It isn't, corporate leadership is effectively a tournament and CEO and executive salaries are the prize for winning it. The point of having a high prize is to get a lot of people to compete for it, not to reward the winners for the effort they've put it (which may only be slightly more than the runner up, while the reward is several times what the runner up gets).
Like, do you think the best golfer in the PGA tournament is "really worth" ten times as much money as the 14th best golfer? Did they work ten times as hard?
I don't think your analogy proves the point you think it does and is in fact a pretty great argument for why CEOs should be paid less and taxed more. A large part of the success of both athletes and CEOs is up to circumstances and chance, as well as the other competitors who may be deficient in a number of ways. Being less incompetent than your peers doesn't mean the same thing as being competent.
Pay isn't just about giving people what they deserve. It's also about providing the incentives that create the behavior you want. You can do that through piece-work compensation, but that's extremely hard to manage for executives (and knowledge workers, too, but that's a separate discussion).
If your workers aren't risk-adverse, then a tournament is exactly as efficient as a piece-work compensation scheme for convincing them to put in effort. The proof is a bit more involved than I'm willing to digest for a HN comment, but there's a full paper you can read about this here: http://faculty.smu.edu/Millimet/classes/eco7321/papers/lazea...
Anyhow, what all this means is that providing a prize to top performers means you can get a ton of effort, training, and work from people competing for it, and you don't have to incur the expense of monitoring the inputs into performance or quantifying the overall performance rendered. You only have to care about the rank-ordering of performance.
Makes you wonder if being a CEO is really worth 400X the average salary of the plebs or if it's all just luck and smooth talking.