It's not always a fallacy. Examples: 1. You're trying to achieve some objective, and the difference between achieving it and not achieving it swamps all other differences between credible outcomes. It may then be rational to assume that your desired objective is achievable. (You have nasty symptoms, which can be caused by two diseases. One will kill you in a week whatever you do. One is treatable. If it's at all difficult to distinguish the two, you might as well assume you've got the treatable one.) 2. You're trying to achieve some objective, and you know it's achievable because others have achieved it, or because the situation you're in has been crafted to make it so. It's rational to assume it's achievable. (There's an example in J E Littlewood's "Mathematician's Miscellany": he was climbing a mountain, he got to a certain point and couldn't see any way to make progress, and he reasoned thus: I know this is possible, and I know I've come the right way so far, so there must be a hidden hold somewhere around there ... and, indeed, there was.)