The almond experiment sounds ridiculous! Asking people to imagine eating ten chocololate-covered almonds then to guess how much they’d like to eat another after

The almond experiment sounds ridiculous! Asking people to imagine eating ten chocololate-covered almonds then to guess how much they’d like to eat another after their hypothetical consumption? Does this produce results as reliable as actually eating the almonds in the first place?