Maybe I worded myself poorly. I meant 'really smart people' not in absolute IQ terms, but more in 'hey this dude believes some things about a subject he has knowledge of and from what I can tell he is well regarded among peers and seems smart'-terms.
Depends on which way you look at. I wrote the article from the perspective of a founder, not an investor.
The math analogy is a bit stretched though. Following it, the only thing that matters is where you start. If you started in the hill leading up to the global maximum you'd be good. I have no statistics to back this up, but I would suspect reaching a global maximum is more expensive than reaching a local one. Truly world-changing ideas, therefore, don't lend themselves as well to the lean methodology.