Skip to forum
Notifications
Clear all

[Closed] Math question

58 Posts
27 Users
0 Reactions
12 K Views
jbpatzer
Joined: 23.11.2009

Consider the following game with three balls (could do 10 balls, but 3 is much easier, and illustrates the point). Player Y puts 3 balls in a bag. He chooses n red balls and 3-n green balls with probability y_n for n = 0,1,2,3. Player X then takes two balls from the bag. He then guesses whether the next ball will be red or green. If he guesses red and is correct he wins $1, otherwise he loses $1. If he guesses green and is correct he wins $G, otherwise he loses $G. If he has seen m green balls so far, he guesses red with probability x_m, for m = 0,1,2. What y_n should Y choose, and what x_m should x choose in order not to be exploitable?

I'll leave out the details (maybe someone could check this?), but the profit for X is

(1+G)(x_0(-y_0+y_1/3)+(2/3)x_1(-y_1+y_2)+x_2(-y_2/3+y_3))+y_0+y_1/3-y_2/3-y_3

so we find that Y should choose y_0 = y_3 = 1/8 and y_1 = y_2 = 3/8. In other words, he should choose all green or all red a quarter of the time. Interestingly, this is independent of G.

We can also write this as

y_0(1-(1+G)x_0)+(y_1/3)(1+(1+G)(-2x_1+x_0))+(y_2/3)(-1+(1+G)(2x_1-x_2))+y_2(-1+(1+G)x_2)

so X should guess green with probability 1/(1+G) and red with probability G/(1+G), independent of the colour of the balls he takes out of the bag (x_0 = x_1 = x_2 = 1/(1+G)). If either X or Y follows this unexploitable strategy, their profit in the long run is zero.

Notice that if G is large, X should hardly ever guess green, but if he never guesses green, Y can exploit him by biassing the balls he puts in the bag towards green, and make a profit. It's just like poker. If someone pushes all in on the river a lot, you have to call them very infrequently in order to stop their bluffs being profitable.

Anyone fancy trying the general case of N balls in the bag?

My intuition says that

i) X should guess green with probability 1/(1+G) for all N>1.
ii) In the case of N=10 and G=1, even when X draws out 9 red balls, he should still therefore guess red and green with equal probability, because Y biasses his choice of balls so that 1 green and 9 red balls is more likely than 10 red balls by just the right (probably large) factor to make himself unexploitable.
iii) In fact, I bet the probabilities for Y are just those in Pascal's triangle
1
(11) / 2
(121) / 4
(1331) / 8
(14641) / 16
etc.
This may give a clue to an easy proof for general N.

So for N=10, Y chooses 9 red and 1 green ball 10 times as often as he chooses 10 red balls, so that the chance of drawing 9 green balls and leaving a red ball in the bag (1/10) is balanced by the fact that he puts these balls in the bag 10 times as often as he puts 10 green balls in the bag. y_10 = 1/1024, y_9 = 10/1024. Looks right to me.


Originally posted by silent21
(the person offering the wager has no information about what's in the set)

There is no villain


Originally posted by roswellx
You wrote the same thing as me... It all depends if the guy who has offering a bet knows about the number of green and not green balls. If he doesn't know then it's 50:50, if he knows then it's still 50:50 for us but he will know exactly what odds are we getting and can manipulate us to take a bet with wrong odds.

It's not a matter if he knows about their number. It's wether he put them there and chose 9 green balls out of 10 in which case if we wanted to know anything about the last ball we would have to start analysing his psycholohy, OR there is a reason why the frequency of green balls is high. This gives us other chances as I said. Actually its not 95% its 90% ^^.
Anyway maybe the author of the post could share the correct answer from the person who gave him this problem. But not too soon, it's fun topic :)


jbpatzer
Joined: 23.11.2009

Either nobody's reading my posts, or everyone thinks I've got my sums wrong. I feel I've solved this problem. Some feedback would be nice.


Clearly nobody reads mine :P

JBPatzer, there is no villain in the problem, i.e your player Y doesn't exist


jbpatzer
Joined: 23.11.2009

Exactly. If there's no villain, the problem makes no sense.


Originally posted by goldchess

Originally posted by silent21
If we continue this method we eliminate all possibilities except:

GGGGGGGGGG
GGGGGGGGGN

so the answer should be 50:50

Think closer about this, and why the probability of each can't be 50/50

Taking my own advice, and thinking closer about this,
there is no way to work out the probability of each, so yes, you are right, the problem makes no sense.


jbpatzer
Joined: 23.11.2009

But the point is that it's much more interesting if there is a villain. And it's much more like poker. I'm really into this stuff!

What happens if there are three colours? Or M colours?

?(


The problem does make sense. Maybe we can alter it a bit without modifying the final outcome. Lets assume the balls that are in the jar came from a box. The 10 balls were selected to be moved from the box to the jar randomly. Now, if we extracted and put 9 green balls one after another from the box to the jar, what is your best guess of the green ball-non green ball ratio in the box ? If you could calculate this, it would be easy to conclude the chances of getting a non-green ball into the jar ( our last misterious ball) .
Or we could put it like this:
We have a non-standard deck of cards which only contains Kinds and Aces. We shuffle them and put 9 cards on the table: All kings . What is the King/Ace ration in the deck of cards if we want to maximise the chance that by repeating the operation we have the same result ?After we calculate the ratio , what are the odds of hitting an ace on "the turn":).
If I am right, in both cases the answer is simple: The ratio between green and not green, respectively Ace and King is 1 non-green every 10 greens, 1 King every 9 aces.
The most precise odds of us winning our wage on the green ball/ace are in between 9:10 and 10:10 or 1:1 .


I think that you guys are over thinking this. The problem is what are the correct odds to accept a bet. The two possible outcomes of the last event are
1. the last ball is green
2. the last ball is not green

It cant be 50/50 because the previous events ( taking out 9 balls before, all green) influences the outcome of this event. If there was a non green ball in the jar in 90% of the doings of this experiment that ball would have been pulled out somewhere among the first 9 balls and only in 10% of the cases it would have been left last (randomly pulled out). So we need 1:9 odds to breakeven on this bet


jbpatzer
Joined: 23.11.2009

The point in my version of the problem is that player Y puts the balls in the bag in such a way that there's a 50/50 chance that the final ball is red/green, otherwise he's exploitable. i.e. he puts 9 green and 1 red ball in the bag 10 times as often as he puts 10 green balls in.

The game as I've posed it loses some of its interest because the equilibrium strategy shows zero profit for each player. It would be more interesting if one player had a slight edge over the other.

What happens if player Y has the additional option of putting in a black ball? More strategic options usually means more chances of making a profit. There could be various games depending on the consequences associated with the black ball.

For example

1) If the black ball comes out last, Y wins $B1, but if the black ball comes out before than he loses $B2
2) X has the option of guessing black.
3) something else??

No idea what the equilibrium solutions are, but it would make it more interesting. Maybe the black ball would act like the zero in roulette and give Y an edge?


Originally posted by goldchess

Originally posted by silent21
(the person offering the wager has no information about what's in the set)

There is no villain

Lol have you gone mad or didn't read the original post. Villian is who offers the wager. You should have take this bet with EV+ and OP asks what is the worst odds which makes EV=0...

I still think taking first nine balls is a separate event (how is it called in English, unconditional?) Basically events are separate. In the Bernard box paradox, both sides know that there are two black balls, there are 2 white and then one black one white.... Here we speak of Yes or No... History have no influence...


havent read previous posts so this may have already been said;

but i would work out the probability of you doing this if the 10th ball was not green, and then 1-this answer would be the probability that the ball is green? work out your odds from that probability...


Sorry for the confusion roswellx, I was meaning the villain hasn't chosen which balls to put in the bag


joeldowey123
Joined: 09.06.2010

are you putting the ball, which you have taken out, back into the jar?

this would make things a little more complicated - or easier if you didnt put it back.... this is kinda a crucial factor really


jbpatzer
Joined: 23.11.2009

When you put the balls back in the bag, the game is much more interesting. I did the calculation with the three ball game (N ball game looks hard), and the equilibrium strategy is that player Y never puts 3 red or 3 green balls in the bag (free money for X), and puts 2 red/ 1 green and 1 green/ 2 red equally often. X's strategy is to guess red if he sees 2 red balls and green if he sees 2 green balls, and to guess red or green equally often if he sees a red and a green ball. If X wins $1 when he guesses right and loses $L when he guesses wrong, then his profit per game is (5-4L)/9. So for a fair game, L = $1.25. Not obvious! It does however seem reasonable that X has an edge, since he gets to see 2 of the 3 balls.

Here's how to turn this into a con!

Tell your mark about the game and say it's 'obvious' that he has a 2 to 1 advantage, since he gets to see 2 of the 3 balls, and for a fair game, he should get $1 if he guesses right and lose $2 if he guesses wrong. However, since you're generous, you'll give him $1 if he guesses right and he will lose $1.50 if he guesses wrong. If he plays, he loses on average $0.11 cents per game!

Here's a simulation.


But jbpatzer, you are answering a different question that you made up because you knew the answer to it. The question OP posed can only be answered using information not in the question. One would have to posit the probability of details left out of the original question.


sufix645
Joined: 20.09.2009

the odds that las ball is not green is 1/(10!)


Shevtshenko
Joined: 06.12.2009

Originally posted by jbpatzer
But the point is that it's much more interesting if there is a villain. And it's much more like poker. I'm really into this stuff!

What happens if there are three colours? Or M colours?

?(

Best argument in this thread! :f_biggrin:


jbpatzer
Joined: 23.11.2009

Originally posted by cyzo
But jbpatzer, you are answering a different question that you made up because you knew the answer to it. The question OP posed can only be answered using information not in the question. One would have to posit the probability of details left out of the original question.

Well, I didn't know the answer when I posed the question, and the original question didn't make sense, on which point you seem to be agreeing with me (highlighted above)! :f_biggrin: