Just a few thoughts I've been having. I don't like the way Mongols and Huns build cities in Civ series, and I can't say I like the way Hordes collect tax, sit in the cities and hire mercs. I'm pretty sure nobody will take anything serious here, but hell, let me just vent out a few words and waste my time, your time and the precious non-renewable resource called your bandwidth.
1. Tax is buillshit. A province does not have the same value for nomads and settlers alike. The "tax" value should be recalculated based on the terrain information already present in the game: plains and hills should be valuable Lebensraum for nomads; forests, tundra, high mountains and deserts not, and certainly not one-province small islands. This would mean that some base-tax 1 province in Siberia should be much more attractive for the Hordes compared to the Russians, and this should give the Hordes more incentive to actively seek them. If you're a steppe nomad herding your flock, you would prefer a central Asian plain province compared to, say, Barcelona... though you might certainly want to loot Barcelona. (More on it later)
2. Hordes don't recruit; they mobilize. Every adult male herder is essentially a trained horseman. The recruitment time and the maintenance for the Hordes should be dramatically reduced compared to the settlers, who need to recruit, equip and train them. The Horde recruitment would be equivalent to hiring mercs, time-wise.
3. However, nomad population grows much more slowly compared to the settlers. Horde manpower growth (whose maximum should be calculated according to the Horde standard) should reflect that; 50% slower manpower recovery sounds O.K. to me. Since their whole male non-elderly adult population would be potential horsemen, the 50% larger maximum (as it currently is) might be kept intact but the recovery should be far slower, so the manpower actually becomes more precious. Nomads didn't lose because they got weaker; there are just so many, so many, so many more farmers as time went that they couldn't win anymore.
4. "We've been in peace too long" - this shouldn't be mechanically triggered after a certain number of years. Why do the Hordes go to war? Nomadic lifestyle cannot support as much population density, and it is the population pressure which drives them to seek more land to herd their flock. Let's say, if a Horde nation has more than 70% of their manpower reserve, the "Khan is pussy" event should trigger with higher and higher frequency, since now they would be in conflict with each other over limited pastoral lands. That is, a depleted Horde shouldn't be getting these events since there are lands to go around; an overpopulated Horde would be forced into war.
5. Cores. What does it mean for a province to be a core territory of a nomadic tribe? It's not just about forging documents and taking over the administration, as it would be for the settlers. If the settlers still have farms around, it's going to be very inconvenient for them and as victors they'd certainly want to get rid of that - that's the whole purpose of Horde conquest, beside looting and population control. A Horde nation would suffer, say, 150% of normal ADM cost plus the equal number of MIL points, and it should take even longer than usual to core a province... This would imply that not only the province has changed owner, a large portion of population itself is displaced. Instead, the vacated settler territory would be Horde pastureland, so the culture and religion change to that of the Horde. This would also make the settler core to disappear sooner, since the culture changed.
6. Then, what if a settler nation decides to conquer a Horde territory and core it? It should take longer since the area needs to be basically repopulated and you'd be building hell lot of new farms (even if there is a surviving city, in EU4 time frame most of the world was still agrarian), but it would cost them the normal coring cost. And since the nomads are gone and every farmland is reclaimed by then, the culture and religion also follows the conquerer's, be it Persian or Russian or Chinese.
7. What about the Timurids with huge Persian territories, etc? Much of the Central Asia has a rich history of city states and trade routes, and what happens to them if a Horde takes control? Timurids would not have them cored. A player (or the AI) should make a decision whether force-coring the mountainous Persian territory is worth the trouble or not. Instead, a Horde can choose to sack its non-cored territory at a regular interval (while fighting the rebels of course). When a piece of province is no longer valuable and is more of a liability than an asset, a Horde can simply abandon it. This would spawn/flip to whatever core-holding nation with the highest revolt risk.
This would mean that a Horde nation can either choose to continue pillaging everything down to nothing and leave for a new prey; or, if they eventually find a suitable terrain, they would choose to "settle" there and eliminate settlers. (They would "core" it.) As they continue pillaging settler provinces, these non-cored provinces are essentially their life support, so should provide a normal manpower (for the Hordes) but the max manpower and manpower recovery rate should eventually fall down, as the province exhaustion rises higher. That way, Timurids would not try to stabilize Persia as the AIs (and human players) do; they would rush out somewhere else when the Persian / Khorasani provinces become depleted.
8. "Horde" CB should allow cheaper vassalization, perhaps 100% should always enable vassalization. However, that needs to come with a penalty as well: being a vassal to a "Horde" government should permanently set a significant malus (say... -50? is that too much or too little?) and the AI should seek harder to use their "Independence" CB if the overlord is a despicable barbarian. (Or a successful pretender or revolutionary rebellions should break the vassalage) And while the Hordes collect tributes, their tribute income does not increase their force limit nor contribute anything to their manpower. Normal vassalization rules would be in place between Hordes; and if a settler vassalizes a Horde, then the settler will not expect money but the comparable force limit increase.
In addition, a Horde should not be able to annex a settler vassal. Instead, in a truly Horde fashion, it should break vassalage (without stab hit) and go for conquest once again. (You might think this would be exploited, so a tiny vassal could be attacked and attacked... no, that'd be useless because that would not solve the overpopulation (reserve manpower reaching their 10-year maximum) which would trigger "our Khan sucks" events.
9. I partially agree that it doesn't suit Hordes to be building universities and research techs. Okay, fine. Let's make the Hordes completely unable to research anything at all. Instead, a Horde with an amicable relationship with a settler nation (if it is possible) OR a Horde with a settler core in its own territory (probably more likely, and that settler nation probably must also exist) can buy technology only if the seller has a higher tech. Say, if Kazan conquers a Muscovite province, and Kazan is 3/3/3 Muscovy is 4/4/4. Kazan can spend both monarch points AND the extra money they collected from looting to find a treasonous Muscovite from the territory to transfer technology.. (and Kazan would still remain Nomad group with the associated tech costs and units) Westernization, then, will now mean they're just buying a different kind of technology through a different kind of source, and what pisses off the Hordes would be they'd be using unfamiliar weapons instead of what they take pride in. However, this should probably be compensated with an extra horde bonus. Say, army tradition loss -50% or something like that. Eventually, when they cannot pay for the new tech anymore, they'll collapse, and fast, just as many of the historical Hordes did when they couldn't conquer anymore.
10. Hordes should be able to "siege" enemy province without infantry. This would mean that the Horde cavalry unit would pillage everything in sight every siege phase, and this should give them a small income every siege phase, until the defending garrison simply runs out of any potential supply route. But a 100% cavalry burn-everything-make-everything-waste-until-they-starve should take noticeably longer to complete... which you might want to delay for the sake of loot income in the mean time. However, repeating this action should accumulate not only the AE, but another kind of malus.. say, -5 "uncivilized brutes" per province looted in a war.
So, if you play the Oirat Horde and you want to run over Ming, lay complete waste of every province and vassalize Ming in one go, you should be able to do it. But you should also pay the huge AE (for vassalizing Ming) and something like -150 "uncivilized brutes" if you sacked every single one of them. You deal with the coalition, of course and you shouldn't complain if everyone including your vassal Ming comes to kick you out of your pasture land and build farms there.
11. You can't hire mercs. That's for the pussies who can't ride horses and shoot arrows. You can hire mercs if you reform your government to a my-ass-is-too-heavy-to-move government.
Number details don't really matter. That's balancing. But the antagonism between nomadic and sedantary lifestyles have been a very significant part of Eurasian history and the Hordes deserve a more tailored playstyle which suits them better. And if properly balanced, I think the AI Hordes will shrink to insignificance by 1821 with these mechanics, as the Russians, Persians and Chinese will kick them out of their lands and if they still exist, it'd be because their lands are not valuable to the settlers (tax-wise). Which is also historically what happened.
