The Monty Hall Problem

[BarthelloSylph]

double

[Millael]

http://math.ucsd.edu/~crypto/Monty/monty.html

Thanks for helping me with the proof.

--------------------# of Players------Winners----Percent Winners
Switched---------------462------------319-----------69.0
Didn't Switch----------430------------147------------34.2


add total number of players: 892

add total number of winners: 468

total estimate of percent winners: do it yourself

ARE YOU FUCKING RETARDED? Holy christ. Look. The people who SWITCHED won 66% of the time, approximately. The people who DIDN'T won 33% of the time approximately. I've tried being reasonable about this, but I've snapped. I'm sorry. I give up.

If you average the 2/3-odds takers and the 1/3-odds takers, of fucking course you're going to have a 50% ratio.

[GRT]

GRT those are some of the dumbest posts I've ever read on any message board ever.

If you really think you are breaking the boundaries, and discovering some new mathmatical shit, then you are really just a sad excuse for moron.

But all that aside GRT, I think what aurik is getting at is this. And I'll try to speak from the heart here.

-------------------------------------

Begin initial phase. GRT, all I am gonna say is - put yourself in the shoes of the contestant. There are three doors, 2 have a goat behind them and 1 has a car. You pick a door to start off, and then the gameshow host will reveal a goat behind one of the two doors you did not. End initial phase.

Begin decision time. GRT, let's say you have a choice. Stick with your current door, or switch doors. Let's say you have to pick one or the other You can't say "it doesnt matter", you have to pick... which is it?

um

i'm kinda speechless, but i'll speak anyway

i never claimed to have any breakthrough math theories, don't put words in my mouth

statistics already proved 1/2 is the right answer, why are we arguing? or have you guys misunderstood from the very beginning what the question was asking?

personally i will switch, but just as the statistics show, NOT EVERYONE WILL SWITCH, in fact the numbers does also indicate how many people switched and how many didn't, and it is shockingly ( ) 50/50

again, we are talking about probability here, not what i personally will do... if you don't know what i'm talking about, go back to school and study statistics/probability

[GRT]

http://math.ucsd.edu/~crypto/Monty/monty.html

Thanks for helping me with the proof.

--------------------# of Players------Winners----Percent Winners
Switched---------------462------------319-----------69.0
Didn't Switch----------430------------147------------34.2


add total number of players: 892

add total number of winners: 468

total estimate of percent winners: do it yourself

ARE YOU FUCKING RETARDED? Holy christ. Look. The people who SWITCHED won 66% of the time, approximately. The people who DIDN'T won 33% of the time approximately. I've tried being reasonable about this, but I've snapped. I'm sorry. I give up.

If you average the 2/3-odds takers and the 1/3-odds takers, of fucking course you're going to have a 50% ratio.

calm down, i'm sure there are plenty of support groups for people who have difficulty admitting that they are wrong

[Byrd]

statistics already proved 1/2 is the right answer, why are we arguing? or have you guys misunderstood from the very beginning what the question was asking?

The question is: Does switching improve your chances to win the car?

The answer is: Yes. (from 1/3 to 2/3)

Pray tell what question have you been trying to answer?

[GRT]

statistics already proved 1/2 is the right answer, why are we arguing? or have you guys misunderstood from the very beginning what the question was asking?

The question is: Does switching improve your chances to win the car?

The answer is: Yes. (from 1/3 to 2/3)

Pray tell what question have you been trying to answer?

scroll up to the top of this page, it's not hard to do, click that little "up" arrow on the right side of your browser window

What's your chance of getting the Car after the host shows you a goat?

your options are: 1/2 or 2/3

i guess i better get some glasses cuz i don't see "Yes. (from 1/3 to 2/3)"

[Enkidu]

ITT we try to save face.

[Blarg]

Back at the start it was stated that the poll question was worded poorly, read the first post again and then answer based on what he is asking.

[BarthelloSylph]

is it really that hard to understand? i remember doing IQ test as a kid i was only average

lol 80 isnt average, sorry

and the stats above prove our point

IF YOU SWITCH, YOU WIN 66% OF THE TIME
IF YOU DONT, YOU WIN 33% OF THE TIME

HOW ABOUT THAT?? TWO OPTIONS (lol50/50chance) and one wins 2/3s of the time and one wins 1/3

GRT you just got OWNED

[Byrd]

What's your chance of getting the Car after the host shows you a goat?

The truly correct answer to that poorly worded question is "Depends on if you switch or not." Even the OP admitted it's a poorly worded question, though, and that was established about page 2 in the dialogue. You should really read the whole thread in something like this, it helps tremendously.

[BarthelloSylph]

Well even if you don't reword the question, if you assume the person trying to win the car WANTS TO WIN and IS SMART, he will switch, and thus win a car 66% of the time.

Only an idiot would lower his odds to 50%.

[Byrd]

Only an idiot would lower his odds to 50%.

Actually, assuming you know which door you originally picked, you can't get 50% odds unless you do something like flip a coin whether to switch or not. Provided you make a decision, you can only have 33% or 66%.

[BarthelloSylph]

Touche Byrd. You are correct.

I'm sure xfaustx and GRW or whatever will come together tomorrow to prove that and everything else wrong, and go on live incredible lives as breakthrough mathematicians.

Oh wait, I've got some breaking news. That was supposed to be molesters. They will become molesters. Oops.

[GRT]

What's your chance of getting the Car after the host shows you a goat?

The truly correct answer to that poorly worded question is "Depends on if you switch or not." Even the OP admitted it's a poorly worded question, though, and that was established about page 2 in the dialogue. You should really read the whole thread in something like this, it helps tremendously.

Actually I did read the original post (but skipped the rest of page 1 as well as page 2 and 3) which is why I got the 1/3 increase that I posted on page 4, which is not one of the poll options. Then I reread the question on the poll, and I don't know about you, but that reads like "overall" chance of winning the car more than anything else, thus I picked 1/2.

I apologize if my arguements thus far confused anyone. But if you've read them, I have explicitly argued and emphasized throughout that the overall chance of winning is 50%, and the first choice is irrelevent to the entire process in regards to overall probability (which I received waves and waves of illogical rebuttal).

So that's that. If it helps you sleep better, I did get pwned by Byrd who cleared up the question.

[BarthelloSylph]

Why did you continue to argue that when it was obvious that the rest of us were arguing what the thread intended? Are you an idiot?

[Enkidu]

Actually I did read the original post (but skipped the rest of page 1 as well as page 2 and 3) which is why I got the 1/3 increase that I posted on page 4, which is not one of the poll options. Then I reread the question on the poll, and I don't know about you, but that reads like "overall" chance of winning the car more than anything else, thus I picked 1/2.

I apologize if my arguements thus far confused anyone. But if you've read them, I have explicitly argued and emphasized throughout that the overall chance of winning is 50%, and the first choice is irrelevent to the entire process in regards to overall probability (which I received waves and waves of illogical rebuttal).

So that's that. If it helps you sleep better, I did get pwned by Byrd who cleared up the question.

There is not an "overall" 50% chance of winning. It depends solely on the contestants. If you get 100 contestants who believe staying is better, you end up with ~33% winners. If they believe switching is better, ~66%.

To answer the original question with a %chance would involve intense psychological studies and it would never, ever, be foolproof. Like Byrd said, the answer to the original question is simple, it depends on whether they switch or stay.

[Seyrr]

http://math.ucsd.edu/~crypto/Monty/monty.html

Thanks for helping me with the proof.

--------------------# of Players------Winners----Percent Winners
Switched---------------462------------319-----------69.0
Didn't Switch----------430------------147------------34.2


add total number of players: 892

add total number of winners: 468

total estimate of percent winners: do it yourself

I ran one myself!

--------------------# of Players------Winners----Percent Winners
Switched---------------100-------------63-----------63.0
Didn't Switch----------10000------------3219------------32.2

HOLY SHIT OUT OF 10100 PLAYERS, ONLY 3282 WON! THAT MEANS THAT PEOPLE WON LESS THAN 33% OF THE TIME!

It's a meaningless statistic that you computed when you did that GRT. Seriously, look at what your statement is.

"Out of about 900 people, when half of them switched and half didn't, they overall won about 50% of the time."

Look at your own statistics. People who switched won 69% of the time. People who didn't switch won 34% of the time. There's a clear difference in results between people who switch and people who don't.

You're grasping at any straw you can to avoid admitting you're wrong, and it'd be hilarious if it weren't so sad.

EDIT: aw, I'm late to the party. teach me to Besieged instead of finishing my post.

[BarthelloSylph]

The overall chance is never 50% GRE, you are still wrong. GTFO

[GRT]

Actually I did read the original post (but skipped the rest of page 1 as well as page 2 and 3) which is why I got the 1/3 increase that I posted on page 4, which is not one of the poll options. Then I reread the question on the poll, and I don't know about you, but that reads like "overall" chance of winning the car more than anything else, thus I picked 1/2.

I apologize if my arguements thus far confused anyone. But if you've read them, I have explicitly argued and emphasized throughout that the overall chance of winning is 50%, and the first choice is irrelevent to the entire process in regards to overall probability (which I received waves and waves of illogical rebuttal).

So that's that. If it helps you sleep better, I did get pwned by Byrd who cleared up the question.

There is not an "overall" 50% chance of winning. It depends solely on the contestants. If you get 100 contestants who believe staying is better, you end up with ~33% winners. If they believe switching is better, ~66%.

To answer the original question with a %chance would involve intense psychological studies and it would never, ever, be foolproof. Like Byrd said, the answer to the original question is simple, it depends on whether they switch or stay.

nah, in mathmatics human behavior is irrelevent, which means that there's no such thing as studying the result of your choice to help you make the decision, there's only the mechanical random choices in the study of probability

i think that should be right, i'm not a math major but i did hear something like that in upper division statistics class

so basically if you run the entire scenario in a computer over and over again the result will always be roughly 50%, which is what i meant by "overall" probability


btw i'm ignoring the other two guys because one makes up numbers and other one is obviously a conformist idiot

[Enkidu]

nah, in mathmatics human behavior is irrelevent

Right, but then how do you postulate that 50% will choose to stay and 50% choose to switch? That prediction is baseless. Mathematics would answer the question like this: Those who stay have a 1/3 chance of winning and those who switch have a 2/3 chance of winning.