Go Back   EcoModder Forum > Off-Topic > The Lounge
Register Now
 Register Now
 


Reply  Post New Thread
 
Submit Tools LinkBack Thread Tools
Old 02-23-2015, 08:15 PM   #21 (permalink)
Growin a stash
 
Join Date: Jun 2014
Location: Austin TX
Posts: 803
Thanks: 412
Thanked 304 Times in 228 Posts
Quote:
Originally Posted by Old Mechanic View Post
Irrelevant, the host is always going to choose the goat.
Agreed.

Think of the rest this way: you choose door A (33% likelihood). The host shows a goat in door C. Does seeing the goat in door C give you any motivation to change your answer? Does it have any bearing at all on doors A&B?

No it doesn't. It gave no new information. It's like he's revealing the answer in slow motion-- not helping you at all.

__________________


2024 Chevy Bolt

Previous:
2015 Nissan Leaf S, 164 mpge

Last edited by ME_Andy; 02-23-2015 at 09:15 PM..
  Reply With Quote
Alt Today
Popular topics

Other popular topics in this forum...

   
Old 02-23-2015, 09:13 PM   #22 (permalink)
Master EcoModder
 
freebeard's Avatar
 
Join Date: Aug 2012
Location: northwest of normal
Posts: 27,692
Thanks: 7,774
Thanked 8,584 Times in 7,068 Posts
Quote:
You wrote that vos Savant was hot and there was an ad for Asian dating. I forgot, but remembered when it showed me a Philippine dating site.
???? For me link goes to "Marilyn vos Savant Biography, Marilyn vos Savant's Famous Quotes - QuotesSays . COM" like it says on the tin. The picture's link I think I got from the Google search page because the 'CLICK FOR NORMAL VIEW' wasn't working. YMMV.

What are the odds the car was a Pontiac GTO?
  Reply With Quote
Old 02-23-2015, 11:03 PM   #23 (permalink)
Master EcoModder
 
Join Date: Aug 2010
Location: Philippines
Posts: 2,173
Thanks: 1,739
Thanked 589 Times in 401 Posts
Quote:
Originally Posted by ME_Andy View Post
Yep, the key to all of this (as I understand it) is that the host intentionally makes his choice in a way that doesn't give you any more info.
Wrong. The host always gives more info.

The chance of the host showing A goat is 100%. This is the format of the gameshow.

Once the host shows that goat, he collapses the probabilities, making it much easier to win the game. Just because the other possibilities no longer exist, doesn't mean you shouldn't count them in assigning percentages!

Simplest way to put it:

If Car = Your Door (1/3rd chance), host can open either door. Changing mind = 0% wins.



If Car =/= Your Door (2/3rd chance), host will never open the winning door. The remaining door is the Car. Changing mind = 100% win


100% of 2/3rd is 66% chance of winning, overall.


It's that simple.

And statistical testing backs it up.

If you want long, boring tables, though:

Code:
Car	You	Host	You		Win %, Changed mind
A	A	xxx	xxx	xxx	0.00%
A	A	xxx	xxx	xxx	
A	A	xxx	xxx	xxx	
A	A	B	A	win	
A	A	B	xxx	xxx	
A	A	B	C	lose	
A	A	C	A	win	
A	A	C	B	lose	
A	A	C	xxx	xxx	

B	A	xxx	xxx	xxx	100.00%
B	A	xxx	xxx	xxx	
B	A	xxx	xxx	xxx	
B	A	xxx	xxx	xxx	
B	A	xxx	xxx	xxx	
B	A	xxx	xxx	xxx	
B	A	C	A	lose	
B	A	C	B	win	
B	A	xxx	xxx	xxx	

C	A	xxx	xxx	xxx	100.00%
C	A	xxx	xxx	xxx	
C	A	xxx	xxx	xxx	
C	A	B	A	lose	
C	A	B	xxx	xxx	
C	A	B	C	win	
C	A	xxx	xxx	xxx	
C	A	xxx	xxx	xxx	
C	A	xxx	xxx	xxx	
				Chances	66%
  Reply With Quote
The Following User Says Thank You to niky For This Useful Post:
RedDevil (02-24-2015)
Old 02-24-2015, 03:05 AM   #24 (permalink)
Master EcoWalker
 
RedDevil's Avatar
 
Join Date: Dec 2012
Location: Nieuwegein, the Netherlands
Posts: 3,998

Red Devil - '11 Honda Insight Elegance
Team Honda
90 day: 49.01 mpg (US)
Thanks: 1,711
Thanked 2,245 Times in 1,454 Posts
The thing that makes it hard to get right is that it looks like a lottery.
I won a car in a lottery; the Hyundai I10 in my garage, actually. I know how it feels.

But the contest is not a lottery.
The host can pick a door with a goat because there will always be one door with a goat left after your first choice.

I can break it down as simple as this:
The chance that your first choice is right is 1/3.
The chance that the car is behind one of the other two doors, of which one will be opened to reveal a goat, is 2/3.

The two doors you did not choose still have a combined 2/3 chance of having the car behind them. But when you know which of those two does not have the car, you know the other has the full 2/3 chance.

Got it? No?
Let's change the experiment:

Running the contest with 1.000.000 doors
1 car, 1 million doors.
We send the goats on leave this time.

You pick one door at random. You have 0.0001 % chance of picking the car.
There is a 99.9999 % chance that the car is behind one of the other doors.
There are (at least) 999.998 remaining doors that do not have a car.
Now it comes:
The host opens 999.998 doors, carefully skipping the one door hiding the car (unless you've had the uncanny 0.0001% chance chance of picking the right one at first, in which case he skips a random door)

Will you swap or not?

You should.
Going to high numbers clearly shows how selective the host operates.
With every door he opens he does not increase the chance that your original choice was right, though it may feel that way.

Because unlike a lottery, he does not pick doors at random.
He opens only doors that have no car.

Running a lottery by contrast
If it were a lottery, initially youŽd have a 0.0001 % chance that a door reveals a car for each door, just like the contest.

If after half the doors were opened still no car was revealed, then youŽd have a 0.00002% chance that any remaining door would have a car.
But you also would have a 50% chance of being out of competition already.

After 999.998 doors and no car shown the remaining two doors, the one youŽve originally chosen and the remaining one, each have a 50% chance of hiding the elusive car.
But youŽd only have a 0.0002% chance of getting that far. Over the whole the chance that those particular two doors have the car is still just 0.0001% each.

In the game there is no element of chance after your initial choice.
You have 100% chance to make it to the last 2 doors because the host will not pick the door with the car.
Nothing alters the chance that you did pick right first time.
__________________
2011 Honda Insight + HID, LEDs, tiny PV panel, extra brake pad return springs, neutral wheel alignment, 44/42 PSI (air), PHEV light (inop), tightened wheel nut.
lifetime FE over 0.2 Gmeter or 0.13 Mmile.


For confirmation go to people just like you.
For education go to people unlike yourself.

Last edited by RedDevil; 02-24-2015 at 09:42 AM..
  Reply With Quote
The Following 2 Users Say Thank You to RedDevil For This Useful Post:
Gasoline Fumes (02-24-2015), niky (02-25-2015)
Old 02-24-2015, 03:36 AM   #25 (permalink)
Master EcoWalker
 
RedDevil's Avatar
 
Join Date: Dec 2012
Location: Nieuwegein, the Netherlands
Posts: 3,998

Red Devil - '11 Honda Insight Elegance
Team Honda
90 day: 49.01 mpg (US)
Thanks: 1,711
Thanked 2,245 Times in 1,454 Posts
Quote:
Originally Posted by freebeard View Post
When I had my IQ tested in high school, they wouldn't tell me the result, because they didn't "want [me] to get a swelled head". Well, they got their way about that.
I've done several IQ tests over the years, and the funny thing is that I score higher with each test.
Apparently, solving IQ tests is a skill, and I get better at it with practise.
I don't think I really got smarter.
But what do I know
__________________
2011 Honda Insight + HID, LEDs, tiny PV panel, extra brake pad return springs, neutral wheel alignment, 44/42 PSI (air), PHEV light (inop), tightened wheel nut.
lifetime FE over 0.2 Gmeter or 0.13 Mmile.


For confirmation go to people just like you.
For education go to people unlike yourself.
  Reply With Quote
Old 02-24-2015, 12:36 PM   #26 (permalink)
Master EcoModder
 
Join Date: Jun 2008
Location: Earth
Posts: 5,209
Thanks: 225
Thanked 811 Times in 594 Posts
Quote:
Originally Posted by Old Tele man View Post
Uh, spelling wasn't part of those IQ tests, was it (wink,wink)?
Or perhaps he's using the British spelling? https://en.wiktionary.org/wiki/practise I have a bit of trouble with that myself, having read a lot of Agatha Christie &c in my childhood.
  Reply With Quote
The Following User Says Thank You to jamesqf For This Useful Post:
RedDevil (02-24-2015)
Old 02-24-2015, 02:49 PM   #27 (permalink)
one of thOOOse people
 
Join Date: Mar 2009
Location: the cloud(s)
Posts: 293

twitchy - '98 honda civic dx + sir + ls
90 day: 30.2 mpg (US)

Norman - '14 Ford E-350 EXT
90 day: 16.18 mpg (US)

Silver - '12 VW Golf Base
90 day: 26.1 mpg (US)

Sparta - '19 Honda CB300R
90 day: 84.07 mpg (US)
Thanks: 0
Thanked 81 Times in 66 Posts
Quote:
Originally Posted by Old Tele man View Post
Uh, spelling wasn't part of those IQ tests, was it (wink,wink)?

Just blame it on the MS "auto-spell" function -- like I do (ha,ha)!
He may be using Firefox which defaults to english(UK) for the built-in spell checker. Gets me all the time.
  Reply With Quote
Old 02-24-2015, 02:59 PM   #28 (permalink)
Too many cars
 
Gasoline Fumes's Avatar
 
Join Date: Oct 2009
Location: New York State
Posts: 1,573

CRXFi - '88 Honda CRX XFi

Insight 256 - '00 Honda Insight
Team Honda
Gen-1 Insights

Insight 5342 (no IMA) - '00 Honda Insight
Team Honda
Gen-1 Insights
90 day: 66.3 mpg (US)
Thanks: 1,329
Thanked 786 Times in 468 Posts
Quote:
Originally Posted by RedDevil View Post
I can break it down as simple as this:
The chance that your first choice is right is 1/3.
The chance that the car is behind one of the other two doors, of which one will be opened to reveal a goat, is 2/3.

The two doors you did not choose still have a combined 2/3 chance of having the car behind them. But when you know which of those two does not have the car, you know the other has the full 2/3 chance.

Got it? No?
Thanks! Now I finally get it! Did I win a car?
__________________
2000 Honda Insight
2000 Honda Insight
2000 Honda Insight
2006 Honda Insight (parts car)
1988 Honda CRXFi
1994 Geo Metro

  Reply With Quote
Old 02-24-2015, 04:35 PM   #29 (permalink)
Master EcoWalker
 
RedDevil's Avatar
 
Join Date: Dec 2012
Location: Nieuwegein, the Netherlands
Posts: 3,998

Red Devil - '11 Honda Insight Elegance
Team Honda
90 day: 49.01 mpg (US)
Thanks: 1,711
Thanked 2,245 Times in 1,454 Posts
Quote:
Originally Posted by Gasoline Fumes View Post
Thanks! Now I finally get it! Did I win a car?
What you've won is insight in how first impressions can deceive you.

Kudos to my dad.
I explained the problem to him tonight and before I could finish the sentence about the remaining door he said it had a 2/3 chance of having the car behind it. It is bone trivial to him!
Should I have told him he really needs to think it over?
__________________
2011 Honda Insight + HID, LEDs, tiny PV panel, extra brake pad return springs, neutral wheel alignment, 44/42 PSI (air), PHEV light (inop), tightened wheel nut.
lifetime FE over 0.2 Gmeter or 0.13 Mmile.


For confirmation go to people just like you.
For education go to people unlike yourself.
  Reply With Quote
Old 02-24-2015, 04:41 PM   #30 (permalink)
Master EcoWalker
 
RedDevil's Avatar
 
Join Date: Dec 2012
Location: Nieuwegein, the Netherlands
Posts: 3,998

Red Devil - '11 Honda Insight Elegance
Team Honda
90 day: 49.01 mpg (US)
Thanks: 1,711
Thanked 2,245 Times in 1,454 Posts
Quote:
Originally Posted by jamesqf View Post
Or perhaps he's using the British spelling? https://en.wiktionary.org/wiki/practise I have a bit of trouble with that myself, having read a lot of Agatha Christie &c in my childhood.
I use double-dutch spelling, preferredly...
Yarrrr matey, ye need shoa me haw yer spellin is in Dutch and yer betta do good or ya walk tha plank

Spelling is not a part of IQ tests.
If you end up in Diagon Alley there is still hope.

__________________
2011 Honda Insight + HID, LEDs, tiny PV panel, extra brake pad return springs, neutral wheel alignment, 44/42 PSI (air), PHEV light (inop), tightened wheel nut.
lifetime FE over 0.2 Gmeter or 0.13 Mmile.


For confirmation go to people just like you.
For education go to people unlike yourself.
  Reply With Quote
Reply  Post New Thread






Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Content Relevant URLs by vBSEO 3.5.2
All content copyright EcoModder.com