As the manager of a relatively new team, the debate on playing a 5-man rotation or an 8-man rotation is less relevant to me; instead, I am way more concerned about the impact of the current training system.
TL;DR: To release the trainees from being forced to play entire games, 1) relax the 45+ minutes constraints from 18 and 19-year-old players to all trainable players; and 2) give the managers to better control the game minutes among players. At the bare minimum, please allow the coaches to stick with their original rotations after the garbage time.
For conciseness, I will assume the team is on a single position training. Double position training should be more or less the same though.
For the current meta, I believe that most managers want to keep three trainees when we are running a single position training system. If the three trainees are equally important, the optimal solution is to make them all play 48 minutes.
In the past, most of us are content to have the three players each play on an entire game. But what if this is no longer possible?
I believe that the most natural solution is as below. Suppose the three players are A, B, and C.
Saturday: A as the starter, B as the bench.
Tuesday: B as the starter, C as the bench.
Thursday: C as the starter, A as the bench.
But, do they all get 48 minutes in this system? No. There are variations on the game minutes share between each game, even if you choose to "Strictly follow the depth chart". Despite the fact that it is impossible to have players sub in at the exact same time each game, the major reason is that in some games, subs played longer than usual, due to the garbage time.
Therefore, if the goal is to approximate the current meta while enforcing players not to play the entire games, we need a "slack factor", which means to guarantee full training effect even when the player did not play 48+.
Currently, there was an update that guarantees full training for players of 18 and 19 years old who play 45+ minutes. This is not enough for two reasons.
i) The training cycle is way longer than two years. Managers will train an HoF until he is 30 or so.
ii) The variation between game minutes share is way greater than 3 minutes. When choosing "Strictly follow depth chart", the starter/bench minutes split to about 36/12 in a close game, while in one-sided games it is about 24/24.
Therefore, I will personally suggest guaranteeing full training for players under 30 who played 42+ minutes.
In addition, we need to have a way to control the minute share in a lopsided game, so that a trainee played as a starter can stay longer on the court. I have seen many good suggestions in the forum, and here are some of them:
i) (Maybe simpler to implement) Add an option for substitutions, to allow the coach to "Strictly follow the depth chart, even after garbage time". This should largely stabilize the variation on game minutes and make things more predictable.
ii) (Maybe more advanced) Let the manager strictly decide how long each player should play. Players going after that preset threshold will be benched no matter what.
Hope we will reach a better consensus between training-oriented teams and strength-oriented teams.