Monty Hall Problem:
How in God's name does switching doors I'improve your chances?
There are three doors. You pick one. One door is eliminated. Leaving two. 50/50 chance. Monty Hall is bullshit.
There's already a probability for retards thread
>>7980857
You're a retard if you believe that Monty Hall trash. The chance is obviously 50/50 still, and switching doors makes no difference.
>>7980864
>I can't understand it and it's not intuitive so it's wrong WAAAAAAH
Just go in the other thread, it's explained there
>>7980853
The door that is eliminated isn't eliminated at random.
>>7980872
So what?
>>7980878
Right, so you have three doors. One is removed. You have a 1/2 chance. Christ, you mathematicians are retards.
>>7980874
So which door is eliminated gives you some information about which door has the car, since the eliminated door isn't chosen at random.
I'm just saying how, intuitively, it is possible for you to get information from which door is eliminated, and therefore make a more "informed" guess as to which door has the car. Your new guess it more "informed" than your original random guess, and turns out to have a 2/3 chance of being right.
The only reason you don't see what's happening is because of how small a number 3 doors is.
If there are 100 doors, and you pick one, and Monty opens all of the doors that you didn't pick that are wrong, would you switch?
>>7980853
Try this mate.
http://youtu.be/7u6kFlWZOWg
>>7980899
Obviously. There's a 99/100 chance I picked the goat.
The doors are opened NOT RANDOMLY. The host knows in advance which door has the prize behind it, and opens other doors.
If they happened to randomly open without revealing the prize, then the chance would be 50/50 as is immediately intuitive to most.
One only needs to up the total number of doors a bit to figure out how wrong they're being. Imagine there's a thousand doors, and you pick one. Then the host opens all doors but one without revealing the prize. Do you still believe there's a 50/50 chance there's a prize behind your door?
>>7980893
But the situation only arises of Monty is lucky enough to pick the door with the goat. The situation doesn't come about if he picks the door with the car behind it.
>>7980911
So you're saying Monty knows what is behind the two doors? The two door situation is GUARANTEED to arise?
>>7980904
Right. It's the same principle. When you have three doors, and pick one, there's a 2/3 chance that you picked the goat that time, meaning there's a 2/3 chance that taking the switch will give you the car.
>>7980916
He [math]must[/math] pick a door with a goat. It is stated in the problem that he opens a door with a goat. He knows what is behind each door, and will never pick a door with a car.
>>7980919
Yes. It's in the problem!
Direct from the wikipedia statement of the problem:
>You pick a door, say No. 1, and the host, who knows what's behind the doors, opens another door, say No. 3, which has a goat.
>>7980853
the probability is a wave function of the cube root of the square of the two doors left over after eliminating the third one.
For instance, eating one grape and having two left over provides you with the probability that one of the grapes will have a bread new car behind it.
Therefore when eating grapes it is always best to save the last one for last and switch your choice form red grapes to green because green grapes have a higher functional taste distribution.
>>7980903
Brady HARAM
>>7980853
It's appropriate that you posted Erdős.
He was a mathematical genius, yet wikipedia says he remained unconvinced of the solution to the Monty Hall problem until he was shown a computer simulation verifying the result.
Completely bizarre.
>>7980928
more like hiram the original servant of the architect of the universe amirite?
>>7980893
If you're going in knowing you're gonna switch anyway, why don't you just pick a different door from the start? Isn't that the same thing?
>>7980927
I like red grapes better than green grapes, green grapes are sour.
>>7980932
no, because regardless of what door you choose the first choice you make has a 1/2 chance of being correct. therefore half of the grapes are red, when chosen, they shall simplify according to the larenzo conjecture into green grapes.
>>7980937
actually the first choice you make has a 100% chance of being totally correct.
changing doors is an illuminati mind trick.
don't let them steal your grapes.
>>7980869
>implying it will ever get intuitive
I'll give you a hint OP
your first choice is a 1/3 of being correct. Then one door is revealed to be a goat.
>>7980958
That's one crafty ass goat if it can disguise itself as a door.
>>7980967
Nice.
>>7980967
or perhaps all the things we think of as doors are actually goats.
What happens if we're on Deal or No Deal? We pick a case, and then guess away cases until the only two cases left are the $1 and the $1,000,000 one.
There's a one in 26 chance of getting either case on your first try.
>>7980982
The difference is you aren't allowed to switch after being shown a case. If you were allowed to switch after being shown the case, obviously people would switch after getting a large case value and never for a small case value. If you get the $1M case and are allowed to switch off of it, you just remember the case number and always win the million.
If when you choose a case, all of the cases are removed except two, the million and your case (or a random case if the one you chose was the million), then instead of a 1/26 case chance you have a 1/2 chance to get the million. Those are some better odds because the first was chosen when there was a 1/26 chance of winning versus the current 1/2 chance of winning. The second case has a greater chance to be the winner therefore and switching is the best option.
>>7980899
THis explanation is cute, but you are comparing two different things.
In the original problem there are three doors. No matter what, you will always come down to two doors because the host will eliminate a goat door.
I'm gonna say that again:
NO MATTER WHAT, YOU WILL COME DOWN TO TWO DOORS.
Essentially the game makes a goat dissapear whatever the first pick., so you're really only picking between two doors.
>>7982435
Nigga you dumb. It's the change in the situation that drives the change. You have a 1/3 chance of choosing the right door initially and after the removal you have a 1/2 chance. Given that your first choice was based on a lower percent chance to get the right door, switching increases your chances of winning from 33% to 50%.
You're assuming that you can ignore the loss of those extra doors, which you can't.
>>7980853
Probability space.
Three doors, under procedure where car is placed randomly behind one of them would each have 33% chance of being randomly picked as winners.
Monty Hall's fuckery consolidates the probabilities of two remaining doors into one door, that's why the switch situation is a question of whether to pick the the freshly consolidated space of 33% + 33% = 66% or to stick with your originally picked chunk of probability space - 33%.
Same concept applies with increased number of doors. If there are 100 doors and a car is randomly placed, then each door has a 1% chance of being randomly picked as a winning door. When you pick 1 door and Monty reveals 98 other doors as duds, you know that the 1 remaining door inherits all of the winning probability previously distributed in the space of 98 other doors + itself. So, when time comes to switch or not, your choices are 1 - stay with the door where the car had 1% chance of being randomly placed or switch to a door with 99% chance of being the one where the car was randomly placed.
Monty Hall problem confuses people because they ignore the fact that WHEN the car is placed behind the door matters. The 3-door Monty Hall example would only be 50/50 at 2 doors if there were no winning doors until 2 were left and only after 2 were left the winning door was assigned behind the scenes.
>>7980929
Probably not a coincidence. OP has stricken himself to be a fool's fool.
>>7982446
The situation leads to a 50/50 chance, so either door you pick has the same probability of win or loss because like coin flips, the previous choice doesn't even affect the current flip.
>>7982519
If you don't choose and a bad door is removed, then yes, it is a 50/50 choice.
However if you choose first, you are choosing on a 33/33/33 chance. If you stay with that choice, you are betting on lesser odds from beforehand once it changes to 50/50.
>>7982522
If you flip two coins and only have a 25% chance of getting two head, then you flip a third coin, that 25% doesn't affect the odds of getting heads again in any way, previous rounds of choice don't affect the odds for the last round in any way.
Not gonna read the thread, but do what I did: make a babby java program and see what happens.
>>7982534
It does affect it if the question is "What is the chance of 3 heads in a row?" A similar explanation is for the doors. What is the chance of choosing a winning door on the first door? 33%, with 66% for losing. Now we remove a wrong door. What is left is a 50% chance to have the winning door and a 50% chance to have a losing door.
You are statistically more likely to have already chosen a losing door. This is obvious because 66% > 33%. So now when the odds change to 50% vs 50%, given that we already have a losing door, choosing the other door will result in more wins than if you stick with the original door.
>>7982544
I'm lazy as fuck. Just go run the simulator here: http://www.grand-illusions.com/simulator/montysim.htm
>>7982548
>"What is the chance of 3 heads in a row?"
That isn't the question at all, though, you don't have to choose right the first time, it doesn't factor into the odds at all, it is a red herring because flipping two heads, just unlocks the chance to flip for a 50/50 shot.
>>7980853
Think about 1000 doors, pick one, the guy opens 998 of the other ones with goats in them. You still think that first pick of 1 in a 1000 is larger chance than the lone door remaining?
>>7982556
Nigga go run the simulator and prove yourself wrong. Stop treating them as independent events.
>>7982561
The last move is independent of the first, they shouldn't be treated as dependent, if you flipped a coin to decide if you should roll a die, the coin results would have no actual effect on the probability of the die roll, it would just affect the staging and timing of the roll.
Why would I trust your simulator when you are obviously going to bias
>>7982570
>why would I trust a simulator
Obvious bait at this point. Fuck off.
>>7982571
I can make a coin flip simulator that give tails a 66% distribution too, it doesn't make it based in statistical reality.
>>7982580
Go code the problem as stated and you will get a 66% result. Shit, just read wikipedia even.
>>7982548
I finally truly understand it.
Thank you.
>>7982548
>You are statistically more likely to have already chosen a losing door. This is obvious because 66% > 33%.
All three doors were equally statistically unlikely to win, even if you switch doors, that door you switched to still only had a 33% chance the first time, the 66% is a red herring, nothing ever had a 66% chance of being correct.
>>7982602
There are 2 goats and 1 car, totaling 3. Therefore, 2/3 of the choices are goats, 66%. 1/3 of the choices is a car, 33%.
How the fuck is this possibly wrong before the part where he removes one of the goats? Are you just that bad at statistics?
>>7980853
Probability that you win by switching is actually varying between 66% an 0% depending on how often the host would be unhonest. If host would prompt you to switch only when you hit the right door, switching gets you 0% chance of being right.
>>7980929
>Erdős
>His colleague Alfréd Rényi said, "a mathematician is a machine for turning coffee into theorems",[15] and Erdős drank copious quantities (this quotation is often attributed incorrectly to Erdős,[16] but Erdős himself ascribed it to Rényi[17]). After 1971 he also took amphetamines, despite the concern of his friends, one of whom (Ron Graham) bet him $500 that he could not stop taking the drug for a month.[18] Erdős won the bet, but complained that during his abstinence, mathematics had been set back by a month: "Before, when I looked at a piece of blank paper my mind was filled with ideas. Now all I see is a blank piece of paper." After he won the bet, he promptly resumed his amphetamine use.
>>7982609
So how does one door that only had a 33% chance of being a car get the 66% advantage of both goat doors when one door is eliminated?
>>7982622
Because when one door is eliminated both remaining doors have a 50% chance of being chosen. However the additional information of the first choice means that switching gives us a greater chance of success due to the likelihood that the first choice was a losing door being higher. The precise amount of increase/decrease is 16.67% for the winning/losing door, respectively. If you might note this is 1/6, which is our original probability of 1/3 (33%) multiplied by the secondary probability 1/2 (50%).
>>7980853
Partial measures, anon. By seeing that one door, which had a prior 1/3 chance of holding the prize, doesn't actually hold the prize, you've been able to "cheat" and dismiss 1/3 of all possible outcomes. Yes, the probability of winning is now 50-50 with either door, >>7982586
Oh whatever. Chances are you lost with your first pick.
>>7982634
>roulette 50% red/black
>70 reds in a row
>Black!, i'm gonna win for sure!
>It actually has a 50% of winning
>>7982654
Except all of those are independent events while the Monty Hall problem is two dependent events. Also sick gambler's fallacy, it has no application to this problem.
>>7982634
>the likelihood that the first choice was a losing door being higher
It had the exact same likelihood of winning or losing as the door you want to switch to, no door was any higher or lower than any other.
>>7982663
There are TWO chances to pick a losing door at the start and ONE chance to pick a winning door AT THE VERY BEGINNING OF THE PROBLEM. Are you trying to tell me that 2 equals 1? You are more likely to have chosen a losing door than a winning door because 2 > 1.
>>7982668
You don't want to pick a losing door, the goal is to pick a winning door and no door ever has more than a 33% chance at the beginning and 50% at the end when they real game happens and the pointless theatrics of the first round are finished.
>>7982700
I'm wondering if you can add numbers. Or even do any math. Because clearly you don't understand anything.
Tell me, what's the total chance of getting any losing door before any are removed? Hint: if you don't say 66% you're a fucking retard.
>>7982708
I don't have a loser mentality like you, got cry to your momma about how someone else picked the winner.
Protip: Monty Hall was scamming people.
The second round always happened if you picked a winner, 3/5 of the time he would end the game after the first pick if they chose a goat to make it look on the up and up, the second round was to get people to switch from the winning door and this troll thought experiment was made to trick idiots.
>>7980887
who removed the door?
was it someone with prior knowledge?
will he ever remove the prize door, or does he ALWAYS remove non prize doors?
>>7982761
>the second round was to get people to switch from the winning door
so you're saying monty NEVER removed the prize door? huh
>>7982621
Doesn't mean he wasn't a genius.
>>7980853
the door you picked first has a 1/3 chance of being the car, so the door that isn't revealed to be a goat has a 2/3 chance of being the car
>>7983520
No, they always showed which door had the prize at the end, he just had to trick people into picking a wrong one.
>>7980927
Thanks m8, now it's all clear
>>7982660
>A,B,C doors
>Chemicals in your brain and experience are fated to choose A.
>Before opening the door the program automatically tells you : C is fake. Which door will you choose?
>There are two doors.
>You have a 50% chance of winning.
Or
>Play the same game 10^6 times putting the car in the door A.
>Play again
>C is fake
>B door of course!
>>7982637
Then, there are two games
1. Choose 1 of the 3 doors
2. Choose 1 of the 2 doors
What a bunch of useless wordplay
i'll put this shit all to rest by showing all possibilities:
There are three possibilities of your first pick
1. you pick goat1
2. you pick goat2
3. you pick car
the gameshow host will then remove a door with a goat behind it so the possibilities are thus
1. you picked goat1, he removed goat2 so what's left is goat1 (your original pick) and car, so you should switch
2. you pick goat2 and the gameshow host removes goat1, so what's left is goat2 (your original pick) and car, so you should switch
3. you picked car and the gameshow host removed a goat (1 or 2) so what's left is the car (your original pick) and the other goat so you should not switch
2/3 times you should switch and 1/3 times you should not. It's that easy
The reason behind this is, switching after the removal of a goat ensures that you will be switching from your original pick to the other type (goat to car or car to goat). Because your original pick is 2/3 chance of getting a goat then the switch is 2/3 likely to be a switch from goat to car
i think it's a very intuitive concept but wiki said some stats phd's wouldn't believe it until they saw computer simulations with random number generators so i can see why you, OP, might just be as stupid as them
Holy fuck guys
This is conditional probability.
Let's say we have a ocean. (pacific, Indiana, Mediterranean) and we are looking for a particular fish.
We want the fish that only lives in mediterranean ocean but water looks all the same so we have to pick up one water source at random, thus 1/3 probability that we pick meiterranean water.
Given that fact, we define our choice of water as Ocean1.
Notice our chance of picking up the right type of water is much lower (higher possibility that we get it wrong) than the correct one. because 1/3 vs 2/3. (there's 1/3 chance that our choice is Mediterranean vs. 2/3 chance that other two has Mediterranean water)
We still have to focus on our choice though because we got no choice.
Within that selection we start looking for the fish. Then someone calls out that out of unselected water (2/3) one of them is pacific water. Now we are sure that what we have is actually Mediterranean based on the new fact. Then he asks if I want to change the water of other one. Given new information other water might be equally likely than what we have. so 1/2 chance. Remember though your initiative correctness was only 1/3 vs. other one had inititive correctness of 2/3. Thus when you change, there's higher probabiily that you might get the right one
The host is always going to display a goat to you after you have picked a door.
With that in mind, you have two goats and one car, so your first pick has a 2/3 chance of being a goat and 1/3 chance of being the car.
Knowing your chances of selecting one of the two goats at the start was high, and seeing the other goat accounted for, it makes sense to switch your choice to the third door as it is more likely to be a car.
>>7985771
>100 doors bet game
>choose A
>they quit 98. B door left.
>/sci/:"monty hall, man, monty hall!"
Or
Repeat 1000 times the game and see if there is actually a higher probability to win if we pick the other door.
>>7980916
Are people really this retarded? This problem is literally explained the first day of any statistics/data analysis class.
It isn't even hard! The host knows what door has the zonk so he picks it! The knowledge of the host is what makes the problem true.
Who even is on this board? Who are the people here that cannot understand this. Whenever I post a legitimate question I get raped and called a retard. Is everyone just butthurt on this board?
>>7985814
Are your retarded? Just make a matlab program that simulates 1000 trials. The Monty Hall problem holds true. It is conditional probability.
>>7985821
Entry probability questions are some of the best and easiest /sci/ trolls. Despite the math being very basic, you cannot teach intuition.
If each door is 1/3 of being a winner, then the 2 doors that arent yours have a 2/3 combined of being the winner. When you eliminate one of those doors, the remaining door retains its 2/3 chance of being the winning door.
>>7985822
But what I am saying is how can I be shit on by "PhDs" and "masters students" when they cannot even understand this sophomore level course. This is required material for almost any future studies. I think that everyone here is lying about how smart they are.
Explaining it like that usually helps:
imagine that you can choose a door. Once you have chosen one, the presentator allows you to switch on the remaining two doors as long as you discard one goat as prize (there's always at least one goat in the two doors you can switch on). This way of tweaking the problem formulation doesn't change the result since there's no difference between the presentator opening one door before the switch and you giving up on one goat later, but helps visualize that you are giving up on one door for 2 if you switch.
>>7985822
>>7985822
Anon how is this applied in the real world?
>guy:"what's the probability to get aids from a person?"
>you know: a% aids people
>Then you remember that ten years before you had 50% probability of dying by toxic inhalation.
>"a/2%"
Anon, the sample space changes, so the probability of every choice should be reset.
>>7985893
i mean HIV
>>7985893
It is ironic because it is used all the time in the medical industry. I am sure you are one of those kids that asks the professor "when will we ever use this in real life". Can you seriously not understand the Monte Hall problem or are you trolling me?
I understand its mathematical approach.
Nevertheless I don't get what kind of phenomenon happened, when you chose the door A, to affect the result.
There is no phenomenon,, so there must be an error.
You should have changed the sample space before calculating probabilities again.
>>7985821
>The host knows what door has the zonk so he picks it!
And reveals it to you, still leaving a 50/50 chance. You clearly don't understand this problem.
>>7985928
Just make a matlab program to test this! You can even see it work right in front of you! The problem is true holy fuck!
>>7985923
when you first choose you picked a goat 2/3 of the times, as simple as that. In that case if you switch you're sure to get the car since the other goat gets eliminated, so the 2/3 above directly translates to 2/3 canche of getting the car if you switch
>>7985928
you have to consider the probability you first picked a goat when there were first three doors. does >>7985846 help?
>>7985935
>the death goat game
>You are poisoned and have a 50% of survival
>you need to choose between two doors A or B to win, but the poison paralyzes you.
>You survive.
>What's the probability to win?
>"25%"
Anon, when you tell me about the MHP that I already chose the door A and the host eliminated 1 option from the sample space. Weknow that we should reset the probabilities of each option.
>>7985956
you're forgetting that the way the host eliminated the option influences the outcome in this case, since if you chose a goat door (66% of the time), the host is FORCED to show you a goat, so 2/3rd of the time the other door is a car door 100% of the times (in your words, 2/3rd of the time a goat is 100% removed from the sample space and those 2/3rd of the time you are 100% sure you initially chose the goat door, so the car is 100% sure to be on the other door)
>>7985968
>influences
How? What kind of perturbation you emit when you say these words?: "Door A"
You are just playing two games: 3 doors and 2 doors.
Also, not resetting the probabilities when the sample space changes is an error Mathlab does too.
>>7985977
if you chose the goat the host is forced to reveal a specific door and not a generic one. if it were generic we could've resetted the probability without keeping into consideration the host elminiation, like you said, but that's not the case. let me say this again: choosing a goat always implies that the other door contains a car, since the host removes that specific second goat door. always.
if you got this, you only have to consider that you choose a goat 66% of the times, since choosing a goat automatically translates to the car being on the door you're asked to switch to
>>>>7985993
Anon, i already got the math part.
But it doesn't make sense.
What is a generic door?
>>7986019
I really didn't make a math argument, I just tried to use logic. do you struggle to get that you pick a goat 2/3rd of the times and that forcibly implies that the car will be on the other door you can switch to?
>What is a generic door?
here's what I meant:
the host can reveal either door B or C if you choose A only in the case the car is beyond door A. Since the other two doors are goats the host can open whatever door he wants, so he can open randomly and choose a generic door. If you chose a goat however he can't choose randomly (ie. choose a generic door) since he must choose the goat you didn't pick.
>>7986034
%A=1/3
%B=1/3
%C=1/3
Event: You lose one choice.
Logic:"Sample space changed, reset probabilities."
From: A=33% and B=33%
To: A=50% and B=50%
>There is a difference if the host know this
From the guest perspective, every door's result is incognito.
So, the probability shouldn't change.
It doesn't matter. I won't feel it intuitive unless I study statistics.
>>7986063
Nigga you only have a choice of 1/3 when you choose door 1. If door 3 is revealed, then it's still 1/3rd if you choose.
Basically it's the difference between choosing Door 1 and then Door 3 (whixh is wrong) or choosing Door 3 (which is wrong) and then Door 2.
>>7986075
Basically all its saying is that don't let your last guess be the door that is revealed, and start over choosing the door that is wrong.
>>7986063
>Logic:"Sample space changed, reset probabilities."
here's your flaw. you can do this only if the hosts randomly chooses which door he opens, because this way the event simply removes an outcome from the sample space. however if the host follows a logic in opening a door, and, in particular, a logic that ensures a car if you chose a goat, you must consider his behaviour and can't 'reset probabilities' whatever that means.
>rom the guest perspective, every door's result is incognito.
sure, but if the hosts assumes he chose a door with a goat and reasons about the host's behaviour, he can be 100% absolutely certain that the door he's proposed to switch to will contain a car.
I guess I'm repeating myself but this is the way that convinced me about the probability being 67%, so I can show you this perspective best. maybe try some though experiment to see if it feels more natural?
Imagine you have a huge amount of doors and the car is only behind one of those, the rest being empty. the door you picked randomly is almost surely empty. the host asks if you want to choose every other door. wouldn't you switch? this is the same thing, except that for some reason the host opens all the other doors before proposing the switch. the act of opening the door is irrilevant for what concerns the probability that your switch was convenient (for example imagine that the host opens the other doors after you switched. would that change the probability that your switch was convenient, even though both of you knew that every door but one was empty?)
Okay, but what if there are only two doors to begin with?
>>7986087
Then it's half. The question asks you what's the chance of you winning with 3 doors and one of them revealed if you never pick door 3, which would be 2/3rd. Remember, it's not about a 1/2 choice it's about a 1/3 that has been modified to a 2/3
>>7980853
The mathematical ignorance is astounding.
Only a knuckle dragging, slow speaking, inbred, mouth breathing, sister lusting, slope headed, MORON would not switch doors. Always switch doors!
>>7986098
:_(
>>7986085
Anon, I get it why does that 2/3 result happens, but the "reason" why don't they reset the probabilities, I don't get it.
I imply that every door has a 33% probability of having a winner. So If there are 10^6 games, every door would have a result of (10^6)/3.
When a guy try the game, the host choose one of the wrong doors; they are just OBSERVERS, they cam't affect the result. So the statistical results STILL give me a 33% each one.
>>7985923
Draw a card.
Have your friend take the rest.
Have your friend dump 50 of the cards that are not the ace of spades.
He's left with one card.
Who do you think got the ace?
>>7986242
>Anon, I get it why does that 2/3 result happens, but the "reason" why don't they reset the probabilities, I don't get it.
if you get the reasoning I spoke of earlier you should get why the probability isn't resetted. why should it, you lose valuable information on the criterium by which the host opened one door.
>I imply that every door has a 33% probability of having a winner. So If there are 10^6 games, every door would have a result of (10^6)/3.
this doesn't make sense, what do you mean with winner and result? what the fuck is a result that equals 333k?
>When a guy try the game, the host choose one of the wrong doors; they are just OBSERVERS, they cam't affect the result. So the statistical results STILL give me a 33% each one.
it keeps not making sense. maybe you misunderstood the example in the post you replied to? it was a lot of doors, not games, it's the same principle as >>7986250
>>7986242
It's the difference between choosing a possibly correct answer out of 3 and then seeing one answer you didn't choose is wrong or choosing a wrong answer out of 3 and then choosing again.
By choosing to start with the wrong answer, you're effectively choosing two out of three questions.
>>7980853
The monty hall problem is such, 1/3 chances of picking the right door.
I pick a door, so it is probably the wrong one... "I'm gonna open up this other door that doesn't have anything behind it! Do you want to switch?"
Yes.
>>7986253
>if you get the reasoning I spoke of earlier you should get why the probability isn't resetted
I omitted the obligation to reset the probabilities when I changed the sample space. So I got that result.
>100 doors chooice game
>computer programmed to randomly put the car behind one of the doors.
>repeat the game 10^6 times
>While the computer operate the orders, observers do shit and repeat wordplay.
>somebody is called Monty Hall or something; it doesn't matter because they aren't touching the doors.
>they don't move the prize; they are just observers; their actions don't affect the result the computer is deciding.
>10^6 contests done
>Prize in door A: (10^6)/3 times approximately
>Prize in door B: (10^6)/3 times approximately
>Prize in door C: (10^6)/3 times approximately
>>7986301
oops, they aren't 100 doors. They are just 3
>>7980853
Just to be absolutely clear, switching definitely gives you a 2/3 chance of winning.
Its difficult to understand. But you pick one then one of the other two gets removed but the one you picked had a 1/3rd chance yet the other two 2/3 combined. Knowing that the other got removed and it had to be fake that door is worth 2/3 not 1/2. They tested and resulted in actually being a 2/3rd chance with a randomizer.
Sooooooo..... yeh
>>7980916
>lucky
Think about what that means when you're talking about probability
>>7982548
Ran the sim got car 30 times goat 2 times
Fake sim because car stays at one spot on the left always resulting in there being 33% chance of picking the xar due to the randomizer
>>7980872
>you pick the right door
>the other two are sheep
>implying it matters which door is eliminated
>>7980853
Well I'll assume by abduction;
You first start with a 1/3 chance of choosing the right answer.
So your first choice has a 2/3 chance of being wrong as the prize could just as likely be in the other two doors
so you have [ 1/3 | 1/3 | 1/3 ]
which when you bring it down to two choices, considering that it was 2/3 likely not to be in the box initially
[ 1/3 | 1/3 | 1/3 ] can be seen as 1/3 in the chosen door, 2/3 chance the right answer is outside of it
also can be seen as
[ {1/3} | {2/3} ]
And here is the crux of the problem
Now that you know you have a 2/3 probability by leaving that initial choice, you'd have to choose one of the other two options.
one is invalidated..there is still that 2/3 probability by choosing another option
so now you have two options one with 1/3 probability and another with 2/3 because of the resampling
Monty Hall suggest you adopt the policy of treating this 2/3 as always wrong
/thread
>>7980853
If you just walk up to the two closed and one open door then 50/50. From the standpoint of the initial three closed doors its different. Probability is all about perspective
>everything has a 50/50 chance of happening: it either does or it doesn't
>t. stephen
>>7986447
>pick the right door, the other two are sheep, switching loses
>pick wrong door 1, the other sheep is revealed, switching wins
>pick wrong door 2, the other sheep is revealed, switching wins
2/3 chance of winning baka desu senpai
>>7985494
>tfw you'll never be this autismo over a stats problem you dont understand
>>7986777
great
Now i can't understand why this problem is so famous.
If only it was properly explained instead of being turned into a media shitfest then maybe no one would care about it.
Thank you kind anon.
