Current Events > Since CE has been discussing probability problems lately: the Boy Girl paradox

(+) Yeah! (+)

https://www.youtube.com/watch?v=bDZieLmya_I

Not so much a paradox as an extremely unintuitive result in probability. There's a follow-up that explains how the wording of the problem is what matters:

https://www.youtube.com/watch?v=ElB350w8iJo

---

This paradox is transphobic.

---

tldr pls
---

ThisGuyAreSick posted...

tldr pls

Someone tells you "I have two children, and one is a girl". Work through the possibilities and you'll find that there's a 33.3%, or 1 in 3, probability that their second child is a girl. However, if they tell you "I have two children, and one is a girl named Julie", then assuming (for simplicity) that 1 out of 100 girls are named Julie, the probability that their second child is a girl is now 50%. Even more bizarrely, if they say that their girl was born on a Tuesday rather than giving the name, then you find that the probability of them having a second girl is 13/27, around 48%.

---

ah so its one of those theroetical probability garbage distortions that work on paper but never actually in practice, sort of like the three doors and the two goats theory

its a 50 50 shot no matter what. i legitimately get annoyed at people passing off math magic tricks as real outcomes. it legitimately does more harm for math than good
---

ThisGuyAreSick posted...

ah so its one of those theroetical probability garbage distortions that work on paper but never actually in practice, sort of like the three doors and the two goats theory

its a 50 50 shot no matter what. i legitimately get annoyed at people passing off math magic tricks as real outcomes. it legitimately does more harm for math than good

Well there's a near-200 post topic going around about the Monty Hall problem (door and goats).

---

This one is connecting probability from separate meaningless things which have no impact on the other. With enough info you could drive the probability to less then %1 with this ridiculous criteria. Lets learn more insignificant info about our mystery person like birthday, birth place, how they like their eggs, favorite ice cream flavor etc..

Wasssup Now posted...

This one is connecting probability from separate meaningless things which have no impact on the other. With enough info you could drive the probability to less then %1 with this ridiculous criteria. Lets learn more insignificant info about our mystery person like birthday, birth place, how they like their eggs, favorite ice cream flavor etc..

Yeah it's not my favourite of these probability 'puzzles'. However the dude has some really nice videos on the use of probability and statistics in legal cases which is how I originally found him:

https://www.youtube.com/watch?v=bVG2OQp6jEQ

---

Cheese_Crackers posted...

Well there's a near-200 post topic going around about the Monty Hall problem (door and goats).

yeah the monty hall problem is one of those "look what I can do" math problems that don't actually work in the real world

the easiest counter argument to the monty hall problem is that you choose a door, leave the room, and tell someone else that wasn't in the room to pick a door. if it's still not 50/50 according to probability, you are literally assuming the physical universe begins and ends with you essentially changing the outcome with your mind
---

Cheese_Crackers posted...

Someone tells you "I have two children, and one is a girl". Work through the possibilities and you'll find that there's a 33.3%, or 1 in 3, probability that their second child is a girl. However, if they tell you "I have two children, and one is a girl named Julie", then assuming (for simplicity) that 1 out of 100 girls are named Julie, the probability that their second child is a girl is now 50%. Even more bizarrely, if they say that their girl was born on a Tuesday rather than giving the name, then you find that the probability of them having a second girl is 13/27, around 48%.

What the shit

---

Stats on paper never translate properly to real world scenarios.

Tom shady is 6-3 in the superbowl. That does not mean if he gets to another SB his chances to win are exactly 66.6666%

I can assure you that the Monty Hall problem would work in real life.

The more options that are given, under the same scenario, switching will be the option to do. If there were 1 million doors, and after you picked one, 999,998 doors were gone, there is no way in hell that you are going to tell me that you feel confident that your door is the one that is right and the other one is wrong.

---

This is theoretically easy to test.

Find a large enough sample group of families with two children, one of which is a daughter.

And then have certain groups of families give out various amounts of information about their one daughter, and after that, just see if someone going in completely blind with that information can guess the other siblings gender more accurately than someone guessing on a coin flip. >_>

---

PiOverlord posted...

I can assure you that the Monty Hall problem would work in real life.

The more options that are given, under the same scenario, switching will be the option to do. If there were 1 million doors, and after you picked one, 999,998 doors were gone, there is no way in hell that you are going to tell me that you feel confident that your door is the one that is right and the other one is wrong.

again

have someone else make the choice for you

you didnt change the probability by speaking out loud. that is literally against the laws of the universe
---

ThisGuyAreSick posted...

again

have someone else make the choice for you

you didnt change the probability by speaking out loud. that is literally against the laws of the universe

The Monty Hall problem can easily be proven in real life, just play the game using three pieces of paper with one marked as the prize.

Or if you're too lazy, simulate the game.

http://www.rossmanchance.com/applets/MontyHall/Monty04.html

No Tolerance posted...

The Monty Hall problem can easily be proven in real life, just play the game using three pieces of paper with one marked as the prize.

Or if you're too lazy, simulate the game.

http://www.rossmanchance.com/applets/MontyHall/Monty04.html

1. proceed with the paper as normal
2. get someone in the other room to pick a piece of paper
3. odds are 50/50 which is not equal to 33/66 unless you literally fucked with the laws of physics
---

ThisGuyAreSick posted...

1. proceed with the paper as normal
2. get someone in the other room to pick a piece of paper
3. odds are 50/50 which is not equal to 33/66 unless you literally fucked with the laws of physics

By using your rules, you've essentially "reset" the games and probabilities by having a new person choose between two choices. Yes, this will be a 50/50, but this does not apply to the Monty Hall problem.

The Monty Hall problem only works with three or more choices and one prize.

ThisGuyAreSick is not acknowledging that the knowledge gained is what is messing with the probabilities. You picking from the beginning out of 3 means you have the knowledge that your pick has a 1/3 chance of being the right one, and 2/3 chance that the others have the right one. When one is removed, you know that your pick still only had a 1/3 chance of being right. When the other person comes in to pick out of the remaining two, they have no knowledge and thus, yeah, out of the two there is a 50/50.

If you use the million doors example, it's even more evident. You know your pick was for all purposes, the wrong one, and when every door but your pick and another is eliminated, you know that the other one practically has to be it as you would be banking on a 1/million chance you happened to select right. When another person comes in and only sees those two doors, same thing, they won't know about the other 999,998 doors that came before.

I don't know why we are suddenly acting like knowledge isn't something that the monty hall problem includes as a major component of it. Knowledge is important for so many things such as game theory and such.

---

Theres got to be some research on this.

On the kind of people who, when confronted with overwhelming evidence that runs counter to their pre-established intuition, insist on doubling down on that instead of accepting that there was something in the world they didnt know and choose to learn something new. >_>

---

Three doors: A | B | C
Behind doors: Goat | Car | Goat

-You choose A-
Host reveals C.
Your friend comes in. You tell him that you chose A and the host revealed C. Your friend should choose B.

-You choose C-
Host reveals A.
Your friend comes in. You tell him that you chose C and the host revealed A. Your friend should choose B.

-You choose B-
Host reveals A or C, let's say A.
Your friend comes in. You tell him that you chose B and the host revealed A. We recommend your friend chooses C, but this is a loss.

By choosing a door at the start you lock in what the host is able to reveal and what he can't. In 2/3rds of the cases, you win by switching based on your first pick. If you didn't get to lock in the first door, THEN you would have a 50/50. Answer this, "What's the point of picking the first door?"

---

PiOverlord posted...

If you use the million doors example, it's even more evident. You know your pick was for all purposes, the wrong one, and when every door but your pick and another is eliminated, you know that the other one practically has to be it as you would be banking on a 1/million chance you happened to select right. When another person comes in and only sees those two doors, same thing, they won't know about the other 999,998 doors that came before.

I dunno why, but this is the first time this has really clicked for me.

I guess I never saw the exact phrasing "you picked the wrong door."

---

ThisGuyAreSick posted...

ah so its one of those theroetical probability garbage distortions that work on paper but never actually in practice, sort of like the three doors and the two goats theory

its a 50 50 shot no matter what. i legitimately get annoyed at people passing off math magic tricks as real outcomes. it legitimately does more harm for math than good

The Monty Hall problem's solution has been confirmed by, among others, the Mythbusters themselves.
See all the red boxes on the right compared to the left? Right side switched doors every single time and was successful more often than the left side which stayed on each attempt.

---

No Tolerance posted...

By using your rules, you've essentially "reset" the games and probabilities by having a new person choose between two choices. Yes, this will be a 50/50, but this does not apply to the Monty Hall problem.

The Monty Hall problem only works with three or more choices and one prize.

i didnt reset shit

two choices = 50/50

period
---

what youre basically saying is

guy picks door, another door is taken away. if he picks same door its 33%

guy picks door, another door is taken away. new guy comes in. happens to pick same door. now its 50%?

this is why the monty hall problem is bullshit. you're conceding that universal probability changes depending on whether you personally think something into existence. that's astronomically absurd
---

AlisLandale posted...

Theres got to be some research on this.

On the kind of people who, when confronted with overwhelming evidence that runs counter to their pre-established intuition, insist on doubling down on that instead of accepting that there was something in the world they didnt know and choose to learn something new. >_>

It's either arrogance or trolling
---

i have a third answer

not everything that can be proven on paper means it exists physically.

like, why is it that we can "prove" multiverses and shit on paper but the entire scientific community acknowledges that physically it has yet to be proven? but when it comes to probability craziness everyone goes "ITS TRUE THE MATH PROVES IT"

theres so much mathematical "glitches" that result from number manipulation and an absurd set of criteria. take into account imaginary numbers, or .9999999~=1, and so on and so on....
---

ThisGuyAreSick posted...

not everything that can be proven on paper means it exists physically.

Like the human brain in some people.

---

SerperiorThanU posted...

This isnt about universal probability, this is about the knowledge of the user making the choice. Like if the guy cheated and peeked ahead of time and they know the left door 100% is the right one. The choice is 50/50 for everyone else, but 100% for the guy making the choice.

youre equating picking a random door to picking a door you know to be correct

not the same thing at all

again, if you want to concede that the probability goes up based off pure ignorance of the initial door chosen, youre literally arguing the universe changes depending on someone else saying something out loud. otherwise the percentages would be the same.
---

ThisGuyAreSick posted...

what youre basically saying is

guy picks door, another door is taken away. if he picks same door its 33%

guy picks door, another door is taken away. new guy comes in. happens to pick same door. now its 50%?

this is why the monty hall problem is bullshit. you're conceding that universal probability changes depending on whether you personally think something into existence. that's astronomically absurd

having knowledge of the whole scenario is the important part here

---

ThisGuyAreSick posted...

ah so its one of those theroetical probability garbage distortions that work on paper but never actually in practice, sort of like the three doors and the two goats theory

its a 50 50 shot no matter what. i legitimately get annoyed at people passing off math magic tricks as real outcomes. it legitimately does more harm for math than good

---

averagejoel posted...

having knowledge of the whole scenario is the important part here

it's still 50/50

the odds for the other door don't go down just because someone else entered the room and told you what new door to pick. if it did we need to be researching this shit right away as it could lead to the discovery of multiverses
---

SerperiorThanU posted...

Thats the whole point with the Monty Hall problem. Someone who knows which door is correct makes it more likely for you to get the right door.

if someone cheated but decided to have someone else pick the odds are still 50/50

thanks for playing
---

Cheese_Crackers posted...

Someone tells you "I have two children, and one is a girl". Work through the possibilities and you'll find that there's a 33.3%, or 1 in 3, probability that their second child is a girl. However, if they tell you "I have two children, and one is a girl named Julie", then assuming (for simplicity) that 1 out of 100 girls are named Julie, the probability that their second child is a girl is now 50%. Even more bizarrely, if they say that their girl was born on a Tuesday rather than giving the name, then you find that the probability of them having a second girl is 13/27, around 48%.

What is this pseudoscience bullshit?

---

ThisGuyAreSick posted...

it's still 50/50

the odds for the other door don't go down just because someone else entered the room and told you what new door to pick. if it did we need to be researching this shit right away as it could lead to the discovery of multiverses

this has nothing to do with multiverses

there are 3 doors. you pick one. one door is removed.
if you switch, you get the prize 2/3 of the time. if you stay with the one you chose, you get it 1/3 of the time.

it's already been explained multiple times itt. it's also more intuitive with larger numbers

---

so what youre saying is the odds of the other door go down simply if someone else enters the room and picks it

so what youre saying is the laws of physics can change on the back of shrodinger's cat

fucking nuts
---

ThisGuyAreSick posted...

it's still 50/50

the odds for the other door don't go down just because someone else entered the room and told you what new door to pick. if it did we need to be researching this shit right away as it could lead to the discovery of multiverses

Feel free to get yourself some pieces of paper and a very patient friend and try it out for yourself a thousand times. You'll see that switching doors wins more often.

---

Hornezz posted...

Feel free to get yourself some pieces of paper and a very patient friend and try it out for yourself a thousand times. You'll see that switching doors wins more often.

ThisGuyAreSick posted...

i have a third answer

not everything that can be proven on paper means it exists physically.

like, why is it that we can "prove" multiverses and shit on paper but the entire scientific community acknowledges that physically it has yet to be proven? but when it comes to probability craziness everyone goes "ITS TRUE THE MATH PROVES IT"

theres so much mathematical "glitches" that result from number manipulation and an absurd set of criteria. take into account imaginary numbers, or .9999999~=1, and so on and so on....

---

Yo, ThisGuyAreSick is getting so mad in here, I think he just said imaginary numbers dont work.

Id love to see his explanation of why the million door version is a 50/50.

---

Jerry_Hellyeah posted...

Yo, ThisGuyAreSick is getting so mad in here, I think he just said imaginary numbers dont work.

Id love to see his explanation of why the million door version is a 50/50.

its 50/50 for the guy entering the room. he literally just changed probability by just existing.

thats some craziness that needs to be investigated immediately
---

YO

THIS GUY JUST SAID "NOT EVERYTHING CAN BE PROVEN ON PAPER MEANS IT EXISTS PHYSICALLY" IN RESPONSE TO BEING TOLD TO USE PAPER.

You can use fucking coconuts or something. Unless.....COCONUTS DONT EXIST???

---

Jerry_Hellyeah posted...

YO

THIS GUY JUST SAID "NOT EVERYTHING CAN BE PROVEN ON PAPER MEANS IT EXISTS PHYSICALLY" IN RESPONSE TO BEING TOLD TO USE PAPER.

You can use fucking coconuts or something. Unless.....COCONUTS DONT EXIST???

so when time travel and multiverses happening?

the paper proves them yet scientists refuse to believe they are real. smh silly dumb scientists the paper says so!
---

ThisGuyAreSick posted...

i have a third answer

not everything that can be proven on paper means it exists physically.

That's why you should try it out physically. Make 3 doors, place a prize without a friend seeing, have them pick a door, open one goat door and let him decide if he wants to switch. Take note of whether or not he chose to switch and whether or not he won.

Repeat a couple thousand times and report back here.

---

Hornezz posted...

That's why you should try it out physically. Make 3 doors, place a prize without a friend seeing, have them pick a door, open one goat door and let him decide if he wants to switch. Take note of whether or not he chose to switch and whether or not he won.

Repeat a couple thousand times and report back here.

and what happens if i get another friend to pick a random door

is it the first step in entering the multiverse?
---

What is it you exactly get out of pretending you don't understand basic things?

---

Hornezz posted...

What is it you exactly get out of pretending you don't understand basic things?

i want to know why you people believe that a person entering the room changes the laws of mathematics

like that sounds crazy to me
---

ThisGuyAreSick posted...

i want to know why you people believe that a person entering the room changes the laws of mathematics

like that sounds crazy to me

Can you.....can you really not grasp the concepts of "information" or "learning"?

---