The Pareto principle is that 80% of result comes from 20% of the effort. So 80% of the 10k hours worth of skill (8k) comes from 20% of the actual time (2k). If you continue multiplying the hours worth of skill by 0.8 and the actual hours involved by 0.2, you get 5120 hours worth of skill in 80 actual hours of practice.
Of course, if you keep going, you get the first 800 hours worth of skill in less than a second (and the first 300 in a mere millisecond), so I'm not sure the repeated application is justified.
"Half the money I spend on advertising is wasted; the trouble is I don't know which half." - John Wanamaker
The question is WHICH of those 80 hours account for 51.2% of the value of 10K hours of practice. It's not necessarily the FIRST 80 hours.
It's very possible you could get 8% of your skill in less than a second; it just might come at a lightning-strike eureka or moment of epiphany during hour 8,679.
I think people also forget that the point of the 10k hours is not just the time, but also the fact that experts actively spend time figuring our their weaknesses, and focus on that, while making sure their strengths don't start to wilt. Planning rather than just doing, are also important.
> so I'm not sure the repeated application is justified
Probably not. I don't think that whoever came up with 80/20 had this in mind, but that's the fun with mathy-sounding rules of thumb - you get to mix them and/or fold them on themselves and see what comes out ;).
More seriously, you may try to rescue the rule and such application by noting that it is the right 20% of work required to get 80% of result. Finding that work in the space of all work to be done requires effort on itself. Finding the right 4%, or right 0.8%, requires even more bits of knowledge, thus more meta-work.
Fortunately, we may cheat a little bit, because those who came before us and mastered the skill we want often already did the big part of figuring out which work has most payoff at the beginning.
I meant the "Pareto principle"[0], or 80-20 rule, that says that you can get 80% of effect with 20% of work. If you apply the rule to itself, you get the 64-4 rule, and if you do this again, you get 51.2% of results from 0.8% of work ;).
