A good analogy is that expertise is like an area of a rectangle: Both innate factors and practice contribute to it.
Which contributes more? It depends on the variance on each dimension. We see from Fig 3 (page 9) in the paper that the unexplained variance is highest for elite athletes at 99%. Why? Because almost all of them practice as much as it is useful to do so. In other words, the heights of their rectangles are about the same, almost all of their area differences depend on other factors [1]. For other groups, the variation in practice time explains more about their performance differences.
[1] These may include innate talent, techniques, psychological factors, etc. With more than two dimensions, we may need to change the analogy which I do not have the time to get into now. Also, as with most analogies, it is not perfect. Practice time is likely not independent of talent.
Which contributes more? It depends on the variance on each dimension. We see from Fig 3 (page 9) in the paper that the unexplained variance is highest for elite athletes at 99%. Why? Because almost all of them practice as much as it is useful to do so. In other words, the heights of their rectangles are about the same, almost all of their area differences depend on other factors [1]. For other groups, the variation in practice time explains more about their performance differences.
[1] These may include innate talent, techniques, psychological factors, etc. With more than two dimensions, we may need to change the analogy which I do not have the time to get into now. Also, as with most analogies, it is not perfect. Practice time is likely not independent of talent.