Slitherine

Abandon action is your best friend. That will cost you Legacy, but prevents issues described above and deprive them from Legacy sources. Now some context:

When you conquer regions of different Religion and small regions you have Abandon option. Abandon ALWAYS prioritized/Replaces/Overwrites Pillage when region has Just Conquered modifier. You will have to wait for 5 turns before Pillage activates. That's why you have to sandbag some regions strategically, Abandon others to create chokepoints.

Furthermore, when you wait for 10 turns Abandoned regions become fully neutral. The logic is the same as with Rebels taking over or 3rd side neutral to Original owner hitting 2nd side that temporarily occupies Original owner region. Abandon makes sure region won't return to them after Peace deal. So it's to cripple big Empires without overstretching your Demesne limit by a wide margin.

Take a note that you can use Absorb Minor to get them later. And if opponent captures that region between the turns Abandon happens anyway on the next turn. You can chose to Abandon regions with no walls (or queue up Perma FortificaTion destruction) in order to Retaliate against them infinitely for XP points. If Just Conquered status elevated AND region above 50 Loyalty you will have Raid option that triggers at the end of your stack movement. Pillage and Abandon obviously triggers before your stack moves.

RAID is your roll against their roll. Modified on your side by Combat Power. On their side by AntiRaid structures bonus, Garrison Combat Power, Regional Fortification Value multiplied by 3, Stack that sits on top Combat Power. So thats why you want to have about x1.5 higher than their Garrison STR.

If you Pillaged region after big combat or Foraging roll fail, then you can't Abandon. Its prudent to take extra utility Peasants to boost Siege Value when needed, place 1 Peasant on top of the almost starving (negative food income) region and move out to GUARANTEE sortie on the next turn, which returning stack gonna put down in the field instead of city walls. The last thing is leaving one Peasant in the Pillaged region, so they can Abandon it on the next turn while you main stack moved on already. And use Peasants to assign disloyal Generals, so if they Secede they won't tag away your proper regions.

In Empires and Kingdoms Administrative Burden/Court Expenses calculated off:

- your current bracket = the optimal one is coffers filled with lower than 2000 gold, but you are fine up to the 4000 gold. If you can't stay below that watermark you are not playing efficiently. One of the best money dumps = having high Combat Power (300+) stack in your National Capital with 75+ Loyalty General AND/OR constant Gifts AND/OR Regions/Units buying off the opponents (look for the overseas launching pads first). Authority above 50 introduces a new bracket of the extra 10%. Authority above 90 is even more suffering = 25%. Court Expenses are moving average similar to how CDR in Empires was, so you might not notice changes immediately. So you see gradual moves to the new stabilization point.

- nation size = regions formed in provinces half that penalty, but difficulty increases it AND you have free 10 regions discount. Sometimes giving up/exchanging regions is a good idea. As Byzantines on Very Difficult and above you want to create natural buffers by donating to the Nations neutral to the ones bordering you e.g. wall off from Seljuqs to curb stomp their chances of DOWing you. That way you can dispose of low value low loyalty regions that otherwise require major investments. On top of that you don't have nearly enough Governors to oversee them, you need free up Demesne limit for the conquests AND outvalue Court Expenses.

- Buildings amount which punishes densely populated regions. Thats why you need to restructure your regions into value buildings listed in the message above after the initial growth focus IF you don't rush. The food needed to grow an additional population is cubic to city size (quadratic for small cities), and soon becomes much larger than infrastructure cost. The real reason to have Infrastructure focused regions are Metal, Iron, Wood, Tools, Marble, Stone, Glass, Copper. Thats why my initial message vouched to Prioritize Infrastructure and Commerce Regions, while food regions bonuses are not significant due to the structures produced goods AND can be replaced by Import from AIs who love to spam them. Likewise, the contribution to Court Expenses of a structure is proportional to # of structures^0.75. That is, the benefit of a buliding rises with the number of Tier 2 structures of same type in the region, while the cost rises with number of structures (across all types) in the region. It is therefore optimal to have as few types as possible in a region.

- Civilization level, where higher is more painful, but you can't do much about that except going for perpetual Renaissance. Typically you want to progress unless you are rushing for Sudden Death wins, in which case Ascending status of your starting Government Tier is the only realistic goal. RULER Administration value better be above 2, where 2 is the cut off point. Assign your Ruler to lead your suicidal armies that has compositions like 8 Peasants, 4 Light Archers, 4 Xbows (another way to dump the extra cash) to maximize his chance of dying. Unfortunately in Kingdoms chances were lowered due to constant complaints on Steam. Use Assassinate Ruler on self without Deploying Spy Network or having tons of Luck producing structures. Going into Negative Authority never really worth it.

Loyalty degradation from Mercs is overstated. its a rare chance to have -5 Loyalty trigger mostly later in the game.

Relevant factors that influence the chance of Secession are Character loyalty, National Authority (or Local Authority as extra factor if a Peer of the Realm, even if deployed as general), Character Treasury and units under his command (mercs increase a chance), Ruler Administration, Ruler Piety (only if not pious and the would-be rebel is), lack of son for the ruler and a loyal general with units in the capital with at least 100 Combat Power.

The latter is very important as an army with a loyal general in the capital will improve the chance of resisting a coup by 50%, and help Generals to Rebel less often. If you are over Demesne limite, chances increase significantly with each region over.

A less able ruler (with some Administrative scores of 0 or 1) is more likely to be overthrown in a coup while an able ruler (with some score of 4 or 5) is less likely to be affected.