[ctchocula]

Regarding the first article, they discovered 9331 studies that studied the impact of practice on performance, and narrowed it down to 157 studies based on their criteria. The breakdown in terms of genre, number of studies, total number of participants, and % of variance in performance capable of being explained by deliberate practice was:

Music 28 1,259: 21%

Games 11 1,291: 26%

Sports 60 2,633: 18%

Professions 7 321: 1%

Education 51 5631: 4%

So not programmers specifically. I didn't read the paper very thoroughly and examine their methodology (perhaps their criteria were such that they were allowed to cherrypick results), but it seems to me that the professions category seems to suffer from particularly low sample size.

It's important to note that under deliberate practice, they did not include starting age. To explain the other ~80% of variance, they proposed starting age, general intelligence and working memory capacity. Pm me if you would like to see the article.