The "3 years" varies, but what I've observed with talented juniors is that it takes roughly the same amount of time to get from 2000 to 2200 USCF as it takes to get from 1000 to 2000.
The improvement timetrack for talented players looks like this

Also, the later a talented kid starts chess the quicker he'll get to 2000 and to 2200. At some point in young adulthood this trend probably reverses.

My statistics prof would have thrown ppl out of the seminar saying such things.
Obviously he wasn't a talented prof. ;)

I went from 2000 to 2200 in one year, and I was only averaging 2-3 hours of study per day (I kept track of this). If you were really aiming to be a top GM I imagine 7-8 hours a day wouldn't be out of the question. What I'm saying is if you study the best learning resources, spend the right amount of focused hours per day training, and on top of that have good health, a good mentality, and real drive you can probably grow extremely quickly. However everyone seems to slow down a lot in the 2600-2700 range.

3 years from 2000 to 2200 is frankly a long time for talented / quickly improving youth from what I've seen.
Which makes sense considering there are 2800 players in their 20s. If it took them 3 years to gain 200 points at the 2000 level, then they'd never be a top 10 player

Then sure, 2200 in 6 years is talented IMO... I mean, most never get there (or even close!)
I haven't paid enough attention to have a cool observation like yyoochess. FWIW It does seem that pre-teens get suck at (or close to) 2000 while teens improve at a more steady rate. Not super fast, but not getting stuck for a year or two either.

Then sure, 2200 in 6 years is talented IMO... I mean, most never get there (or even close!)
I haven't paid enough attention to have a cool observation like yyoochess. FWIW It does seem that pre-teens get suck at (or close to) 2000 while teens improve at a more steady rate. Not super fast, but not getting stuck for a year or two either.
Another interesting phenomenon I've observed is that when the most talented American pre-teens get near 2200 they often jump over it in a big way. One day they are playing 2100 level chess and then, boom, it's 2250/2300 level though it may still take a little while for their ratings to catch up. We saw this with Hikaru Nakamura, Awonder Liang, etc. It's as if something has clicked in their brains. That clicking happens around age 10 regardless of what age they started their chess. I wonder if psychologists have an explanation for it?

My statistics prof would have thrown ppl out of the seminar saying such things.
Obviously he wasn't a talented prof. ;)
Obviously u didn't get my posting

And you're getting this from what? or simply making a claim. And I agree 3 years of serious study to reach 2000 isn't that good. It's definitely not bad but it's probably around average.

I went from 2000 to 2200 in one year, and I was only averaging 2-3 hours of study per day (I kept track of this). If you were really aiming to be a top GM I imagine 7-8 hours a day wouldn't be out of the question. What I'm saying is if you study the best learning resources, spend the right amount of focused hours per day training, and on top of that have good health, a good mentality, and real drive you can probably grow extremely quickly. However everyone seems to slow down a lot in the 2600-2700 range.
LMAO oh ONLY 2-3 hours per day... "I worked EXTREMELY hard for 1 year to go from 2000-2200." is what should have been said.
Another 3 years to reach 2200 master