Mastering a Skill?

I was wondering, what do most coaches consider mastering a skill, specifically in tumbling? I mean, does the athlete have to be able to perform it a specific amount of times in a day, or make a certain number of the skill out of however many they perform.  And at what point should the gymnast move on to the next skill.  After seeing how perfectly the foundations program is laid out on when to move on in progressions, it kind of dawned on me that i was tumbling blindly.  I never really knew when I could consider myself proficient at a skill, and I moved on to the next skill without really knowing if I should or not.  Because of this, I always seemed to gain and lose skills and never really made any consistent linear progression.  I always seemed to bounce around.  I don't know, just a question that really just hit me.  So basically I am asking...

1) What defines mastery of a skill (mostly in terms of tumbling) in gymnastics?

2) And at what point should a gymnast move on in progression?  For example, after i develop a good layout, can I start twisting before it's completely mastered?

1. Mastery of a skill depends on what you want to do with it. Some skills are extremely open-ended.

2. You must be supremely confident in your layout before you begin attempting to twist it.

My vague guess at a definition of mastery from a gymnastics perspective would be being able to perform it reliably (> 80% of the time), with no execution deductions.

This is a definition promulgated by Nick Blanton in his video: http://www.youtube.com/watch?v=BPYMBNtWHV4  (Strategies in Gymnastics Routine Training) and it seems like a decent enough definition for those of us that tend to obsess about numbers and statistics when we approach training (guilty as charged).

Other than that, what Hari said above is pretty accurate - you can have your own definition of mastery depending on your use for a certain skill. I for one believe that no one can truly "master" a round off, even if they can perform it > 80% of the time without execution deductions. The roundoff is a skill where you're trying to maximize transfer of momentum; hence, there will always be ways to improve it. (Similar to a rowing stroke - even the best rowers in the world do not have a perfect stroke, though they do multiple thousands of them in practice every single day).

Yeah, a round off is particularly open-ended. If you had a perfect round off backhandspring, you could get the power to do a triple back on floor, no problem. But for most people, that is not a useful standard of mastery.

On a high level, what you want when learning a new skill is to have such a level of proficiency in the underlying mechanism that even when you're distracted by thinking about how to perform a new variation and nervous about the possibility of injury, your execution will still be good enough to see you safely through.

I never really thought about it in terms of a round off or skills like that.  Thats really interesting!  I think between both of these replies I have a pretty good idea of what i need.  I like the 80% rule as the guideline, it makes sense that if I cant do a skill well 8/10 times, i probably have no business moving on.  At the same time though, it should be at least semi easy, so to the point where you don't have to be completely focused on this one skill, and you can afford some distraction, just as Hari said.  The idea of logging the number of skills/connections that the video talked about seemed really useful!  I'm definitely one of those people that like to look at numbers too, so i'll have to give that a try.  Thanks for the responses guys.  I already know that as I'm practicing I'll probably have more questions about this topic, but this helps so much.  Like I said, I never really know when to move on to the next skill or progression, so I think this will be invaluable!  

Hashem Bakr said:

I never really thought about it in terms of a round off or skills like that.  Thats really interesting!  I think between both of these replies I have a pretty good idea of what i need.  I like the 80% rule as the guideline, it makes sense that if I cant do a skill well 8/10 times, i probably have no business moving on.  At the same time though, it should be at least semi easy, so to the point where you don't have to be completely focused on this one skill, and you can afford some distraction, just as Hari said.  The idea of logging the number of skills/connections that the video talked about seemed really useful!  I'm definitely one of those people that like to look at numbers too, so i'll have to give that a try.  Thanks for the responses guys.  I already know that as I'm practicing I'll probably have more questions about this topic, but this helps so much.  Like I said, I never really know when to move on to the next skill or progression, so I think this will be invaluable!  

You have absolutely no business attempting to twist a layout unless you can land it safely 100% of the time. Anything else would be putting yourself at high risk for a serious injury. You need to be at a level where you don't even have to think about it to do it correctly.