Friend or Foe? TV game show
If both players choose Friend, the money is split.
If one chooses Friend and the other Foe, Foe gets 100%.
If both choose Foe, each get 0.
Before you choose you try to convince the other player why they should pick you as friend. My question: Is there any reason (money related, nothing to do with being a "man of your word") to choose friend? You only get money if the other person selects Friend anyway, and you get double for picking Foe.
EDIT: Didn't know if this should go in gambling or math.
Assume that if Player A chooses, Player B will choose each option half the time. So if Player A chooses Friend, he will either get 50% of the pot, or nothing. The EV of choosing Friend is therefore 0.25. If Player A chooses Foe, he will win the whole pot half of the time, and nothing the other half of the time. The EV of choosing Foe is therefore 0.5. The same reasoning applies for Player B, so each has a "dominant strategy"---to choose Foe. However, the situation wherein if each player chooses the dominant strategy, each suffers a worse outcome than if they had chosen the "dominated" strategy, is the classic "Prisoner's Dilemma", and the reason that it's called a "dilemma" is that there IS no optimal strategy.
The only resolution to Prisoner's Dilemma games comes about through repeated iterations, but this game only is played once per pair of contestants.
Quote: mkl654321They should have called it "Prisoner's Dilemma", since that's what it essentially was.
It was exactly the prisoner's dilemma, but I would doubt if most people are familiar with the term.
Assuming the other guy has a 50/50 chance of picking friend or foe, the combined EV is 1 picking friend, and 0.5 picking foe. Of course, maximizing combined EV is not necessarily everyone's goal. I read an academic paper on the show once. As I recall, slightly more than 50% picked foe, and surprisingly women were more likely to pick foe than men.
Quote: Wizard...and surprisingly women were more likely to pick foe than men.
Not to sound like EvenBob, since he takes things a bit far on occasion, but I'm surprised that you find this surprising.
@mkl - respectfully, I believe you are in error when you say there is no optimal strategy. In the game show the weak equilibrium strategy is to betray. You are certainly wrong to say that prisoners dilemmas in general don't have a solution - betray is a pure Nash Equilibrium strategy in a classical prisoners dilemma because the cooperate+cooperate choice has a small cost associated with it.
Quote: sunrise089Not to sound like EvenBob, since he takes things a bit far on occasion, but I'm surprised that you find this surprising.
@mkl - respectfully, I believe you are in error when you say there is no optimal strategy. In the game show the weak equilibrium strategy is to betray. You are certainly wrong to say that prisoners dilemmas in general don't have a solution - betray is a pure Nash Equilibrium strategy in a classical prisoners dilemma because the cooperate+cooperate choice has a small cost associated with it.
Betray/Foe is only an "optimal strategy" in a vacuum. It has a higher EV than cooperate/Friend, but only for any one of the players--the proper term for ONE player is "dominant" strategy, in any case.
Betray has a much higher cost than cooperate if BOTH players choose one or the other. And since in the classic prisoner's dilemma, both players have equal standing, the "optimal strategy" would be to do what you think your opponent will do. In subsequent iterations, "tit-for-tat" has proved to be the dominant strategy.
The point is, in the FIRST iteration, there is no strategy that will work, because paradoxically, the dominant strategy will lead to the worst possible outcome.
In order to answer the question, "Which way would you vote?", you first need to choose between:
A) I would rather someone get some money than neither of us.
B) If I can't have all the money, then no one gets any!
These of course correlate to A being Friend, and B being Foe. I've seen some really upset people who were swayed by the arguments of their teammate and voted Friend, only to be backstabbed. I think your best bet is to decide BEFORE going on the show which way you would vote. Don't be swayed by puppy dog eyes or sob stories. They're lying anyway.
Friend or Foe (also in the UK, see the game show "Golden Balls" for a similar game situation), as stated before, has the problem that you only get one shot at this. I refer back to the A/B choice above. Instead of going for "optimal" strategy, you just need to go with "personal" strategy.
Looks like there are a handful of episodes and clips on Youtube. Just search for "Friend or Foe Game Show."
Agreed, I think you should go in knowing what you are going to pick. I watched that show quite a few times, and those who picked foe were usually the ones who seemed the most believable that they would pick friend. Of course everybody says they will choose friend. Those who really do tend to keep their statements shorts, while those who pick foe make an impassioned case why picking friend is the right thing to do, and then stab you in the back.
In the book Hollywood Blackjack author Dave Stann said he was on the show, and chose foe. He said something to the effect of, "I was a starving actor who need the money back then so screw my partner." I tried to shame him about it at the Blackjack Ball, but he did not seem very repentant.
Quote: MoscaThere is a difference in the game in that for the prisoner, the wrong choice could work out to a negative outcome (longer sentence), whereas in the game, you can do no worse than even: "started with nothing, can't leave with less than nothing, therefore might as well go for it all." I think that's why you see more "foe" choices. If the money was coming out of her bank account, I think you'd see more friend/friend outcomes.
I don't see it that way. Lots of times on game shows you see contestants making bad risks, because they came with nothing. For example, somebody posted a link of a player in the punch board game who won $5,000, and risked it for about a 1/100 chance at $10,000. The ironic thing is that he got it. My point being is you shouldn't play foolishly because you don't have the money in your hands yet. Same goes for casinos. You shouldn't bet more aggressively after winning because it is the "casino's money."
In the Friend or Foe case, let's say there was $1,000 at stake, I voted friend, and the other player voted foe. I would view it as $500 being my fair share, and the other player took it from me.
Quote: MoscaThere is a difference in the game in that for the prisoner, the wrong choice could work out to a negative outcome (longer sentence), whereas in the game, you can do no worse than even: "started with nothing, can't leave with less than nothing, therefore might as well go for it all." I think that's why you see more "foe" choices. If the money was coming out of her bank account, I think you'd see more friend/friend outcomes.
No, because at the decision point, each player has a 1/2 stake in the prize amount. Thus, they aren't "starting with nothing".
Quote: WizardI think your vote in that game says a lot about who you are as a human being.
I think that in turn, that is influenced by just how painful a person views the outcome, "He got everything and I got nothing". I've read a number of books on behavioral psychology that outlined how our choices are not rational in this regard. In a number of studies, Person A preferred an outcome where, say, Person B lost $300 and Person A got nothing, to the outcome where both got $50.
One of my favorite words is "Schadenfreude"--German for "pleasure at the misfortune of others". In Prisoner's Dilemma games, especially when the environment is adversarial, there's all too often a sense of, "I don't know what I'm getting, but I sure don't want that OTHER bastard to get anything at my expense." So such players pick the "dominant strategy"--they choose "Foe"--and the one who acts responsibly--"Friend"--gets the shaft. And if both pick "Foe"--"well, at least HE didn't get anything, either." That this is actually a worse outcome for the person thinking this way than if the players had cooperated is a secondary consideration in his mind.
The reason this doesn't work in the long run for society in general is that next time, the betrayed person remembers, and retaliates. Out in the world, the person who is cooperative gets screwed once--and that's it for any future possibility of cooperation. That's why we still have so many political divides--cooperation is such a fragile and not easily replaced thing. I think that the Prisoner's Dilemma is at the heart of many conservatives' refusal to pay taxes. They say, what if I pay, and the other guy doesn't? He'll get a free ride--and the possibility, however real or remote of someone--ANYONE--getting that free ride sends conservatives into a frenzied lather.
Quote: WizardI don't see it that way. Lots of times on game shows you see contestants making bad risks, because they came with nothing. For example, somebody posted a link of a player in the punch board game who won $5,000, and risked it for about a 1/100 chance at $10,000. The ironic thing is that he got it. My point being is you shouldn't play foolishly because you don't have the money in your hands yet. Same goes for casinos. You shouldn't bet more aggressively after winning because it is the "casino's money."
In the Friend or Foe case, let's say there was $1,000 at stake, I voted friend, and the other player voted foe. I would view it as $500 being my fair share, and the other player took it from me.
I understand, and mkl you make the same point. But lots of people don't think that way. They're going to be thinking, "At least I got on a game show while I was out here. That's something." And they pick foe. Thing is, in that instance, as a "friend" you gain nothing by picking foe, except the satisfaction of having guessed your partner/opponent right. Following that to its real world conclusion, this is not a game for those who are predisposed to choose "friend" in the first place, and will get a preponderance of "foe" choosers as applicants for contestant slots.
I'd be a "friend" picker, too. But in reality, the whole concept of the game irritates the hell out of me (not as a mind problem but as something to actually do), and I wouldn't participate, nor watch. Joining that artificial situation on purpose would just make me frustrated. My dad watches it, and tries to get me interested, but I'd rather watch 10 year old "Cops" reruns.
Quote: MoscaFollowing that to its real world conclusion, this is not a game for those who are predisposed to choose "friend" in the first place, and will get a preponderance of "foe" choosers as applicants for contestant slots.
It occurs to me that if the contestants were pre-selected for their Foe-ness, you'd never have to give away a single dime in prize money.
Sort of reminds me of one entry in Letterman's Top Ten List, for "Ways NBC could save money"--"10. Really brutal Final Jeopardy question."
I understand that in games that have substantial top prizes, like "Deal or No Deal", the producers of the show actually do sweat the money, and construct the show in subtle ways so as to minimize the amounts the contestants win.
Quote: mkl654321It occurs to me that if the contestants were pre-selected for their Foe-ness, you'd never have to give away a single dime in prize money.
Sort of reminds me of one entry in Letterman's Top Ten List, for "Ways NBC could save money"--"10. Really brutal Final Jeopardy question."
I understand that in games that have substantial top prizes, like "Deal or No Deal", the producers of the show actually do sweat the money, and construct the show in subtle ways so as to minimize the amounts the contestants win.
I would think the producers would want somewhere between 30 and 40 % friend; often enough to get some friend/friend and a reasonable amount of friend/foe, and a plurality of foe/foe. After all, it's a game show designed to sell advertising, not a paid contest of wills or a real world test of The Prisoner's Dilemma. (Although it uses the dilemma as its hook.)
Quote: MoscaI would think the producers would want somewhere between 30 and 40 % friend; often enough to get some friend/friend and a reasonable amount of friend/foe, and a plurality of foe/foe. After all, it's a game show designed to sell advertising, not a paid contest of wills or a real world test of The Prisoner's Dilemma. (Although it uses the dilemma as its hook.)
Even a 40% proportion of Friends would only produce a Friend-Friend outcome 16% of the time; Friend/Foe 48% of the time; and Foe/Foe 36% of the time. I wonder what they think the audience wants, if the money at risk isn't that much of a consideration. Which societies would respond most favorably to each kind of outcome? Do Americans want to see everyone get some, one person get it all at the other guy's expense, or nobody get anything, bwahahahahaaaaa? (Given the rather vehement and acidic responses I've gotten to questions I've posted about social reform and income distribution, I tend to think that the I win/you lose outcome is the one that appeals to the most Americans.)
Quote: mkl654321Even a 40% proportion of Friends would only produce a Friend-Friend outcome 16% of the time; Friend/Foe 48% of the time; and Foe/Foe 36% of the time. I wonder what they think the audience wants, if the money at risk isn't that much of a consideration. Which societies would respond most favorably to each kind of outcome? Do Americans want to see everyone get some, one person get it all at the other guy's expense, or nobody get anything, bwahahahahaaaaa? (Given the rather vehement and acidic responses I've gotten to questions I've posted about social reform and income distribution, I tend to think that the I win/you lose outcome is the one that appeals to the most Americans.)
I was wondering that, too, and actually wrote out some of my suppositions, then decided I didn't really have a clue as to what would make a good show, or who they were looking for as an audience: the folks who recognized TPD? The Jerry Springer crowd?
If I were the producer, I think I'd want just enough friend/friend to make it interesting, around 10%. That's where I got my guess, figuring between 30-40%. I'd want about 50% of the matchups to be no payout.
1. Choose Foe.
2. If the other guy chooses friend, then write him a check for half the money.
Under this strategy the other guy won't be rewarded for voting Foe. I'd rather see the money kept by the show than go to a Foe voter. However, this strategy breaks down if the other player adopts it as well. Then you could have two honest, yet vindictive, players get nothing.
Quote: WizardHere is the Wizard Friend or Foe strategy:
1. Choose Foe.
2. If the other guy chooses friend, then write him a check for half the money.
Under this strategy the other guy won't be rewarded for voting Foe. I'd rather see the money kept by the show than go to a Foe voter. However, this strategy breaks down if the other player adopts it as well. Then you could have two honest, yet vindictive, players get nothing.
Wouldn't the strategy have to include "3. Tell the other guy that you're doing this."? Then his only hope of winning anything would be to vote Friend; he wouldn't be likely to vote Foe just for the satisfaction of seeing you get nothing, if doing so cost him money.
The next question would be, "why should he believe you"?
Quote: mkl654321Wouldn't the strategy have to include "3. Tell the other guy that you're doing this."? Then his only hope of winning anything would be to vote Friend; he wouldn't be likely to vote Foe just for the satisfaction of seeing you get nothing, if doing so cost him money.
The next question would be, "why should he believe you"?
As you wrote, he might not believe you, and he might get satisfaction out of seeing you get nothing, even if it cost him. I would.
As an example, suppose you are on the show, you choose Friend, and the other guys chooses Foe. After paying the Foe voter, the host makes a surprise offer. He whips out an additional $10,100. He says either (1) you both get none of it, or (2) the Foe player gets $10,000 and you get $100. The choice is yours. I would choose both of us get none of it. It would be worth the $100 to me to punish the Foe player.
How about just signing a contract with the other guy that you are going to split the money, so long as at least one of you chooses Friend? If no one welshes on that deal, everyone comes out OK financially unless both vote Foe for some reason. There would be incentive to abide by the deal and none to violate it (unless you welsh). The result -- both would vote Friend and get the added benefit of being perceived as honorable people, even if the contract is dishonorable in the context of the show.Quote: mkl654321Wouldn't the strategy have to include "3. Tell the other guy that you're doing this."?Quote: WizardHere is the Wizard Friend or Foe strategy:
1. Choose Foe.
2. If the other guy chooses friend, then write him a check for half the money. ...
Quote: DocHow about just signing a contract with the other guy that you are going to split the money, so long as at least one of you chooses Friend?
I can't speak for Friend or Foe, but on Survivor they make it very clear that deals to split the winning prize are forbidden. Chances are they wouldn't like it on Friend or Foe either, as it would get boring if everyone voted friend because they signed contracts. Under my "Wizard" strategy I would give the other player the check discretely when nobody from the show was watching.
Regarding such deals, here is a story about a friend of mine was on Jeopardy. He told me that backstage before the show he made a deal with one of his opponents in which if the two of them were in first and second place, and the first place player had more than twice the money as the second place player, that in Final Jeopardy the first place player would wager his score less double the second place score. For those who don't know, in the event of a tie, both players advance to the next show. So, under this strategy, the first place player has nothing to lose, and the second player has a chance to advance. I saw a tape of the show, and that is exactly the situation that occurred, with my friend in second, and the other guy in first. However, the first place player reneged on the deal, and wagered enough to stay in first by at least $1. The first place player turned out to be wrong in Final Jeopardy, my friend right, so his betrayal caused my friend not to advance. When he was confronted after the show, he made some excuse about such deals not being legally binding.
Yes, that's why I said the people agreeing to the (secret) contract could appear to be honorable, even though the contract would not be, at least in the context of the show.Quote: Wizard... they make it very clear that deals to split the winning prize are forbidden. ...
Quote: WizardAs you wrote, he might not believe you, and he might get satisfaction out of seeing you get nothing, even if it cost him. I would.
As an example, suppose you are on the show, you choose Friend, and the other guys chooses Foe. After paying the Foe voter, the host makes a surprise offer. He whips out an additional $10,100. He says either (1) you both get none of it, or (2) the Foe player gets $10,000 and you get $100. The choice is yours. I would choose both of us get none of it. It would be worth the $100 to me to punish the Foe player.
Which would be horribly irrational. Someone walks up to you holding out a $100 bill. You smile and reach for it, but the guy says, "Before I give this to you, you have to agree that I give another randomly chosen person $10,000. If you refuse, you don't get the $100, and neither does the other person get the $10,000." Who on earth wouldn't say, "Whatever, dude. Gimme the $100"? And is there any real difference if the prospective recipient of the money is standing next to you on a sound stage?
Of course, I understand what you're getting at, but someone I dislike getting a large sum of money wouldn't be sufficient negative consequence to get me to forgo $100. But I try to maximize utility. The outcome of me getting $100, and the other guy getting a million dollars, is still superior to the outcome of us both getting nothing.
For one thing, I can meet the other guy backstage afterwards, and mug him.
Quote: WizardI can't speak for Friend or Foe, but on Survivor they make it very clear that deals to split the winning prize are forbidden. Chances are they wouldn't like it on Friend or Foe either, as it would get boring if everyone voted friend because they signed contracts. Under my "Wizard" strategy I would give the other player the check discretely when nobody from the show was watching.
Regarding such deals, here is a story about a friend of mine was on Jeopardy. He told me that backstage before the show he made a deal with one of his opponents in which if the two of them were in first and second place, and the first place player had more than twice the money as the second place player, that in Final Jeopardy the first place player would wager his score less double the second place score. For those who don't know, in the event of a tie, both players advance to the next show. So, under this strategy, the first place player has nothing to lose, and the second player has a chance to advance. I saw a tape of the show, and that is exactly the situation that occurred, with my friend in second, and the other guy in first. However, the first place player reneged on the deal, and wagered enough to stay in first by at least $1. The first place player turned out to be wrong in Final Jeopardy, my friend right, so his betrayal caused my friend not to advance. When he was confronted after the show, he made some excuse about such deals not being legally binding.
But all that the leading player's reneging did was restore a situation that would have existed without the deal: he could ensure a win simply by making a small enough wager. So your friend's failure to advance, strictly speaking, wasn't caused by the other player's reneging on the deal; it was caused by the fact that he had less than half the money the leader had when the Final Jeopardy round was reached.
As the leader, I would actually HOPE for the game to finish in a tie, and in fact, if the amount was small enough, I would wager the appropriate amount and then deliberately get the Final Jeopardy question wrong. Why? Well, I've just beaten that second place guy rather soundly--I have more than twice as much money as he does. So wouldn't I like to play against him again, NEXT time? For example, I have $15,300, the second place guy has $7,500. I would only wager $299 or less under the obvious "guarantee a win" strategy, so perhaps I would do better to bet $300, hope for a tie, and ensure that I play against that guy who I know I can beat, rather than some unknown quantity.
Quote: mkl654321Which would be horribly irrational. Someone walks up to you holding out a $100 bill. You smile and reach for it, but the guy says, "Before I give this to you, you have to agree that I give another randomly chosen person $10,000. If you refuse, you don't get the $100, and neither does the other person get the $10,000." Who on earth wouldn't say, "Whatever, dude. Gimme the $100"? And is there any real difference if the prospective recipient of the money is standing next to you on a sound stage?
That is not the situation. I'd be happy to take the $100 if the person getting the $10,000 were a random person. However, the player who chose Foe not only screwed me, but also probably lied to me, because everyone promises to vote Friend in the discussion before the choice. You were the one who brought up schadenfreude, which is exactly the reason behind my decision.
Quote: mkl654321As the leader, I would actually HOPE for the game to finish in a tie, and in fact, if the amount was small enough, I would wager the appropriate amount and then deliberately get the Final Jeopardy question wrong. Why? Well, I've just beaten that second place guy rather soundly--I have more than twice as much money as he does. So wouldn't I like to play against him again, NEXT time?
Very good point. Something I have thought of. I'm not saying you're wrong, but you should also consider that the second player place at least beat the third place player. This might make for a good math problem, but I'd bet you're right. In my hundreds of times watching Jeopardy I've never seen this happen.
Quote: WizardThat is not the situation. I'd be happy to take the $100 if the person getting the $10,000 were a random person. However, the player who chose Foe not only screwed me, but also probably lied to me, because everyone promises to vote Friend in the discussion before the choice. You were the one who brought up schadenfreude, which is exactly the reason behind my decision.
Sure, but if that joy ("freude") costs me $100, that's a different story altogether, than the abstract concept of enjoying someone else's misfortune. To flip it around, would I pay $100 to see someone else lose $10,000? I'd better really HATE that person if I would, and even then, I'd be cutting off my nose to spite my face, to use an expression that I hate.
Again, I default to the concept of maximum utility. Let's put this concept a little differently. There is another person--whomever--who you consider your worst enemy; he's stolen your money, had sex with all three of your wives, borrowed your hedge trimmer without returning it, etc. One moonless night, Mephistopheles appears to you and tells you that he will kill your enemy instantly, with no obligation, other than that you must agree that you will die one year sooner than you otherwise would have (you aren't selling your soul, just part of your life). Would you take the deal? How much would you have to hate that guy to take the deal?
Quote: mkl654321Sure, but if that joy ("freude") costs me $100, that's a different story altogether, than the abstract concept of enjoying someone else's misfortune. To flip it around, would I pay $100 to see someone else lose $10,000? I'd better really HATE that person if I would, and even then, I'd be cutting off my nose to spite my face, to use an expression that I hate.
I think that one reason society works is that good people band together to punish those that don't contribute to the common good of society. Hopefully we can at least agree that this comes down to weighing the utility of the schadenfreude vs. the utility of the money. For example, if the amounts were $10,000 to the other guy and $5,000 to me, I'd take the money. I'd like to think if the amounts were $10,000 and $1, you'd sacrifice the $1.
Quote: mkl654321Again, I default to the concept of maximum utility. Let's put this concept a little differently. There is another person--whomever--who you consider your worst enemy; he's stolen your money, had sex with all three of your wives, borrowed your hedge trimmer without returning it, etc. One moonless night, Mephistopheles appears to you and tells you that he will kill your enemy instantly, with no obligation, other than that you must agree that you will die one year sooner than you otherwise would have (you aren't selling your soul, just part of your life). Would you take the deal? How much would you have to hate that guy to take the deal?
I don't want to promote the vigilantism implied in your question. I'll just leave that alone. It is easier to wrap my mind around these hypothetical situations if constructed in the form of a game show.
Things that don't have monetary value take on value in people's minds (like screwing someone who screwed you. Yeah!). People don't make economically rational decisions, just decisions that are satisfactory (less than optimal, but close). Most people don't have a perfect grasp of the probabilities in front of them, so they make 'mistakes', even though they might believe they are acting rationally.
I'd take the $100 if it was a random person, but I'd pass if I have a negative history with the guy. I'm not even sure the guy would learn a lesson, which would certainly make the decision irrational. But that's what my gut says. I've become more interested in trying to figure out WHY we make these odd decisions, as opposed to trying to find the right one.
I realize it is easy to Monday morning quarterback this analysis, but here is what I think gives away both players.
Gareth: First, the stealer, if there is one only, is usually the first to talk. Often it seems they have a prepared statement. Stealers often beg the other to pick "share." You can see he said "Please split." (1:33). Another clue is that he bows his head at one point (1:42) as if in shame. Now, some stealers are more confident than this guy, but I think he is not a great liar, and trying to keep a poker face.
Selena: She is just a doll. Notice how she has a genuine smile throughout (until she got the knife in her back). You can tell a real smile in that is uses the whole face, and not just the mouth. It takes a good actor to fake a true smile. Sharers tend to look the opponent straight in the eye, altough a good stealer can do this too, but overdo it. It is harder to explain what makes a sharer, but I think something about a sharer is just more likeable. She is just the sweetest thing.
So, shame on that guy. I hope every person who he might need to build trust with, like a girlfriend or business partner, watches the show.
Seriously, this is a great show.
Unfortunately, this skill is completely useless, even to a competitor in the game. Knowing what they are going to do doesn't help you at all.
When people explain themselves after the show, and say something like "I chose steal because I didn't know what she was going to do, and I wanted to play it safe", I wonder if they are really that logically challenged, or just trying to justify themselves so they don't look as bad on TV.
My favorite one, by far, is this: http://www.youtube.com/watch?v=S0qjK3TWZE8 I had heard about this before, from a friend of mine who is really into game shows, but never seen the video.
"You're walking away with my money because you're an idiot". Best line ever.
Quote: AxiomOfChoiceUnfortunately, this skill is completely useless, even to a competitor in the game. Knowing what they are going to do doesn't help you at all.
True. The only benefit is that if you are a sharer then at least you can have the satisfaction of revenge if you know the other person will pick steal.
Here is another episode that fits my behavior model.
The stealer tries to convince the other player to share. Meanwhile, the sharer explains why she is going to split, while leaving the other player to her own decision. As usual, the stealer is the first to talk. The stealer also has a hard time keeping eye contact, looking down often, while the sharer looks the other straight in the eye.
George Burns
Quote: WizardTrue. The only benefit is that if you are a sharer then at least you can have the satisfaction of revenge if you know the other person will pick steal.
Excuse me, I'd like to return this jacket
May I ask why?
For spite...
I heard of an experiment that was done (unfortunately I don't have a link to anything, so it's all 2nd-hand and from memory).
Basically, 2 strangers were approached. Some amount of money (say, $1000) was offered to them to share. The catch was, one of the players (player A) was told that he was allowed to offer to split the money between the two however he wanted (player B has to be quiet while player A is deciding -- it's not a negotiation). Player B would then be allowed to accept or reject. If Player B accepted, both would get the assigned amounts. If Player B rejected, both would get nothing.
So, if you're player A, how much do you offer player B in order to try to get them to accept? If you're player B, what's the minimum that you'll accept?
Apparently, the "70-30" split came up a lot. I have to say, I have to agree. For non-life-changing amounts of money, I probably wouldn't accept less than 30% if I was player B. Player A is a harder question. You could just offer a 50-50 split. You could try to take a little more, but still under the 70-30 level that (apparently) many people would accept. Or, you could try to read the other person, and figure out their minimum and extract maximum value from the situation.
A friend of mine suggested that for large amounts of money, if he was player A, he would offer slightly more than 50% to the other person to ensure that they would accept. Eg, if it was $100k, he would offer $51k to the other person, figuring that no one would turn that down (again, it's not a negotiation, so there are no counter offers... the other guy either accepts, or no one gets anything)
Quote: AxiomOfChoiceMay I ask why? For spite...
Exactly. I think one of the reasons we evolved is that humans are social creatures. However, to ensure that everyone in the pack acts altruistically, when one member is selfish he is ostracized by the rest. This behavior can also be observed in primates.
Quote:I heard of an experiment that was done (unfortunately I don't have a link to anything, so it's all 2nd-hand and from memory).
We discussed that experiment at length here a long while back.
Quote: WizardWe discussed that experiment at length here a long while back.
Yes we did in Mach of last year.
https://wizardofvegas.com/forum/questions-and-answers/casual-corner/4720-ultimatum-game/
Quote: Wizardif you are a sharer then at least you can have the satisfaction of revenge if you know the other person will pick steal.
Oh, I get it, losing a round does not mean you are eliminated? That's what I was thinking, that you got eliminated.
Quote: odiousgambitApparently I am insufferably stupid, as my questions are invariably ignored.
I'm sorry odious, but I am not familiar enough witht he whole game to answer your question. I've only seen the final choices a few times. I don't know how the "killer" ball functions, but am curious to find out.
Quote: odiousgambitApparently I am insufferably stupid, as my questions are invariably ignored.
Speaking only for myself, I wasn't sure if you were referring to Friend or Foe, Golden Balls, or the WoV challenge. However, I didn't want to look stupid by asking.
Somewhere on YouTube there is an entire espide of Golden Balls, broken into parts. The lead up to the ending is also interesting, and is why I prefer it to Friend or Foe.
Quote: WizardSomewhere on YouTube there is an entire espide of Golden Balls, broken into parts. The lead up to the ending is also interesting, and is why I prefer it to Friend or Foe.
OK, if we play again I'll try to have watched it. Sorry to be petulant, it was the sting of defeat [g]
Quote: dwheatleyI used to be a pretty firm believer in maximum utility, but that belief has waned over the past few years. Much easier to explain people's behaviour with concepts like subjective rationality or behavioural economics.
(...)
Most people don't have a perfect grasp of the probabilities in front of them, so they make 'mistakes', even though they might believe they are acting rationally.
(...)
But that's what my gut says. I've become more interested in trying to figure out WHY we make these odd decisions, as opposed to trying to find the right one.
I'm surprised nobody mentioned the evolutionary aspect.
Repeated Prisoner's Dilemma shows it is "optimal" to develop a Tit-for-tat strategy. But it is certainly not easy to see it as any more individually rational than other possibilities. What happens is, a population of players will reach a better outcome. They cooperate, it does not appear individually rational, yet they end up better off than the conservative-types. (Look at the social-political systems: I'm glad I'm in Europe.)
So, many species (animal or vegetal) develop such "behaviours" of cooperation. I'd rather you not ascribe it to some conscious rationality; it is instinct, it is genetically imprinted, it is the result of evolutionary selection.
Well, when we humans act in a seemingly "irrational" fashion, we are more than probably exhibiting instinctive reactions, developed by evolution. And one of these is the retaliation instinct. Most of us would admit to a little Schadenfreude, but essentially when Schade is hurting someone who did us wrong. And, in the Wizard's example, we all somehow get the "not rational but yet I understand" feeling. That is because our brain fights between reason and instinct, which it BOTH finds acceptable.
This explains why: 1° we react differently if it is for a random person (no retaliation); 2° we (our brain) are trying to rationalize our instinctive reactions (reduce the cognitive dissonance)
3° the same retaliation attitude appears in the Sharing Game. A sum is to be shared among two persons. The first one gets to choose the amount each receives. The second person then accepts the deal, or refuses -- and then nobody gets anything. It is rational for the first person to propose a ridiculously small amount to the other, because it is rational for the second to accept anything higher than zero. But in the real world, most second players reject scornfully offers where they receive less than 40% of the sum. And in experiments, most first players propose close to 50%. And nobody thinks it strange or irrational: we all have the same instinct of punishing the exploiter.
Quote: kubikulannwe all have the same instinct of punishing the exploiter.
I enjoyed your comments, and agree with them. The second round of my Golden Balls game was designed to keep people honest. Steal in an early round and it could cost you in a later one, especially with a big multiplier. I hope you'll sign up. You have until 8:00 PM today, Pacific time.
