First thought: no control group.

Alternate hypotheses not considered at all, which is really a whole category of problems/thoughts.

Didn't seem to care that they were measuring different things in adults and kids.

As it is, the article is spending about half of its two pages just to say "the way we define 'gifted' is totally different in kids and adults, and so people who are in one group aren't necessarily in the other."

Seems kind of foggy, but there's some use in piling up anecdotes, I think. On the other hand, the first anecdote is about a gifted child who's eventually knocked out by mental illness. This is something which could happen to anyone, and it would be nice to have statistics for the proportion of people who are partially or totally disabled by mental and physical illness in the general population.

I expect that most people accomplish much less than they could, it just shows up more with gifted children.

I also expect that there's a matter of fit between the child's and surrounding adults' temperament. Some people are highly competitive. Some people resist being driven by anything other than personal satisfaction. If you're just looking for accomplishment, then it seems plausible if competitive children have competitive parents, as long as the parents aren't so competitive that they resent the child, and internal satisfaction children have similar parents. Unfortunately, it's hard to arrange this.

IIRC, Howard Gardner claimed that people with great accomplishments were apt have parents who valued hard work without putting huge emphasis on success at every stage.