My friend posed this problem to me. He wrote a computer program to generate all possible combinations of dice and counted the cases with pairwise $7$'s. I thought that the problem had a simpler mathematical answer. My theory was:
- There are $6$ Possible combinations among four dice (A,B,C,D).
- Each pair of dice has a probability of $5/6$ of NOT totaling $7$.
- The probability that none of the Six pairs will total $7$ is $(5/6)^6 \approx 0.334897977$.
- The probability of rolling one or more pairs of $7$'s is the complement or =$0.665102023$.
The problem is that we had a discrepancy of $.01$ in our calculations. I don't really feel like debugging his computer program. Could you please comment on my mathematical approach to this?
thanks, Bob K
the probabilities in step (3) are not independent. So probability of the compound event is not the product of the simple events' probabilities...