A big portion of the job is just being able to not crumble apart, over a long period of time. Another metric: How prone are you to making a horrible decision when under extreme stress, instead of an average one?
The depths at which the worst case scenarios can bottom out are deep. However, when the downside does not materialize, all that is left is an average decision. And average decisions are easy to critique in hindsight, for the CEOs themselves, as well as for most other people.
In that sense, programming is similar: When you steer clear of horrible code, consinstently and over a long period of time, the plain looking stuff that comes out the other end can look like a given to somebody who hasn't been through enough code filled with bad decisions to appreciate their absence.