Three Kingdoms Guide - Cutting through the complexity

[Mox_the_disorganised]

This game has a lot of complexity that the player can investigate (eg WuXing bonuses, Character bonuses, Army composition, different building approaches, items, courts, spying) - I found it a little mystifying and not all of it has a good in-game payoff to the player. Hopefully these ideas make the game more approachable for the average player. Fair warning, these are general rules therefore are not an optimal approach for any given Faction.

This guide is for players who already have 5-100hrs of game time & want to play Han factions on very hard/legendary difficulty, rise of the Warlords, patch 1.6, with a Com/Industry focus. If your looking at something else, this wont apply. Others have done a great job of explaining game mechanics & tradeoffs, links below. This post adds two things:
-economics of mega vs average commanderies, and
-additional focus on game stage (EarlyGame turns 1-40, MidGame turns 40-120, LateGame turns 120-180) to the strats that others have created

The TLDR is:
-specialize provinces into food, megaIncome & income. Rough rule of thumb is if there are two blue/purple counties in a commandery -> megaIncome. With the remainder of commanderies, If 1-3 food creating counties -> food province. Rest is income/free for support buildings (conscription/TradeInfluence/Prestige)
-Low payback periods (return from investment in <12 turns) are important for snowballing
-Buying provinces can have a better payback period than conquering, both early & lategame
-suggested 10 techs for first 50 turns, so you dont have to agonize over the tech tree

Game mechanics
https://www.reddit.com/r/totalwar/comments/bwepxi/a_guide_to_buildings_and_economy_in_three_kingdoms/

https://www.reddit.com/r/totalwar/comments/btzbcq/3k_tips_tricks_army_empire_and_character/

More advanced:
https://steamcommunity.com/sharedfiles/filedetails/?id=1762205509

Very detailed / advanced: big thanks to the work Serious Trivia has put into casting about this game, who you can find on youTube. It's worth the time, but needs 20+hrs to cover all of it.

-----------

***Game summary:
faction's economy enables army & buying territory, armies increase territory, victory comes from 500 prestige points (territory/buildings give prestige) AND armies to take seats/force abdications. You can do it differently, but this seems a dominant strategy for the Han factions

***Generating, Keeping & Using your income

There are 73 commanderies / 9 passes on the map. A strong economy requires some cities to build tall, which consume a lot of food (24 to 46 food). To identify & feed these cities, you should mentally sort the map into:
~15 big money makers, ~10k income at full build
~25 small money makers, ~5k income at full build
33 food makers (<1k gold & 25-60 net food),
Splitting provinces this way means you have a mental shortcut of what to build and when, rather than thinking out each province and the 25 buildings types. See below for a list of MegaProvinces, level of community debate will increase as you get to the end of the list.

Provinces should have a good investment payoff:
Payoff means the turns it takes to recoup your building upgrade costs. If you're into doing calculations then take the cost of upgrade and divide by all additional income from building upgrade (flat rate gain plus % based gain). Rule of thumb is a building's payoff should be less than 12turns.
-if your not into calculations, just stuff your money into upgrading mega provinces & food provinces (see below)
-Reduce payoff time by:
1) reducing build costs
oversee construction sentinel assignment, sentinel administrator discount, hire contractor/industrial focusevents, cost reduction ancillary sets, WuXing building cost reduction, -10% cost reduction & -20% mine cost reduction techs
2) increase income gains
% income tech gains, administrator with high expertise, blue/purple assignments). In the early game, build cost reductions can be the difference between being able to afford a building or not.
- If you stack administrator + item reduction + events + WuXing, this can be 90+% off build costs, which has a very large impact on payback times. See table below for some examples.
Making upgrades with better payback periods than your opponents has at least four advantages:
-more income means you can field more army stacks than your opponent. Even with legendary penalties, at 2-armies-to-1 odds you can wipe the floor with opponents that have similar or more territory than you.
-if two factions invest money at different rates of return, the advantage to the better investor increases at an increasing rate over time. See graph below for the compounding returns, over time, for 50,000 gold invested at different payback rates (linked graph is effectively a compound interest graph for any finance otaku's out there)
https://imgur.com/a/lBcip61
-taking less, but very valuable provinces enables near zero corruption in mid game, further enhancing income advantage / snowball potential.
-on average, less territories means you've met less factions, which means less declarations of war. Doesn't apply if you've relocated your empire to do this Strat.


For a similar level of $ investment, mega provinces have a 100% faster payoff than average provinces:

	Avg Income Payoff	Food Payoff	Prestige	Funds invested	Investment discount by 25%	Raw Payoff (turns)	Discounted payoff (turns)
MegaProvinces (lv10)	-10k	-46	80+	158k	126.4k	15.8	12.6
MegaProvinces (lv8)	-9k	-24	60+	130k	104	14	12**
AvgProvinces (lv7)	~4k	-16	50+	110k	96	27.5	24**
AvgProvinces (lv5)	~2k?	-6	40+	60k	n/a	30	n/a
FoodProvince	~1k	20-60	30+	32k	n/a	n/a	n/a

This chart is very summarised so users dont have to getting bogged down int the detail. Skip the next paragraph if that sounds like you
Notes:
-MegaProvinces are: top12: Jianning, Danyang, Poyang, Jiangyang, Taiyuan, Zangke, Changsha, Cangwu (peasantry build), Badong, Hedong, Wudu, Jincheng
the next ones are more speculative: Luoyang, Dong, Wulin, Yulin, Nanhai, Yuanan, etc
for more detail see https://www.youtube.com/watch?v=OEGaHJyPoxc&t=575s, min12+
-Avg income is important to the calculations, and up for debate. Happy to see your figures & adjust if needs be.
-Ive used an average discount of 25%, which IMO is on the low side. Consists of sentinel administrator with 20% worth of expertise + 3% Avg WuXing building discount + avg of 2% supervise construction assignment + no events/item combos
-this assumes industry/Commerce build, as I've been unable to scale peasantry builds/tech in *early/mid game* in enough provinces to compete with a blue/purple build. My take is that there are 15 industry/com provinces that scale to 9-12k income p/turn vs 10 peasantry friendly ones which scale to 6-12k, but more slowly (peasantry builds cost less, but pop growth to 9.5m takes 30-80 turns with different builds)
-turns to build to lv8/t5 buildings is interdependent on tech pathing, public order, funds available, starting state, characters, events, wars etc
-there will be some provinces that don't fit into this model: silk/spice, flex provinces that have no clear best use, temple/horse/tamer provinces. As per above guide is trying to reduce complexity to give you a general approach, so it's not optimised for those areas.
-remember that administrators give a -X turn buff to construction time & every +1m people gets you a +1 build slot in a province.
-assumes blue & purple assignments for at least 1 of your commanderies (+% commerce & +% industry)
-mega province build is:
https://imgur.com/a/8fZ9F8Q
-Purple buildings give discount to blue buildings, so construct that first. Flat income buildings come before percentage based ones, until city level 5-7
-public order, this build assumes 1-2 public order tech reductions, and sentinel administrator with Stability. Add speciality buildings, upper court traits/items, and garrison for more PO.
-you generally wouldn't put an administrator in a food/income province, hence n/a
-im not suggesting that you never put money into income provinces, merely that megaProvinces are first priority & cash is scarce early game.

Corruption:
As you grow, corruption will become a problem at an increasing rate (this is the game's anti-snowball mechanic). It reduces upto 90% of a provinces income if you take no action. three points:

1)Corruption costs increase exponentially through the game:
so your effort at mitigating corruption should be similar - spending 20k on a coin maker in your first 20turns will net you almost no return
2) Payback period for corruption reduction:
on average reducing corruption on your 500 gpTurn farming province isn't a great payback & reductions that are factionwide have a much bigger impact than localised ones
3) Suggested approach:
for <7 commanderies:
add administrator
1st yellow corruption reduction tech
heir with 'honest' trait
(cumulative -40% corruption)

for 7-20 commanderies:
boost emperor satisfaction (-10% or +10corruption factionwide)
general satisfaction (-10% or +10% factionwide)
2nd corruption tech
(-64% corruption)

for late game:
copper mine tech + 2 or more Lv5 coin mines
build lv4 Private Workshop/Coin Maker (purple) in provinces that border your MegaProvinces
Lv4or5 Admin building in megaProvinces (office of seals)
discourse of states item on leader or heir
Stack upper court w/ corruption reduction
(over 100% reduction)

Income enables you to obtain more provinces - you'll need to use your judgment between:
- spend upto 10k recruiting + 3.5k p.turn on an army, muster for 1-4 turns, take a territory, then disband the army or take next province. Given the setup costs, it's best to take a few provinces before disbanding to make the army creation worth it. If it took you 10 turns to conquer 1 commandery (2-3 settlements), how many turns will it take to repay the 45k you've outlayed? It could be better to....
-spend 3-50k on buying a territory (<1-10 turns worth of income, can be cheaper than raising an army, or dealing with coalition war that comes afterwards). Note factions will not sell their commandery capital, the ones with a star
- build buildings, at an average payback of 9-30turns, which lets you do more of the above two strategies
-I find most 'Spend income on this event/council proposal' to be much worse return on income than above three methods.


***Fighting:
Army composition: there's lots of debate in the community, feel free to Google. To shortcut needing to research each of the ~100 units in the game, try:
1 champion with 4 spear guard in shield wall out front of army + 2 ji infantry held back for anti cavalry (replace 4 units with defenders of earth late game)
1 strategist with trebuchet + 5 highest class ranged unit you can get (militia -> archers -> Onyx dragon or unique unit or crossbows) & shoot enemy with flame shots
1 commander/vanguard with 1-6 spear militia horsemen for anti cav + anti character+ infantry flank charges. Replace with shock cavalry late game

Rationale is that legendary computers get stat buffs, and it's *much* more efficient to shoot buffed units & force morale loss with flaming weapons than melee + cav charge to win.
- See Onyx Dragon rush for nice synergy with Commerce focuses factions
-about half the units in the game are compararively bad & should not be made (there's a better version of the unit, it's too expensive for the stats, or locked behind a tech tree advance with too high an opportunity cost)
-question for readers, is there some CaoCao Strat that's enabled by getting 50% more techs than most factions?)

World Map. In legendary mode,
1) You have a morale disadvantage (~8pts?) which becomes very hard to mitigate if the enemy has
-night battle / ambush (-15 morale)
-flaming arrows (-7 morale)

2)someone will declare war on you every X turns, and send stacks of armies against you to their own detriment (eg lose territory just to throw armies at you). Respond by setting ambushes in front of half stack 'bait' armies (still in reinforcement range), or attacking with multiple stacks reinforcing. Opponents will use night battles to cut off your reinforcements, so be careful how you arrange you troops at end of turn.
In short, the military adage applies: if you find yourself in an even fight, your doing it wrong

3) Movement speed of Han armies in Nanman lands. Naman move about 2x your speed - meaning that they bring more armies to a specific fight than you & pick off your armies very easily. Only way Ive found that works is to either encircle them with 4-10 armies on ambush OR control very specific chokepoint. Aim is that they never initial a battle that isnt an ambush. Either way its time consuming & expensive (+20 turns to conquer equivalent lands than fighting in North or East)

***Other
Reform Tree
Tech tree in this game is visually well designed, but is a bottleneck for most factions. You mostly can't speed it up how fast you tech, and need tech for this strategy. Think about what you need in your first 100 turns, and pick the 20 techs in advance.
-until you get a feel for the whole tech tree aim to progress to certain techs that are very valuable. Number of techs needed in brackets, many techs need certain buildings to allow research:
~10 Early game: ,+1 Trade agreement (1), +1 administrator (1), +25 to +100% food (upto 5), unlock inn upgrade & coin mint (4), corruption reduction/PO buff (yellow, 2 techs)
13 Mid game:
-8% recruitment costs (1), corruption reduction/PO buff (yellow, 2 techs), Onyx dragons (5), +45% Industry/T5 industry (4), +10% replenishment (5),
16 Late game: Tier5 Admin Bld (3), shock cavalry (1), Coin minting (4), T5 Industry bld (3)
More detail here: https://forums.totalwar.com/discussion/260574/my-tech-tree-overview-biased

Population
Adding population in your megaCommanderies is IMO free money / snowball speed increase. Use purple labour building or
1.5m people = 50% all income buff, -2build turns with administrator, -6 PO
2.5m people = 75% all income buff, -3? build turns with administrator, -14 PO
Mix and match these to offset PO debuffs: Administrator with Stability (+5 PO) + lv1 yellow tech (+2 PO) + lv3 yellow tech (+3PO) + lv2 yellow tech (+5PO)
When you have enough PO buffs, you can increase your tax rate by +10% (-6PO)

Spies:
give it a pass until you've got a handle on the game. read some guides if you're really into the topic. Note that it got buffed by the game devs b/c it was not very useful

Characters. Until you want to spend several hours researching various traits, then
-avoid greedy, disloyal, ambitious, reckless
-look for honest, humble, brilliant, wise, tranquil, brave, vigilant, creative, clever, bright.
-advanced players: individual traits range from net -2 to +12 attribute points. Aim for heroes with a sum of 10+ of the attribute your looking for (expertise/cunning mostly). Characters with 100+ (needs checking) attribute points in a class become legendary & get 'wounded' instead of dying if losing a battle
- Faction leader traits are handy, but by mid/late game is much more important to have multiple army stacks than +10% melee evasion or +15% ranged damage. Use the upper court (PM, leader, heir) to offset any issues your having (corruption, PO, satisfaction, redeployment costs)
For more details, see https://www.reddit.com/r/totalwar/comments/hwlxbz/ever_have_a_hard_time_desire_who_will_be_your/

Satisfaction / How big should your court be?
-Early game: aim for no one to be idle (legendary has a satisfaction penalty so it's a lot of effort to hold on to surplus characters. Secondly, you don't want to miss cost reductions assignments). Don't allocate lower court ranks unless you have no other choice (cost outweighs return early game)
-Mid game: hire 1-7 sentinel administrators (skill the bottom row for economic gains) + generals. Use titles (+10 to +40 satisfaction) to increase low satisfaction generals you want to keep until you find a permanent solution (administrator/lower court/friend/prestige buffs). Note 'low satisfaction' causes a faction wide corruption debuff. It's ok to fire generals if you can't hold onto them
Late game: 15 army stacks means you need 45 characters just for generals!

Weapon / item / ancillaries combos.
These are pretty interesting, but theres a lot of RNG, so just use as stat boosts until you want to drop a few hours of your life into what sets are optimal with which heroes / roles.

Commanderies:
-Commandery tab on right of screen will save you a bunch of time once you get to 5+ commanderies, even if it takes a bit of getting used to.
-This game largely doesn't reward 'recruitment' provinces like some other TW games (horses the exception). Food is for early game trading, then enabling tall builds on your mega provinces
-1-2 farm commanderies support 1-2 lv8 cities, depending on your green techs
-if your faction creates an emperor seat, move your capital to a mega commandery for the % boosts before you rank up (not included in above calculations)
-Should you upgrade your megaProvinces to lv8/lv10?
Lv8 yes, the 6th build slot pays for itself either as income buff or corruption reduction
Lv10: Late game maybe, in 1-3 Commerce focused commanderies. At this point of the game you should have 200k-1m gold, so money is kind of irrelevant.
Difference between Lv8/10 city from Lv7:

	extra Food	extra expense / payback turns	extra population	extra allIncome	extra Public order	extra slot	Com/All income gain from buff	for peasantry build
Lv8	-8	8k 11turns	+1m	+15%	-14?	+500 (marketplace/ labour)	225
Lv10	-20	30k 18 turns	+3.5m (~)	+35%	-20	+500 (marketplace/ labour)	1125	(+300% of ~700 base gold)

'Establishing Order' in empty territories:
Generally, don't. Easier to let other factions spend the 30k/X turns & then conquer them if you need them. Think about the payback period of your army's salary vs building it yourself.

Rushing buildings (coin icon on build menu)
Generally no (because it reduces return on building investment), but situationally useful for:
-farms, to stop going into faction wide food deficit
-settlements, for incoming sieges
-anti corruption building, if they touch a lot of provinces
-upgrades that take more than 5 turns, particularly city upgrades.

Rank upgrades
-Early game, get extra administrators & assignments. You get bonus armies from extra administrators. Trade agreements are situational
-Late game. hover your cursor over the end circles for each category of bonuses. You get a buff for finishing any given row. The -15% army upkeep is a huge bonus, -25% character salary is nice but by this point you won't care about 1k gold per turn.
-as above I don't use spies

Counters
Mid game snowball is weak against an early game rush from multiple opponents. Will leave it to posters to suggest defenses against this

***General recap, your strategy should be:
-survive the scripted events for your faction's first 20turns
-obtain & scale 1-5 mega provinces to city lv7/8, while keeping 10+ surplus food
-manage corruption
-then hire 15-20 army stacks & take territory / emperor seats
If your new-ish to the game, try Sun Jiang, conquer south then east

Have fun.

[Kitten Food]

Great guide and tips! When 3k originally came out, I played around 40 hours and left. Came back because of Serious Trivia's videos and now have over 1k hours. Game gets better once you understand the mechanics. Can't wait to see what CA does with the next 3k game!

[Darklordnj]

Did you mean to post this in the guides section of steam?

[Herobrine]

i agree

[Bleiser]

So this doesn't work for 1.7.2?

[E.R.L.]

Originally posted by Bleiser:

So this doesn't work for 1.7.2?

Yes it does. 1.7.2 doesn't change any gameplay or mechanics.