Gustav, --- In [email protected], Gustav Fredell <[EMAIL PROTECTED]> wrote: > > So you say it's a 100% that I win if I stay... awesome conclusion. Just > awesome...
Excuse me but i did not say that. > You can't drag the law of large numbers into this the way you are doing. > It has nothing to do with odds only being true for many trials or not. > It merely states that if you do something enough times it will converge > towards the "true" odds. Which in this case is 1/3 and 2/3. You just said that I can't can't put in the realm of the law of large numberfs, and then you immediately defend its position within the realm of large numbers. You can't have it both ways - either it applies or it doesn't. The number of games is part of where you are going wrong. There is one trial is the classic example - what effect does it have on one game? None. In the same way that you cannot predict when a radiactive atom will decay. > So what it > in fact can be used for is to make the choices many times (like others > suggested), ad them up and calculate odds that come nearer and nearer to > the real odds. In this way you can "prove" that the odds are in fact 1/3 > and 2/3 if you do it enough times. I challenge you to repeat Bob's 100 > trials and and test your hypothesis against a 95% confidence interval or > so... I wrote a program in Basic on an Atati 800. Making a choice between 1, 2, and 3 generated at random, I decided never to switch. Out of 100 "games" I "won the car" 76 times. > With a solid year or so at university in probability and statistics and > very confident in this. Ask all the probability/statistics professors in > the world if the odds are 1/3 and 2/3, and I strongly believe that the > law of large numbers will make the will make this converge towards 100% > agreeing. In fact I challenge you to ask a single probability/statistics > professor about this and let us all know about his/her answer. Why should I answer a challenge which is dependent upon how well another person thinks? > Take on the one or both of the challenges above and then come back and > say it's not true. I bet you my upcoming 2x2x2 average WR that you can't > do it ;) I'm sorry you lost the bet. You probably had your heart set on the 2x2x2 average WR. Regards, David J > /Gustav > > d_j_salvia skrev: > > >Hi Duncan and Stefan and Pedro and Evan, > > > >Sorry, you guys, but you aren't correct. > > > >In probability there's a thing called the law of large numbers. If you > >generate a long enough string of numbers "randomly" that eventually > >you would have every digit the same number of times. N oparticular > >number is favored. Your answer relies upon this law of randomness. > > > >What is actually wrong with the standard answere you gave is that one > >is not dealing with a large number of choices, and, as such, odds *do > >not apply.* > > > >I went to a site with the software and made my choice and did not > >switch and I won. Doing it more than once is outside the boundaries of > >the game. > > > >Cheers, > > > >David J > > > > > >--- In [email protected], "Duncan Dicks" > ><[EMAIL PROTECTED]> wrote: > > > > > >>I Had a freind who wouldnt believe this no matter how often I > >> > >> > >explained the > > > > > >>maths to him so he set up spreadsheet to test it out. Very easy to > >> > >> > >do and > > > > > >>confirmed what the maths tells you - you should switch! > >> > >>Duncan > >>----- Original Message ----- > >>From: "aznseashell" <[EMAIL PROTECTED]> > >>To: <[email protected]> > >>Sent: Sunday, December 25, 2005 7:04 AM > >>Subject: [Speed cubing group] Re: (Off topic)3 doors... > >> > >> > >>Haven't we had this dicussion before? Or was it in another cubing group? > >> > >>The game is set up so that if you switch, a winning choice would > >>become a losing choice and vice versa. In the beginning you had a 1/3 > >>chance of winning and 2/3 chance of losing. Staying with your choice > >>doesn't change your odds of winning (the host will always be able to > >>show you a door with nothing behind it no matter which door you > >>picked), but switching will turn your probabability of winning to 2/3. > >> > >>If my explanation makes no sense, consider the situation with 100 > >>doors and one door with a prize. You pick a door, and the host shows > >>you 98 doors with nothing behind them. Now it's much more obvious that > >>you should switch, right? > >> > >>Shelley > >> > >> > >>--- In [email protected], "richy_jr_2000" > >><[EMAIL PROTECTED]> wrote: > >> > >> > >>>It is counter intuitive, but if you are in this situation, your > >>>chances would be better to switch to the other door. It's actually > >>>quite interesting. > >>> > >>>-Richard > >>> > >>>--- In [email protected], Pedro > >>><[EMAIL PROTECTED]> wrote: > >>> > >>> > >>>>Ok, this is off topic, but is interesting... > >>>> > >>>> (please forgive if I make some mistake on the english...) > >>>> Suppose you are at a TV show, where you have 3 doors. 1 of the > >>>> > >>>> > >>>doors has a car, and the other 2 don't have anything. So, the show > >>>presenter asks you to choose a door. So, you choose, but he doesn't > >>>open your choosen door. He opens an empty door. Then, he makes a > >>>question: do you want to continue with your first choice or change to > >>>the other door? > >>> > >>> > >>>> What do you do? > >>>> What situation gives you more chances of winning? > >>>> > >>>> Think about it... > >>>> > >>>> Pedro > >>>> > >>>> > >>>>--------------------------------- > >>>> Yahoo! doce lar. Faça do Yahoo! sua homepage. > >>>> > >>>>[Non-text portions of this message have been removed] > >>>> > >>>> > >>>> > >> > >> > >> > >> > >> > >> > >>Yahoo! Groups Links > >> > >> > >> > > > > > > > > > > > > > > > > > > > >Yahoo! Groups Links > > > > > > > > > > > > > > > >__________ NOD32 1.1340 (20051226) Information __________ > > > >This message was checked by NOD32 antivirus system. > >http://www.nod32.com > > > > > > > > > > > ------------------------ Yahoo! Groups Sponsor --------------------~--> Get fast access to your favorite Yahoo! Groups. Make Yahoo! your home page http://us.click.yahoo.com/dpRU5A/wUILAA/yQLSAA/MXMplB/TM --------------------------------------------------------------------~-> Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/speedsolvingrubikscube/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
