I actually don't think this is true. As I understand it, a given routine will have a number of high-risk tricks that may work, say, only 1-in-3 times. If they do work, they're highly effective, precisely because they're so implausible, and if they don't work you can just pretend it's a bluff and move onto plan B. It's all about effective showmanship and managing expectations.
(Disclaimer: not a magician!)