This does not disagree with the point. I've encountered some of the research that they are referring to in http://www.amazon.com/First-Break-All-Rules-Differently/dp/0.... What Gallup has done is developed their own methodology to identify who will be effective managers, and their claim is that the managers that they identify as effective have a huge impact on performance. But their identification has a fairly low correlation with what most in the organization perceive as who is effective.

If you re-read the blog keeping what I just said in mind, you will see that it fits. And you will further see that your point is in complete agreement - the people who are perceived as having good management potential have a low correlation with actual performance.

The blog points out that this happens at the bottom level because people who are good at non-management tasks get promoted to management and may or may not be a fit. But as you go up the latter you find that a lot of what tends to get rewarded is visible success. Which gives an unfair edge to self-promoting narcissists who manage to make their occasional successes more visible than they should be. The result being that perception and reality tend to diverge.

CEO, vs Manager? I dunno, I know they're one and the same in a start-up, but I took this article to be talking about larger companies, with teams that all have managers. In my experience, the article is pretty bang-on, too.