* The loss of 12.7IC/day from making peace.
12.7% out of 22 ic is only 2.8 ic. Later it will be ~8% out of a higher effective ic.
*The effective loss of 12 infantry and 3 mountaineer divisions, all with 60 EXP, which protects China's flank in the event of Shanxi falling. The additional loss of 350 manpower assuming the CHC AI builds more units.
How can those 15 divisions be obtained? The only way i can think of is to get them as exp forces and steal them, but for this CHC would need to agree to an alliance, which they will not. CHC must be annexed earlier or later. Later will be cheaper in terms of military losses, earlier will allow to use 2.53 base ic earlier. Over 720 days that would make a difference of over 2000 icd. On the orther hand real high losses might reduce money from unused manpower, but that will be less than 0.5 ic and decrease further because dead people donnot age and retire.
*Allows a reduction of military salaries (which, assuming the player keeps the starting army, is -$6.45/day [or ~2.58IC/day] if moved down to 0 military salaries - but then attrition will limit this benefit).
During peace however attrition will be lower anyway. But i donnot think peace is likely to be in chinas best interest anyway. All national territory but mongolia must be gained back soon. Mongolia follows during barbarossa.
I am all ears, how is that possible without baiting the AI out of it's provinces, all mountainous with 3 land forts each? Especially when Nat China forces start with 8-11% ESE on the CHC border, with just ~20% ESE after all the 15 dissent is reduced months later.
[...]
*Took heavy losses capturing it - meaning more IC needed for reinforcements.
Once alliances with shanxi and guangxi are created china has 93 Inf divisions at its disposal. That is 31 divisions per 5 in the three provinces of CHC. The 2 mountains can be attacked from 4 provinces, the hills can be attacked from 3 sides only. It should be obvious that CHI can take the 2 vp provinces at some ease, albeit losses might be significant. But even if all 93 divisions take 10% loss, than only 330 icd are required for reinforcements. I would be more worried about supply consumption. Also ESE is quite an issue, attacking from the desert provinces might not be the best of ideas or fighting losses might be topped by attrition loses.
*The loss of the Xi'an event, meaning the warlords have to be annexed via war, thereby losing all their units and generating dissent from the declaration anyway, or waiting until Japan declares war on China (so if Japan is also human, then he/she could wait until 1939 or whenever, ensuring the human China can do very little in the build up unless he/she invades the warlords).
It is probably best to assume that marco polo triggers normally. But even if not you should be aware that the warlords can be annexed by capturing all vps. Than all their manpower and all their military left will be property of the Republic of China as im case of inherit event or in spanish civil war. The the goal must be to take those provinces with minimal losses on both sides. Exploiting exp forces can do that with zero losses. But even if that is not done again losses would probably not mount to more than 10% per division which should be considered acceptable. Those losses will save more icd at maintance than reinforcing icd will be needed later. War finances war and some experience on leaders and divisions is nice to have, too.
It should be noted what effects it has it to take events 90+91 in their historical choices or avoiding both events by annexing all 4 warlords by military means:
The events give -10% dissent, the integration events will add 20% dissent and a later DoW on CHC will cause further 4.44% dissent so that is +14.44% in total. Declaring war on 4 warlords instead will cause 4 times 4.44% dissent, which adds up to 17.76% dissent.
The events give +2 democratic, +2 interventionism, +2 freedom and +1 political left.
The events will kill 1 minister(-15% supply consumption, somewhat useful at low ese) and 2 leaders, one of them a good one.
The 4 DoWs will cause no belligerence. Crushing those insubordinate warlords within the borders of the Republic of China does not disturb international affairs.
I donnot see a decisive advantage for either case. Taking the events is simply less time effort for the human player.