Very interesting. Now that we know that training has very little effect with regards to developing players, i think it would be good to see a test of what actually works other than playing them in the first team, as it sometimes isnt an option to play a much worse player in first team games just because they have potential. for example if you have a world class 5* striker then buy a 20 year old 2.5* wonderkid backup, most people are not benching the 5* just because it helps the 20 year olds development.

perhaps there can be some sort of test made with identical players, say clone the same 20 year old wonderkid, too good for your under 21 team but not good enough to start for your first team, and put him in different scenarios

1.put him in u18 squad (im not sure if this is even allowed as he is 20, but its worth trying)
2.put him in u21 squad
3.keep him in first team squad, he never plays for the first team but make him available to play in u18 and u21 matches
4.loan him to a low league team with poor training/coaches where he will play every week
5.loan him to a good team where he is more of a backup/squad player
6.baseline of just not playing him at all in any game
7.optionally you could maybe add any other scenario you can come up with other than actually using him in first team matches, (something like where he never starts games but comes off the bench in cup games, wouldnt be able to work with holidaying through the season though).

then holiday for the year, or play the season out a few times and see which is most effective.

it would also be interesting to see what coaching attributes matter if at all. does having a coaching team where certain attributes such as working with youngsters make any difference to the development