So it boils down to this: can you think of a mechanic that doesn't harm people in a way to dissuade them from waging war, doesn't overpower the strongest in a way that would dissuade others from war, and is not easily bypassable by gimmicky false wars and the like, and does not penalize people during a normal peace cycle after (or before) an important war?
I
think I can, though it's not simple, and there is, of course, still no guarantee that it won't suffer from some of the same problems as the others. Note that all of this, while I have written (at least parts of) it down before, is still highly speculative and if it were to happen, it wouldn't be in the immediate future.
It involves making some changes to recruitment centers that I've wanted to do for a while, to combat the homogenization that's occurred over the past several years. Essentially, I would add the possibility of values above 100% for equipment and training, but in different ways.
Equipment above 100% would be something centers could rarely generate with, and in a similar manner to the current RC-selection system, you wouldn't be able to create more than a small number of them in your realm at a given time. (You could, however, capture them from other realms—adding a potential drive for war.) This would be "Expert Equipment", and would
normally be unusable to the soldiers, because its use is too complex.
But that's where the extra training comes in. RCs could not generate with extra training, but as units engage in battle, they would have the ability to gain training above the usual 100% (just training the unit would not be able to surpass that limit: battle would be required for it). Their total training level would be a limiting factor on their ability to use their equipment, much like adventuring/combat gear level is a limiting factor on an adventurer's ability to use their skills.
But as yet, none of this is particularly a reason to go to war. This extra training should give two reasons:
1) The training developed from fighting decays after a while. The game would keep track of when the unit had been in battles, and after it's been a while since it's been fighting regularly (not just a single battle every few months), its skills would start to get rusty. Slowly at first, then with increasing speed over time. It would
only lose the training from battles: the training it was hired with and the training from training sessions would stay.
2) Though RCs would not
generate with the ability to hire troops with expert training, there would be a way to allow them to have it: You can retire a unit to the RC as expert trainers, which adds a portion of their expert training to any unit hired from there. It would have to be a unit of the same type as the RC, each RC could have only one set of expert trainers, and, as I said, it would only be a
portion of the expert training the unit retired with. Furthermore, and most importantly for this, after a period of time, they would get old, and begin to die off or retire.
This would be something the game would warn you about well in advance, and even when it was coming very close, you could still hire expert-trained units from the RC that could help you in fighting your next war, so even letting it go past the "deadline" would not cripple your fighting ability. But it would leave you with less, and should be a pretty clear incentive to engage in warfare telegraphed
RL months before you would actually need to worry about any reduction in capability.
(I have some ideas lying around, too, about adding a "war fatigue" mechanic that incentivizes breaks in wars that have dragged on for a long time; that one
would involve region stats starting to fall if you ignored it too long, but since its purpose is to get you to find a way to declare peace, hurting your regions is actually in line with that goal, and shouldn't be as counterproductive. The basic reasoning behind this is twofold: 1) Players get bored when all they're doing is marching to the front, fighting one or two battles, then marching home for weeks or months on end; and 2) Giving
both sides an incentive to come to some sort of agreement is likely to increase the frequency, and thus the normalcy, of peace terms that involve a small number of regions or large sums of gold changing hands, rather than abject surrender.)
