Really this is really a statement about maximizing the expected benefit of success minus the expected cost of failure. This delineates the cases where we would

Really this is really a statement about maximizing the expected benefit of success minus the expected cost of failure. This delineates the cases where we would like to follow this rule vs the cases where we do not want to. If the payoff is convex, like in "if you've never failed, you don't take enough risks" then the advice is wise. If the payoff is concave like "if you've never broken every bone in your body you haven't gone bungee jumping enough" then it is less so.