Agent upgrades - an alternative model

[Eddie]

I am also not satisfied with the vanilla stat progression. I have been thinking on better ideas for some time. Here is what I came up with, if anyone is interested. I guess this is a good spot to post it. I don't know if it would actually improve gameplay.
My main gripe was that I wanted to reduce micromanagement of soldiers (Has that soldier hit anything on this mission so far? No? then I need to find something for him to shoot or he gets no stat improvements).

The new concept requires training rooms to be available from the start. Stat improvements come only from training, not from missions. Going on a mission will double training efficiency for x amount of days after the mission (the mission experience motivates the soldier to train harder). Going above the training stat cap is possible when the mission bonus is active, but progression is then not double but standard.

Further possible additions:
- Stat loss when not training. When training and skill is above training stat cap, it will slowly go down to that cap unless the mission bonus is active. -> To keep being a supersoldier, you need to see regular active combat.
- Soldier retirement. A soldier of sufficient rank can be converted into an instructor (convert soldier to item). An upgraded training facility can be build that allows for higher training stat caps or some other bonus. To build it requires x amount of instructors as resources.

A hybrid model of this concept and vanilla (stat gains on missions) is also possible, but with greatly reduced stat gains from missions. For example, all stat gains work like moral, where each xp point gives 10% chance for a stat improvement, but the improvement is 1 point and not 10.

[BTAxis]

Just going by what you wrote, I think that approach would have problems of its own. As a player, the system you describe would encourage me to never take my soldiers out of training until they couldn't get any stats for free anymore. By that time they would all be at the default training cap, which means they might as well be clone troopers. They would remain clone troopers after that as well.

With your idea your main goal was to eliminate the vanilla system's incentive to game it for optimal gains. I would like to add two more goals: firstly to give the player an incentive to take soldiers on missions, and secondly for soldiers to differ from each other in terms of stats. I mean, it's all the personality they get. Let's not take it away.

So let's try a variation on your original idea:
- No stat gains from missions directly.
- Soldiers earn "mission XP" for going on missions, a la XCOM 2012. Exact method of earning this XP is up in the air, but the focus is on not giving the player explicit feedback that will incentivize gaming.
- When enough XP has been earned to "level up", a soldier can go in training for X days. Once training is complete that soldier gets stat gains in a random range per stat, a la Fire Emblem.
- Some maximum limit to gains earned this way so that a soldier can max out.

In this idea I chose to send soldiers into training to improve rather than improving them automatically because that way the player has more feedback on the process, plus it encourages building a deeper roster.

For me personally this would work better. That said, I never really had a problem with the vanilla system anyway, so I'm not really too invested here.

[Eddie]

The clone trooper thing is the most obvious problem. It depends on game design though.

How much do your soldiers and training facilities cost? When they are costly you can't just have a large inactive roster in training and wait for them to max out.
Another way to deal with the problem: Set training time and training cap without the mission bonus to levels that discourage inactive soldiers. For example accuracy cap 60, +5 average stat gain per month.

Having a hybrid model where you get max +1 stat gain per mission also reduces the clone trooper problem, while toning down gaming for stat gains.