As a different data point, I am training wildly inefficently (in terms of the elastic effect) on both my primary and Utopia teams because I am doing stuff that is probably also going to be wildly ineffective, but felt like it was going to be fun.
With that said, I think the negative effect of the elastic effect is understated at least in the training spreadsheet developed based on the training data project. Heavily understated. On my primary team, I have been training almost exclusively JR for SG with a level 6 trainer for the past two seasons, and as 19 year olds it was taking close to 5 weeks per pop (even the one 6'0 player, so the height of the other two is not *that* severe an impedement). My secondary team has been training 1v1 for FW those two seasons again with little exception, and again those pops are coming in far less frequently than the spreadsheet would have predicted.
This probably means I will be continuing a third season on these training regimes, which will be extremely fun as the players are age 20, as I have little hope that a player entering a season with a high 12 skill and exiting it with a high 15 will miraculously get to high 19/20 in less time.