You are right that "successful in data scientist roles would not be successful (or at least, much less so) either as programmers or statisticians", assuming they do not train a bit more and become better statisticians (a subject in which world-class experts have wildly different opinions about the very foundations of the field) or programmers (a 1X, say).
What I had maybe issues articulating is that the Dilbert's creator is focusing on the system of skills (and, what does "better than average" mean? Better than the population average or better than the average of people professionally working, say professional comic artists? Because we are talking about two different things) instead of the application of the skills. He had a brilliant (a posteriori) idea that did not require to be neither Caravaggio nor Dave Chapelle. The lesson is that you can have success drawing like a 6-year old (which is better than what I can do) and some deadpan humor about office work which is funny one every ten times. Which is, IMO, way below the average skills of professionals.
His "lesson" is between "nothing new" and "so what": nothing new because we have seen many successful people having near-zero talents or skills (that would be a long book for me), and so what because it is not the above average skills (and I ask again: average of what?) but the more or less fortuitous choice of a profession or activity.
We have vague-casting, vague-posting, and, in this case, vague-philosophizing.
A decent generalist that nobody may want, though.
I don't love Adam's formulation, but it is close to something real.
But the argument is being above average in two or more areas can make you as impactful as being top of your field in one area, even though you are nowhere near that talented in any different field.
Even with your example, I would counter that a huge number of people successful in data scientist roles would not be successful (or at least, much less so) either as programmers or statisticians. It seems many of the earliest wave of data scientists (e.g. pre bootcamps etc.) have failed to thrive at at least one of the two before ending up in "data science" (caveat, agree this is poorly defined).