tchaiko sky gets the maximum sentence of 12 years for being recalcitrant. (f course, the tables will be turned if the conductor stands firm and Tchaiko sky gi es a false confession and implicates the conductor /12 years for the conductor, one year for Tchaiko sky0. .f both gi e false confessions and implicate the other person, then both get a reduced sentence of 34 years. .f neither one of them confesses nor implicates the other, they each get three years. These options are clearly laid out for the two prisoners, who, of course, are ne er allowed to talk to each other. The conductor reasons as follows5 6e knows Tchaiko sky is either /a0 confessing and implicating him or /b0 holding out. .f Tchaiko sky confesses and implicates him, the conductor gets 12 years by holding out, but only 34 years by confessing and implicating the other person, so it is to his ad antage to confess. .f Tchaiko sky is holding out, the conductor gets three years if he holds out and only one year if he confesses and implicates Tchaiko sky, so it is to his ad antage in this scenario to confess and implicate Tchaiko sky. Thus, confession is clearly the conductor7s best strategy.

Tchaiko sky is no dummy8 hes sitting in his cell doing the same mental calculations. 6e comes to the same conclusion. The result is, of course, that both men confess and implicate the other and are sent to !iberia for 34 years /the #$% ha e played this game many times and know they will get something on both men, regardless if it is true or not, and be able to fill their 9uota of prisoners0. +hen the two men meet in the $ulag :rchipelago, they compare stories and reali;e that they are both innocent and that they ha e been duped. .f they had both held out and said nothing, they each would ha e gotten only three years instead of the 34 they wound up with. 6owe er, the temptation to get sent away for only one year by confessing and implicating the other was so o erwhelmingly tempting at the time that they could not resist, and thus were in for 34 years. The decisions the prisoners had to contemplate is best illustrated by using two decision tools, a decision tree /<igure 3 below0 and a payoff matrix /<igure 10 below. The term )rat, or )not rat, is used to indicate whether or not one prisoner will implicate the other by making a false charge against him. The numbers in the decision tree and the payoff matrix are arranged with the conductors conse9uences of ratting or not ratting first, then %oriss conse9uences of ratting or not ratting.

"i#$re %
De(isi n Tree 34, 34

Ra' B ris Ra'

N ' ra'

3, 12

C n)$(' r N ' ra' B ris

Ra' 12, 3

N ' ra'

=, =

"i#$re &
Pay ** Ma'ri+ B ris Ra' N ' Ra'

Ra' C n)$(' r N ' Ra'

34, 34 >

3, 12

12, 3

=, =

/!addle point0

> 'onductor 34, %oris 34

The decision tree and the payoff matrix show the different conse9uences or outcome of each decision in years in the $ulag. The saddle point in the payoff matrix is that decision in which both sides get the same payoff ? e9uilibrium. The saddle point, or e9uilibrium, is not the maximum benefit for either side, but is the optimum benefit for both sides. The Mini-!a+ The re! )The theorem states that in ;ero-sum games in which the players interests are strictly opposed /ones gain is the other others loss0, one player should attempt to minimi;e his opponents maximum payoff while his opponent attempts to maximi;e his own minimum payoff. +hen they do so, the surprising conclusion is that the minimum of the maximum /mini-max0 payoffs e9uals the maximum of the minimum /maxi-min0 payoffs. Neither player can impro e his position, and so these strategies form an e9uilibrium of the game.,ii )The general proof of the mini-max theorem is 9uite complicated, but the result is useful and worth remembering. .f all you want to know is the gain of one player or the loss of the other when both play their best mixes, you need only compute the best mix for one of them and determine its result.,iii )To put it in plain language, the mini-max theorem says that there is always a rational solution to a precisely defined conflict between two people whose interests are completely opposite. .t is a rational solution in that both parties can con ince themsel es that they cannot expect to do any better, gi en the nature of the conflict.,i .n other words, )*ou know that the best you can expect is to a oid the worst., Ba(,#r $n) @ohn on Neumann concei ed of the basic precepts of game theory in 3A1B when analy;ing the game of poker. &on Neumann was a brilliant mathematician, so he put the precepts of game theory into a mathematical format, which he later expanded to apply to economics. %ut on Neumanns first theorem, the mini-max theorem as stated abo e was the basic foundation of game theory. &on Neumanns original insight was that when playing a ;erosum, two-person game, a person must make mo es based on the probable mo es of his competitor, assuming both players are rationale, and try to minimi;e the competitors maximum gain. Therefore, e ery mo e in a game must be based on an assessment of what a competitors possible mo es are. &on Neumanns original game theory, the mini-max theorem, works only under the following conditions5 3. There are only two players.

1. %oth players are rationale and desire to win and to a oid a loss. =. .t is a ;ero-sum game /one players win is -3 and the other players loss is -30 and, thus, no cooperation is possible. C. The players ha e only two choices or strategies. 2. There is perfect knowledge8 in other words, all the possible mo es are known to both players and there is no hidden information. D. The game is played once. Eater refinements of on Neumanns original mini-max theorem, which he pro ed mathematically, included the Prisoners Dilemma game /multiply games played0, Nashs F9uilibrium /more than two or many players0, and economic risk theories /multiple games, multiple players, and multiple strategies0. The Pie Ga!e Probably the best and simplest example of the mini-max theory is the Pie $ame. .n 3AA=, . was conducting a negotiating seminar for the .owa %roadcasters :ssociation /.%:0. :s . went into the seminar . passed a table that displayed se eral in iting .owa home-made pies. . bought one of the pies and took it to the seminar room, but before . entered the room, the executi e director of the .%: introduced me to his 32-year-old son and asked if the boy could attend the negotiating seminar. )(f course,, . said, wanting to be accommodating and reali;ing that the boys presence ga e me an excellent opportunity to play the Pie $ame. %efore . began the seminar, . put the pie . had purchased on a first-row table and asked the boy to come forward, which he did with some trepidation. . asked him if he liked pie. ). sure doG,, he replied enthusiastically. )He, too. . lo e pieG,, . gushed. )+ell, .ll tell you what .m going to do,, . said. )!ee this scrumptious pie hereI .m going let you choose how much of it you want. +ere going to play the Pie $ame., Then . said slowly, pausing between the two rules5 )There are only two rules in the Pie $ame, which are5 "ule J3 is that you can cut the pie any way you want. /pause0 "ule J1 is that . get to choose the first piece., +hen . told him the first rule, the boys face lit up8 he was almost drooling. +hen . told him the second rule, his face dropped and his brow became furrowed. 6e pondered for almost a minute. Then, with a big smile on his face, he cut the pie with extreme care exactly in half. . ga e a big sigh of relief because the boy had figured out the correct strategy, as game theory and the mini-max theory had predicted. 6e had minimi;ed the maximum si;e of my piece of pie /mini-max0 and he had maximi;ed the minimum si;e of the piece of pie that would be left after . chose /maxi-min0. %ecause the alues of the mini-max and the

maxi-min strategies must be the same /which is what mathematical calculations pro e0, he had cut the pie in half. . doubt if the boy knew about game theory or the mini-max strategy or could work out the mathematical proof, but he liked pie and was rational. !o as a reward for being rational and intelligent enough to figure out the correct strategy, . ga e him the entire pie, which delighted him, the audience, and me because . had made a dramatic point to begin my negotiating seminar. The Pris ners Dile!!a an) Mini-!a+ The point of the Prisoners Dilemma game is that if both had not been greedy and had not tried to maximi;e their own ad antage /oneyear sentence0 at the expense of the other, and had cooperated silently, they both would ha e been better off /lower right-hand box in the Payoff Hatrix0. .f both had pursued a mini-max strategy and accepted the second best conditions /three years by both holding out0, both would ha e been out se en years sooner. +hen making any strategy decision it is a good idea to use a decision tree and a payoff matrix8 because these decision tools can help you isuali;e and analy;e your alternati es. !ee )$ame Theory ? Programming, and )$ame Theory ? !ales, in the )Papers by 'harles +arner, link on www.charleswarner.us for examples of how to use decision trees and payoff matrices in business strategy situations. :lso, when making strategy decisions it is best to employ a mini-max strategy. !imply stated, )Dont get greedy.,

: andish Dixit and %arry Nalebuff. 3AA3. Thinking Strategically. New *ork5 +.+. Norton K 'o. : inish Dixit and %arry Nalebuff. 3AA3. Thinking Strategically. New *ork5 +.+. Norton. Pp. 3LB-3LA iii .bid. i +illiam Poundstone. 3AA1. The Prisoners Dilemma. New *ork5 Doubleday. p. D1. .bid.
ii