it's a complement problem. "At least one match" is the classic signal: computing the many ways to succeed is a nightmare; computing the single way to fail is easy. Complement is the whole engine here.
Key idea: P(at least one match) = 1 − P(no match at all). The engine is that the "no match" event is a single clean counting task, while "at least one" fans out into cases.
Set up the deck state.
Total deck: 8 different cards × 3 copies = 24 cards.
3 non-matching cards have been dealt — three different card types, one copy each.
Remaining in deck: 24 − 3 = 21 cards.
Of those 21, split them by what a match would mean. The 3 dealt types have 2 copies each still in the deck:
- Matching cards (would pair with a dealt card): 3 types × 2 copies = 6
- Non-matching cards (the other 5 types, untouched): 5 × 3 = 15
Check: 6 + 15 = 21
Compute the complement — draw 2 more, neither is a match.
"No match" means both new cards come from the 15 non-matching cards.
$$P(\text{no match}) = \frac{\binom{15}{2}}{\binom{21}{2}} = \frac{105}{210} = \frac{1}{2}$$
Flip it:
$$P(\text{at least one match}) = 1 - \frac{1}{2} = \frac{1}{2}$$
Answer: (E) 1/2
Note on the "3 of the same card" clause: it's a red herring — drawing 2 copies of a dealt type gives you 3 of a kind, but that's already inside the "at least one match" event. The complement handles every success case at once without enumerating them, which is exactly why complement is the right tool.
The only move that mattered: turn "at least one" into "one minus none." Once the deck was split into 6 matching / 15 non-matching, the answer was two combinations away.