>Rick,
>
>This is not correct and it has been proven with empirical testing. The issue of confusion is the removal of one door. At first you have a 2/3 chance of selecting the wrong door, odds are you will be wrong. After one of the wrong doors is opened then there are two doors left, the one you chose with 2/3 chance of being wrong and the other with a 1/2 chance of being right. If multiply the two probilities you get 1/3 chance of the other door being correct or wrong while there was still a 2/3 chance that the first choice was wrong and only a 1/3 chnace it was right.
>
>The mathematics say that over an extended series of tests the other door will more often be the correct one.
>
>The first time I heard this puzzle I was a non-believer too. Then I was shown the math and the emperical tests that proved the math was right.

Looks like I was wrong. I hate it when that happens. I tweaked Christof's code so I could run it both ways, and in fact the empirical results do show about one win in three if you don't change versus two wins in three if you do change. So, now I'm a believer too... but it's still counter-intuitive.

BTW, do you remember Cheech and Chong's take-off on Let's Make a Deal?