Started By
Message

Marilyn vos Savant and the history of the Montel Hall question

Posted on 2/22/15 at 8:57 pm
Posted by Big Scrub TX
Member since Dec 2013
33446 posts
Posted on 2/22/15 at 8:57 pm
quote:

When vos Savant politely responded to a reader’s inquiry on the Monty Hall Problem, a then-relatively-unknown probability puzzle, she never could’ve imagined what would unfold: though her answer was correct, she received over 10,000 letters, many from noted scholars and Ph.Ds, informing her that she was a hare-brained idiot.

What ensued for vos Savant was a nightmarish journey, rife with name-calling, gender-based assumptions, and academic persecution.




LINK
Posted by Walking the Earth
Member since Feb 2013
17260 posts
Posted on 2/22/15 at 9:29 pm to
I vaguely remember the controversy. The Monty Hall Problem is pretty counter-intuitive and, in fairness (it was touched upon in the article) in "real life" it works a little different than it was presented in the original hypo.

Still, a resounding victory for the genius and proof that the Internet didn't invent the shrill response crowd, it just perfected it.
Posted by Kafka
I am the moral conscience of TD
Member since Jul 2007
142047 posts
Posted on 2/22/15 at 10:09 pm to
quote:

Montel Hall
I remember Montel Hall. He hosted this weird game show where if you wanted to be picked as a contestant you had to dress up as an interracial transsexual unwed mother
Posted by HVAU
Far, far away
Member since Sep 2010
4589 posts
Posted on 2/22/15 at 10:23 pm to
Enjoyed the read.
Posted by Big Scrub TX
Member since Dec 2013
33446 posts
Posted on 2/23/15 at 11:43 am to
quote:

proof that the Internet didn't invent the shrill response crowd, it just perfected it




Good point.
Posted by VetteGuy
Member since Feb 2008
28193 posts
Posted on 2/23/15 at 11:47 am to
quote:

Marilyn vos Savant


She was pretty good-looking back in the day.
I've always had a thing for really smart chicks.





Dumb ones too, come to think of it.
Posted by fightin tigers
Downtown Prairieville
Member since Mar 2008
73681 posts
Posted on 2/23/15 at 11:52 am to
quote:

in "real life" it works a little different than it was presented in the original hypo.


The crux of this argument is that a door containing nothing will always be removed first and the person opening the doors know which is which.
Posted by USMCTiger03
Member since Sep 2007
71176 posts
Posted on 2/23/15 at 12:09 pm to
Doesn't it matter more whether the opened door was chosen at random or picked specifically because it did NOT have the car? This issue confuses me even after reading the details.
Posted by fightin tigers
Downtown Prairieville
Member since Mar 2008
73681 posts
Posted on 2/23/15 at 12:17 pm to
quote:

Doesn't it matter more whether the opened door was chosen at random or picked specifically because it did NOT have the car?


Yes, there is no randomness factored in.
Posted by CockHolliday
Columbia, SC
Member since Dec 2012
4517 posts
Posted on 2/23/15 at 12:56 pm to
Assuming the game host would never open a door containing the car, there are 12 possible scenarios here (4 separate ones for each door containing the car):

If Door 1 contains the car:

1) Guest picks Door 1, host opens Door 2. If guest switches, lose. If guest stays, win.

2) Guest picks Door 1, host opens Door 3. Switch: lose. Stay: win.

3) Guest picks Door 2, host opens Door 3. Switch: win. Stay: lose.

4) Guest picks Door 3, host opens Door 2. Switch: win. Stay: lose.

Theoretically there would be 2 more scenarios for Door 1 containing the car, but those would involve opening Door 1 which we've already ruled out for purposes of the contest. So far, switching is 50/50.

If Door 2 contains the car:

1) Guest picks Door 1, host opens Door 3. Switch: win. Stay: lose.

2) Guest picks Door 2, host opens Door 1. Switch: lose. Stay: win.

3) Guest picks Door 2, host opens Door 3. Switch: lose. Stay: win.

4) Guest picks Door 3, host opens Door 1. Switch: win. Stay: lose.

Still 50/50 by switching vs. staying.

If Door 3 contains the car:

1) Guest picks Door 1, host opens Door 2. Switch: win. Stay: lose.

2) Guest picks Door 2, host opens Door 1. Switch: win. Stay: lose.

3) Guest picks Door 3, host opens Door 1. Switch: lose. Stay: win.

4) Guest picks Door 3, host opens Door 2. Switch: lose. Stay: win.

Once again, 50/50.

What am I missing here?

Posted by Willie Stroker
Member since Sep 2008
12904 posts
Posted on 2/23/15 at 1:16 pm to
I don’t follow her logic. We’re ultimately presented with information that a prize is behind one of two doors. Why should it matter which one is called door number 1, 2, or 3? Why should it matter that the original question was posed as if the odds were 1 in 3?

Let’s work it backwards. There are two doors. A prize is behind one. The odds of correctly guessing which door the prize is behind is 50/50, right? You choose one. I ask you if you want to change your mind. If you do and choose the other door, did the odds of correctly guessing suddenly change? Or is it still 50/50 odds? If so, explain. What about if I suddenly say, “Would you change your mind if I told you there was another door that did NOT have a prize behind it, would you change your mind?” Would that additional question and fact suddenly change the odds to 1 out of 3? What if I said there were 100 additional doors that a prize was NOT behind? Does that make the original question anything different than a 50/50 chance?

I still don’t understand how her answer is regarded as correct.
Posted by WG_Dawg
Hoover
Member since Jun 2004
86501 posts
Posted on 2/23/15 at 1:23 pm to
quote:

I still don’t understand how her answer is regarded as correct.


I'm with you since I"m not a math/probability guy AT ALL, but how someone explained it to me was this.

At the start when you pick a door, you have a 33% chance of guessing correctly. After 1 door is revealed to be nothing, now your chances of picking the correct door are 50/50 between car and goat, so you're "supposed" to switch. I don't really get it, but that's how it was told to me.
Posted by studentsect
Member since Jan 2004
2259 posts
Posted on 2/23/15 at 1:33 pm to
quote:

I'm with you since I"m not a math/probability guy AT ALL, but how someone explained it to me was this.

At the start when you pick a door, you have a 33% chance of guessing correctly. After 1 door is revealed to be nothing, now your chances of picking the correct door are 50/50 between car and goat, so you're "supposed" to switch. I don't really get it, but that's how it was told to me.



Look at it this way:

There are 3 doors. You pick door 1. The host says, "You can either keep door 1 or have BOTH of the the other doors."

You know for sure that at least one of these other two doors have a goat behind it, but you will still make the switch because two doors are better than one.

The actual problem follows the same logic.
You pick a door, and then are given the option of keeping your door or switching to the other 2 doors. The only difference is that this time they show you that one of the other two doors has a goat behind it...but you already knew with certainty that one of these two doors had a goat behind it, so him showing it to you doesn't change anything.
Posted by Monk
Baton Rouge
Member since Aug 2007
3660 posts
Posted on 2/23/15 at 1:57 pm to
I thought you were going to reference the time that she stated that there is a 50/50 chance of flipping heads even after 9 heads in a row.

I recall that caused a stir.
Posted by WG_Dawg
Hoover
Member since Jun 2004
86501 posts
Posted on 2/23/15 at 2:23 pm to
quote:

I thought you were going to reference the time that she stated that there is a 50/50 chance of flipping heads even after 9 heads in a row.

I recall that caused a stir.



Are you saying you disagree with that?
Posted by Tommy Patel
Member since Apr 2006
7558 posts
Posted on 2/23/15 at 2:25 pm to
I learned something from TD.com today.
Posted by nes2010
Member since Jun 2014
6763 posts
Posted on 2/23/15 at 2:30 pm to
quote:

You are the goat!
Glenn Calkins
Western State College


Posted by Monk
Baton Rouge
Member since Aug 2007
3660 posts
Posted on 2/23/15 at 2:52 pm to
quote:

What am I missing here?


I'm guessing it has to do with the fact that you are considering 4 options instead of focusing on only 3 doors/options.

If you pick the right door initially, the host has 2 options on doors to open. As you basically noted, if you pick wrong initially, the host only has 1 door to open.

In your test, you are allowing the contestant to pick the correct door twice. In reality, there are 3 doors and he gets 1 initial pick, regardless which of the 2 goat doors the host decides to open next. So 1 and 2 in your test is really 1 OR 2 so ONE win, not two.
Posted by Monk
Baton Rouge
Member since Aug 2007
3660 posts
Posted on 2/23/15 at 2:54 pm to
quote:

I thought you were going to reference the time that she stated that there is a 50/50 chance of flipping heads even after 9 heads in a row.

I recall that caused a stir.
Posted by Ace Midnight
Between sanity and madness
Member since Dec 2006
89552 posts
Posted on 2/23/15 at 2:57 pm to
This actually a fairly smart question for fairly smart people.

The answer is counterintuitive - but it is correct from a probability standpoint.

However, it is the nature of the 1/3 and of the laws of probability itself.

The original choice is a 1 in 3. The fact that it eliminates one of the 2 bad choices inherently increases your odds - you cannot switch from a bad answer to a bad answer but only from bad to good or good to bad.

All of the failures occur on the initial choice (picking the right one) - all of the successes occur on the switches (switching from the bad one, with the other bad one eliminated).

Weird, but it works.

(Because they never show you the good choice if you pick one of the bad - that's the other factor that makes this work.)
Page 1 2 3 4 5 6 7
Jump to page
first pageprev pagePage 1 of 7Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on Twitter, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookTwitterInstagram