Even so, if you are suggesting dyslexia (in these cases) probably has or probably hasn't affected outcomes, that's problematic. The examples you give provide zero information either way; without enumerating a statistically relevant section of people considered successful, these examples seem to provide information, but actually provide nothing.

You'd also have to adjust for other factors such as parents trying harder to find areas (other than academia) where dyslexic children can consider themselves successful. It's almost intractable to think of a study that could remove these correlations from any statistically relevant top level study...