[General Battuta]

I'd never go above $16. If someone is stupid enough to say $17, someone is probably going to be stupid enough to say $18. I seriously doubt anyone I hang around with will offer any amount >= $19. But who knows...

However, since I'm so distrustful of easy money, I'd probably just avoid bidding altogether if I'm given the option.

Well, remember that you've just been tossed in to the scenario and don't have time to think about it. In probability a large chunk of 'you and the people you hang around with' would get sucked in.

[mxlm]

Please explain this further. If it means what I think it means, both players would gain most monetary profit by basically moving their hands from one side to another as fast as possible...

I would think the optimal strategy would be for me to smack my hand on the desk repeatedly and rapidly, and then split the money with the other guy. Raising/lowering a few milli/centimeters must be faster than a 180 degree rotation, yes?

[Timerlane]

You are going to play a game for ten seconds. You will be partnered with someone else.

From this moment on, you cannot speak or communicate, and you must close your eyes and keep them shut. Please assume the arm wrestling position!

Now, you're going to arm wrestle, and every time your opponent's hand touches the desk, you will receive $1. You have 10 seconds. Go.

I imagine you can tell what the optimal strategy in this game is, and I also imagine you can guess what pretty much everybody actually does.

Please explain this further. If it means what I think it means, both players would gain most monetary profit by basically moving their hands from one side to another as fast as possible...

My smart-aleck rules-lawyer sense wants to point out that, technically, both peoples' hands touch the desk either way(the back of the 'losing' hand, back fingertips of the 'winning' hand). Therefore, a dollar should be given to both people when either side 'wins'.

Therefore, the ideal strategy is for the weaker individual to let his arm go almost limp, and the stronger individual to raise and lower his opponent's hand as fast as possible. Of course, it depends on what specific rules are required to define "arm wrestle". Must both arms return to a roughly 90^o vertical position before another 'win' can be counted, for example? Back and forth would then be ideal, to ensure that dollars will always be given for each 'win'(especially since both players have their eyes closed, and may not be able to verify what angle their arms are positioned at, at a given moment, and in trying to perform a repetitive action in quick succession, it's likely people will fail to raise their arms all the way up between 'wins').

(Yes, I understand and see the scenario as it was 'meant' to be; I just enjoy being difficultsilly.)

[Bob-san]

If it was 2 people going at it, they'd probably not get very far. Using that logic, I'd say the scenario is a bunch of people (5+) that are all bidding. So, as the bids start to take off I'd probably stop at $15 or maybe $18 and call it a day. Someone else will take my position as first and then second place, as there will still be value left in the bids. Then, it's all up to the rest. The problem is the winner will pay $19 less than the loser, and nobody wants to the loser and pay more.

[Turambar]

I'd never go above $19

But if someone bids $20, you've lost $19 (since you're in second place).  If you then one up them to $21, you only lose $1.

yeah, then i stop, and i've lost nothing, and that moron just paid $20 for a $20.

if i paid $19 for the $20 i could take my $1 earnings and go buy a small bag of chips or a mcDouble.

[General Battuta]

I'd never go above $19

But if someone bids $20, you've lost $19 (since you're in second place).  If you then one up them to $21, you only lose $1.

yeah, then i stop, and i've lost nothing, and that moron just paid $20 for a $20.

if i paid $19 for the $20 i could take my $1 earnings and go buy a small bag of chips or a mcDouble.

No, if you stop, you pay $19 and get nothing. Re-read the rules.

If it was 2 people going at it, they'd probably not get very far. Using that logic, I'd say the scenario is a bunch of people (5+) that are all bidding. So, as the bids start to take off I'd probably stop at $15 or maybe $18 and call it a day. Someone else will take my position as first and then second place, as there will still be value left in the bids. Then, it's all up to the rest. The problem is the winner will pay $19 less than the loser, and nobody wants to the loser and pay more.

Most people set a limit at about $15 and then proceed to break it.

You are going to play a game for ten seconds. You will be partnered with someone else.

From this moment on, you cannot speak or communicate, and you must close your eyes and keep them shut. Please assume the arm wrestling position!

Now, you're going to arm wrestle, and every time your opponent's hand touches the desk, you will receive $1. You have 10 seconds. Go.

I imagine you can tell what the optimal strategy in this game is, and I also imagine you can guess what pretty much everybody actually does.

Please explain this further. If it means what I think it means, both players would gain most monetary profit by basically moving their hands from one side to another as fast as possible...

My smart-aleck rules-lawyer sense wants to point out that, technically, both peoples' hands touch the desk either way(the back of the 'losing' hand, back fingertips of the 'winning' hand). Therefore, a dollar should be given to both people when either side 'wins'.

Therefore, the ideal strategy is for the weaker individual to let his arm go almost limp, and the stronger individual to raise and lower his opponent's hand as fast as possible. Of course, it depends on what specific rules are required to define "arm wrestle". Must both arms return to a roughly 90^o vertical position before another 'win' can be counted, for example? Back and forth would then be ideal, to ensure that dollars will always be given for each 'win'(especially since both players have their eyes closed, and may not be able to verify what angle their arms are positioned at, at a given moment, and in trying to perform a repetitive action in quick succession, it's likely people will fail to raise their arms all the way up between 'wins').

(Yes, I understand and see the scenario as it was 'meant' to be; I just enjoy being difficultsilly.)

No, you're not being silly. This is actually the correct strategy.

[el_magnifico]

I'd never go above $16. If someone is stupid enough to say $17, someone is probably going to be stupid enough to say $18. I seriously doubt anyone I hang around with will offer any amount >= $19. But who knows...

However, since I'm so distrustful of easy money, I'd probably just avoid bidding altogether if I'm given the option.

Well, remember that you've just been tossed in to the scenario and don't have time to think about it. In probability a large chunk of 'you and the people you hang around with' would get sucked in.

I'm not given enough time to think about it? In that case, I'm definitely NOT going to bid. At all. Period. There are moments in life when you don't have time to think thoroughly and you just have to risk it. This is not one of those moments. If I'm not given enough time to think, then something definitely smells fishy.

Cuando la limosna es grande, hasta el santo desconfía.

EDIT: I've just found an approximate translation:

When alms are generous even the saint is distrustful.

[CP5670]

Interesting topic. I was just grading the final exams for a game theory class earlier today.

As Herra said, a coalition that involves everyone would be the optimal strategy here, but your rules don't allow that, so in that case the best thing is not to bid at all.

[Timerlane]

My smart-aleck rules-lawyer sense wants to point out that, technically, both peoples' hands touch the desk either way(the back of the 'losing' hand, back fingertips of the 'winning' hand). Therefore, a dollar should be given to both people when either side 'wins'.

Therefore, the ideal strategy is for the weaker individual to let his arm go almost limp, and the stronger individual to raise and lower his opponent's hand as fast as possible. Of course, it depends on what specific rules are required to define "arm wrestle". Must both arms return to a roughly 90^o vertical position before another 'win' can be counted, for example? Back and forth would then be ideal, to ensure that dollars will always be given for each 'win'(especially since both players have their eyes closed, and may not be able to verify what angle their arms are positioned at, at a given moment, and in trying to perform a repetitive action in quick succession, it's likely people will fail to raise their arms all the way up between 'wins').

(Yes, I understand and see the scenario as it was 'meant' to be; I just enjoy being difficultsilly.)

No, you're not being silly. This is actually the correct strategy.

Well, my "two dollars for each tap", and "define the exact parameters of the game" was the obnoxious rules-lawyering(IMO), but I meant that I get the idea that people with "must win!" on their minds, and no time to prepare, will, by default, pull as hard as they can, and if relatively evenly matched, end up with nothing(or at least far less than they 'theoretically' could have).

The only way to 'win'(or rather, "not lose horribly and/or look like a moron in retrospect"), is not to play. Always question(so to speak) the process, and if you aren't allowed to, for whatever reason, walk away, especially if it's not a matter of life and death.

[Qent]

Another thing about the $20 game is that it really makes a difference whether or not the players think they have to pay. Did the executives actually pay $2000 and $1999? Did they think they were going to?

Maybe I would bid up to $25 or $30 and then hold there as the second bidder to deny someone just a little pleasure in posting it on the internet.

My smart-aleck rules-lawyer sense wants to point out that...

Who says they have to stay in arm-wrestling position? Will either hand trigger a win, or just the right ones?

[karajorma]

No, you're not being silly. This is actually the correct strategy.

Unless you have no knowledge of whether the other person will split with you. In which case the smart thing to do is to immediately go limp and then try to get the other guy to do the same. If you bounce back and forth like a metronome you still win without risking the other guy stabbing you in the back.

It's well worth remembering that there are a large number of tards out there who would simply believe that they won somehow and never split the money.

[Bob-san]

And the thing being missed here is that the actual irrationality is likely only in a handful of the bidders; it's a GROUP of people that had the auction going higher and higher, but it only takes 2 people to bid and the person in 3rd place when it passes $20 will probably not bid again (knowing the rules say only the first and second place bidders have to pay). The actions of 1 or 2 in a group does not show the rationality or irrationality of the entire group, just those selection irrational individuals.

[General Battuta]

Another thing about the $20 game is that it really makes a difference whether or not the players think they have to pay. Did the executives actually pay $2000 and $1999? Did they think they were going to?

They do, in fact, have to pay, even in the student version.

And the thing being missed here is that the actual irrationality is likely only in a handful of the bidders; it's a GROUP of people that had the auction going higher and higher, but it only takes 2 people to bid and the person in 3rd place when it passes $20 will probably not bid again (knowing the rules say only the first and second place bidders have to pay). The actions of 1 or 2 in a group does not show the rationality or irrationality of the entire group, just those selection irrational individuals.

That's not being missed at all. Since in any group almost all bidding members are liable to end up in this position, it holds for most members of the group.

Remember, statistical tests are used to determine significance.

The heuristics that cause this kind of irrational behavior are present in anyone and are actually exacerbated by groups. They're a fundamental feature of human cognition that can be overriden by effortful processing but not removed.

[Dilmah G]

Oh for ****'s sake, it's a twenty dollar bill! I could get that out of some poor bugger's wallet in the city with less effort. And besides, I'd find it more amusing sitting back and watching the other guys and girls bet money and proceeding to laugh at the wanker who ended up paying in excess of $20 for a $20 bill.

Bunch of wankers.

[Bob-san]

Another thing about the $20 game is that it really makes a difference whether or not the players think they have to pay. Did the executives actually pay $2000 and $1999? Did they think they were going to?

They do, in fact, have to pay, even in the student version.

And the thing being missed here is that the actual irrationality is likely only in a handful of the bidders; it's a GROUP of people that had the auction going higher and higher, but it only takes 2 people to bid and the person in 3rd place when it passes $20 will probably not bid again (knowing the rules say only the first and second place bidders have to pay). The actions of 1 or 2 in a group does not show the rationality or irrationality of the entire group, just those selection irrational individuals.

That's not being missed at all. Since in any group almost all bidding members are liable to end up in this position, it holds for most members of the group.

Remember, statistical tests are used to determine significance.

The heuristics that cause this kind of irrational behavior are present in anyone and are actually exacerbated by groups. They're a fundamental feature of human cognition that can be overriden by effortful processing but not removed.

This test isn't displaying any form of group-think after it passes ~$20. Past $20, as I said, it's 2 participants making fools out of themselves.

[TrashMan]

MEh...

Either don't play, or cheat as hell.
I always find these arbitrary restrictions funny...so artificial.

[Herra Tohtori]

MEh...

Either don't play, or cheat as hell.
I always find these arbitrary restrictions funny...so artificial.

Somehow, this makes your conversational conduct history in the forums so much more understandable.

Thank you for this bit of insight.

[TrashMan]

I have no idea what you're alluding to here, but I don't like it.

I was referring to the test mentioned here - for the most part they operate under restrictions that would never happen in real life (like - no communication of any kind), and IMHO, that cheapens the tests themselves.

And I really don't see what kind of "insight" you think you've gained from this, but I do know it's nothing pertinent.

[karajorma]

Seriously, lets just ignore him. The thread is far too interesting to be dragged into the usual spiral of Trashman's inability to grasp any complicated notion.

And the thing being missed here is that the actual irrationality is likely only in a handful of the bidders; it's a GROUP of people that had the auction going higher and higher, but it only takes 2 people to bid and the person in 3rd place when it passes $20 will probably not bid again (knowing the rules say only the first and second place bidders have to pay). The actions of 1 or 2 in a group does not show the rationality or irrationality of the entire group, just those selection irrational individuals.

As Battuta points out that is irrationality that ends up being displayed by whoever made the last two bids when it gets near the $20 mark. Which basically could be anybody in the group.

[headdie]

also makes you wonder how we all perceive value of something representing a concept