You don't always need hard data to prove a point. I could relate to this article from my personal experience. Smart people tend to be harsh on themselves and on their solutions.
The "most smart people" part could be revised to something else so it doesn't seem like the conclusion was from hard data.