Struggling with how to approach this question. I'm trying to use casework, but it's devolving into a really messy problem very quickly.
You have 6 colored balls of 3 colors, with 2 of each color. For example, 2 red, 2 green, and 2 blue balls. You're drawing for 3 pairs of balls, without replacement. What's the probability that you don't draw a RG/GR pair. For example, a drawing could be as follows:
- first draw a red, then draw a second red (RR)
- (from the remaining 4 balls) draw a green, then draw a blue
- (from the remaining 2 balls) draw a green, and then a blue
Your pairs would be: RR, GB, GB
What's the probability that one of the pairs isn't composed of a green and red ball?
Without loss of generality, let the balls be uniquely numbered as $R_1,R_2,B_1,B_2,G_1,G_2$. This lets us organize our thoughts a bit more freely.
Direct: We know that $R_1$ will need to be paired with something. What is the chance that it gets paired with the other red? What is the chance it gets paired with a blue? If it was the case that $R_1$ got paired with a blue, given that information what is the chance then that $R_2$ gets paired with a blue and not a green?
This line of thought gives us an overall answer of $\frac{1}{5}+\frac{2}{5}\cdot\frac{1}{3} = \frac{1}{3}$
Indirect: Let's consider the opposite... what is the probability that we did get at least one RG pair?
What is the chance that $R_1$ gets paired with a green? What is the chance that $R_2$ gets paired with a green? What is the chance that both of these happened simultaneously?
Combining this information by inclusion-exclusion and subtracting away from $1$ gives us the final answer as:
$1-\left(\frac{2}{5}+\frac{2}{5}-\frac{2}{5}\cdot\frac{1}{3}\right) = \frac{1}{3}$