definitely a forum myth that ratings improvement only slow as they reach their max.

I think we don't see that the difference is still there because all those recruit beginning single digit ratings are now Low potential, and those guys are lucky to break into double digits.

About 10 seasons back (early 2013?) I had a PF that started with a Perimeter rating of 10 with high potential.  He redshirted and I didn't spend a minute of practice time on perimeter for 3 seasons.  After those 3 seasons his rating was 1, but he had high-high potential, which a 1 rating almost never sees.

For his 4th season I put 20 minutes into perimeter practice.  He crept up to a 7 rating after that 4th season. 
from 1 -> 2 took 7 practices.  2 -> 3  took 9;   3 -> 4 took 7; 4 -> 5 took 4;  5 -> 6 took 4; 6 -> 7 took 3.

For his 5th season I increased practice from 20 to 24 minutes.
started at 8.  going from 8 > 9 > 10 each took 2 days.  It took the next 11 practices to go from 10 to 20.
I bumped practice up to 30 minutes.
over the last 19 practices his rating soared from a 20 to a 48.

Summary
-------------------------------------------
from 1 > 5 : took 540 minutes       135 minutes per rating point
from 5 > 10: took 236 minutes      47 minutes per point  (+1 gained offseason)
from 10 > 20:  took 264 minutes   26 minutes per point
from 20 > 48:  took 570 minutes   20 minutes per point

There are multiple factors in skill growth. They are:

WE
minutes played
Growth left in the skill
starts (kinda)
Current rating

Work Ethic and minutes work the same. There may be a point of diminishing returns (or maybe it isn't exactly linear) but the more you have the faster you grow.

Growth left is important. I believe that is part of the reason why LP and PER can grow so much in a season. Generally speaking, the more room you have to grow the faster it will grow.

Starts indirectly affect growth in the sense that starts increase WE which increases growth.

Current rating is the one this discussion is on. It's hard to separate (without superior math skills than what I have) but the scores on the extremes grow slower than in the middle. Below 10 and above 90 definitely grow slower and I think the normal growth factor is somewhere around 17-83.

Posted by oldwarrior on 7/10/2014 12:23:00 PM (view original):
definitely a forum myth that ratings improvement only slow as they reach their max.

I think we don't see that the difference is still there because all those recruit beginning single digit ratings are now Low potential, and those guys are lucky to break into double digits.

About 10 seasons back (early 2013?) I had a PF that started with a Perimeter rating of 10 with high potential.  He redshirted and I didn't spend a minute of practice time on perimeter for 3 seasons.  After those 3 seasons his rating was 1, but he had high-high potential, which a 1 rating almost never sees.

For his 4th season I put 20 minutes into perimeter practice.  He crept up to a 7 rating after that 4th season. 
from 1 -> 2 took 7 practices.  2 -> 3  took 9;   3 -> 4 took 7; 4 -> 5 took 4;  5 -> 6 took 4; 6 -> 7 took 3.

For his 5th season I increased practice from 20 to 24 minutes.
started at 8.  going from 8 > 9 > 10 each took 2 days.  It took the next 11 practices to go from 10 to 20.
I bumped practice up to 30 minutes.
over the last 19 practices his rating soared from a 20 to a 48.

Summary
-------------------------------------------
from 1 > 5 : took 540 minutes       135 minutes per rating point
from 5 > 10: took 236 minutes      47 minutes per point  (+1 gained offseason)
from 10 > 20:  took 264 minutes   26 minutes per point
from 20 > 48:  took 570 minutes   20 minutes per point

the way you phrased this, i thought you were disagreeing :) but now i see the opposite is true. i think you are right - this is hard to see nowadays at the high end, because at high ratings, players are necessarily also getting close to their maximums. and on the low end, now the bottom 10 have been wiped out. very good point... hadn't really thought about it, but it would be pretty damn hard for a newer coach to pick up on these things, or for anyone really to pick up on it based on recent experience.

I just got this in a ticket response from CS:

7/14/2014 10:05 AM Customer Support


As players near their max value in a rating, the rate of improvement slows way down.

Posted by Trentonjoe on 7/14/2014 10:19:00 AM (view original):
I just got this in a ticket response from CS:

7/14/2014 10:05 AM Customer Support


As players near their max value in a rating, the rate of improvement slows way down.

nobody is disputing this, and it doesn't have any impact on the discussion about a factor for improvement being the current rating (where a low and high rating will result in slower growth than a middle range rating).