So far, this thread has done a very nice job identifying what different types of training improve which skills, and how fast. There's another aspect of training that I don't think has really been addressed yet, namely, how much the trainee matters. That is, how much do age and potential effect training speed? When do players hit their potential caps? And so on...
So, here's what I propose. I'm going to start compiling training data for anyone who's interested. If we get enough people involved, we should very quickly be able to get some real results. With enough trainees, we can regress training factors against results and figure out exactly how much a trainee's characteristics affect his training.
I'll of course share the results with everyone who's been involved in this. A caveat, I will need age, height and potential as well as their current ratings. I don't ask for names, but height/age do make the players identifiable. I won't disclose this, but if you'd rather not tell me, no need to participate!
If you want to participate, BB-mail me the following info (I'll also check this thread if you'd prefer to post here, but again, I don't want people to worry about disclosing their players' ratings)
Per team:
Trainer level, Training type, Position trained
Per trained player (even if they had no pops!):
Age, Potential, Height
All pre-training skills (numerically, e.g. 7 6 6 5 7 5 3 3 2 4)
Which skills which popped