doors are labeled 1,2,3
We have 3^2 combinations namely
(1,1),(1,2),(1,3),(2,1),(2,2),(2,3),(3,1),(3,2),(3,3)
If the boxes were pre-programmed and say the program is (blue, red , blue ) for doors (1,2,3)
then we have :
(blue,blue),(blue,red),(blue,blue),(red,blue),(red,red),(red,blue),(blue,blue),(blue,red),(blue,blue) 6/9=2/3
If the program were (Red,red blue) the ratio is unchanged , but the colour combinations will be different
But if they are not programmed , combinations (1,1),(2,2),(3,3) give the same colour but they would have no idea about; would have happened had they opened different doors , say the combination 1,2 which gives (blue,blue) in the first program , but the same combination give (blue red ) in the second program .
That implies the two(or more ) programs superimposed in these non-symmetric combinations. The the probability for Same colour, in those combinations, cancel the probability for different colours , leaving us with 3/9=1/3 certainity about colour similarity namely in combinations that involved opening the same door . Which means that the probability of combination (1,2),(1,3),(2,1),(2,3),(3,1),(3,2) is undecided 50/50 %
So we have 2 different doors A,B , and two colours Red and blue , 2^2=4
Red Red
Blue blue
Red blue
Blue Red X 6 combinations = 24
the 6 combinations expand into 24 colour combinations . 12 combinations show the same colour , and 12 show different colours , and they cancel to Zero leaving us with 1/3 certainity about colour similarity + 0.5 probabillity for colour similarity
It’s probable that the experiment would yield each of the 6 combinations in one colour red/blue for 6 boxes. But as the number of trials increase , opposite probabilities cancel , and limit to zero .