To construct a useful ethical question within the realm of the hypothetic it is first necessary that both options are undesirable. These questions are useful only when they force a confrontation between actions which have both good and bad results. The listener is then forced to examine their intentions as the consequences of each side is examined.
Second, these questions need to have a bit of plausibility.
"Would you have dropped the bomb?" is an example of a useful question.
Here is an example of a pointless question:
"Would you drink the spell of a witch if that gave you powers to destroy all evil, except every third evil person was actually good, and it was the potion which made you wrong, but anyway isn't it cool to destroy evil?