Legendary difficulty - dealing with late game hordes?

[Jacobin]

Playing as Chokosabe I conquered the western half of the map and took Kyoto but then found myself facing the Uesugi and the Ikko Ikki who were allied and assembled a vast horde of six or seven full or near-full stacks and sent them against Kyoto.

I had a killer stack of elite samurai and monks commanded by my Daimyo and a support stack of heavily upgraded ashigaru standing outside and the equivalent of a full stack in Kyoto itself and if they had assaulted the citadel would have easily beat them thanks to the time limit.

But they didn't attack but besieged and when I hit the attack button I was outnumbered 3-1 with almost all the Uesugi stacks being samurai and the Ikko Ikki mostly being Ronin and the auto bar gave me about a 10% chance of success.

I bottled out out and sent the two stacks outside Kyoto away to attack other enemy provinces in the hope that they'd either detach stacks or launch an assault which I could still defeat - but they did neither and just starved Kyoto out - and then sent the same horde to take out my fortresses one by one, in every single case starving them out so I couldn't even inflict any damage on them in siege battles.

Eventually I gave up on this campaign but I wonder whether there was any way of beating this horde?

The Timurids in MTW2 posed a similar problem - but by that point in the game even at VH/VH I could afford multiple elite stacks myself and to just sacrifice one stack after another until eventually they were wore down.

However well you manage your resources STW2 at Legendary does not give you that option.

Now fighting without a timer I could have sacrificed all my three Kyoto stacks and done enough serious damage to reduce the threat - but at Legendary without the radar map or the option to issue orders in pause or at slow speed handling reinforcements is a nightmare.

I could also have used my garrisons (which were usually full stacks) to suicidally attack at the beginning of each siege and aim to take out a large part of at least one besieging stack - but this would just have speeded up their advance as rather than taking 4 or 5 turns per fortress they'd be done in 1.

As suggested by Maltz in the Avoiding Realm Divide thread I could also in theory have used agents to raise rebellions - but as I'd focused on economic and military development my agents were hopelessly outclassed and outnumbered and rarely survived more than a couple of turns before being taken out.

Sending a geisha or ninjas to assassinate their generals wouldn't have helped much either as I find the biggest impact of an AI general is in auto resolve which can't be used against a horde this size - and unlike the original STW you can't destroy a faction by killing every family member.

So any other suggestions?

[Maltz]

If you are not opposed to the most powerful exploit on Legendary: Low-level agents are not a problem if you are determined to find the right random seed to make them lucky. Find any castle to biege with anything (like a Light Cavalry). Every time you choose whether to assault or continue to siege, there is a new autosave generated.

(1) Generate new autosave.
(2) Agent action. If succeed -> (3a). If fail -> (3b)
(3a) Generate new autosave. Backup the autosave in the save folder.
(3b) Load autosave generated in (1), generate a new autosave.

After you are done you can break the siege and pull the unit back, or play chase on the field. Light cavalry is perfect for this.

Before Realm Divide, it would be better to make one powerful ally (such as Uesugi) by gift/alliance/hostage and pull them out of their alliance with Ikko-Ikki.

Note: Low level monks can't recite rebellion, but it is very easy to take them to level 2/3 by army demoralization.

[Jarmam]

This sounds very, very odd. If you control the western half of the map with Chosokabe you should be showering in gold. When I had the two islands alone I was so filthy rich I paid Tokugawa to fight the Takeda/Hojo/Imagawa alliance despite him being 5 provinces to their 30+. On half the map (or, say, 25+ provinces) you should be fielding at least 4 stacks of absolutely top-notch units with the best bling in the game. Ashi are a waste of gold in the middle and late game as their upkeep is hopelessly disproportionate to their effectiveness. Even with rifles.

You need to make sure you develop your infrastructure, even if this takes time you'd rather spend bashing guys in the head. This is particularly the case if you stay Buddhist as Chosokabe as the faction is almost tailor-made for an easy early Christian conversion with all the subsequent riches it provides. Im not sure what you mean with "However well you manage your resources STW2 at Legendary does not give you that option". Apart from simply sabotaging enemy stacks that are closely together and picking them off one by one (ninjas are stupidly, stupidly overpowered for this reason alone) you can simply out-produce the whole enemy alliance if you focus on development from turn 1 (which you should since Chosokabe gets 5 isolated provinces basically for free). Even as a Buddhist you should be able to secure all 5 southern trade posts and have a stable grip on the Shimazu island without any hassle. This means you can do it with primarily Ashigaru - freeing up your own island's provinces for a 5 merchant-guild 5xMetsuke r4 blitz. Since economy is exponential it makes a world of difference whether you do this early or middle game. With that kind of cash, well... you can BUY the enemy stacks by the time Divide sets in.

[Maltz]

I just went to Excel Spreadsheet and crunched some numbers. It is better to leave Markets in the first level, but only upgrade them to Rice Exchange and above in your richest provinces (very fertile soil) where you station a good Metsuke. The extra food consumption is not worth it.

Practically, though, you might end up building Sake Dens instead of Markets in most provinces, as you want to cut down your upkeep cost.

This is what I did in my Chosokabe campaign and it worked great:

(1) Highest level farm upgrade in all provinces (for very high town growth) as early as possible.

(2) Ally with nobody and trade with everybody. If you have all six trade nodes, even without any trade partner, the unsold resources still earn you 10,000 koku per turn.

(3) Northern Kyushu has 3 very fertile provinces: Upgrade Markets + Metsuke (lv 2 from Tsukushi) gives about 38% tax rate. By the time I finished the campaign, these three provinces alone generates almost 10,000 koku income under that fat tax rate. That's half of my clan's tax income. (The number you see on the campaign map is not your income, it is the production value before you tax just a fraction of it.) All other provinces builds Sake Den (don't station your general in the castle town or they get negative attributes).

(4) Pause for a long time before Realm Divide. Disband any unit that has been outclassed and fill the vacancy right before you trigger the Realm Divide. This low-maintenance saving periord should give you at least a hundred thousands koku.

(5) Once RD triggers, forget about building/upgrading economic infrastructures. Use the money you save to build happiness structures in new provinces and hire units.

I actually think Ashigaru is great if you mass hire them during special events in the appropriate provinces. If you build a Castle (an upgrade of Fortress) in Satsuma, you can easily hire level 4 Yari Ashigaru with +5 armor during the "Ashigaru master" event. Worthy to mass produce them. Likewise you can easily get level 4 Bow Ashigaru with +25% accuracy (more practically, +15%) in Buzen during two events. They are all beasts! (drooling)

[econ21]

Was your general a night fighter? If so, he could attack one stack in isolation without adjacent ones reinforcing. I think it's probably the single most important late game general skill (the one reducing siege times coming a close second) and so I gave it to all the fighting generals I could.

Sabotaging an enemy army can also prevent it reinforcing, but I found this a little more glitchy (sometimes they were sabotaged and yet did reinforce - never figured out why).

You would still suffer attrition in one on one fights and if you did not have a big qualitative advantage, it would be costly. But night fighting can be a way for a single elite force to thrash multiple inferior enemy stacks, as veterans of Rome Total War: Barbarian Invasion can testify.

Perhaps as you did, I found the AI rather passive when faced with a single elite player army - it does not seem to want to try to use its multiple stacks to overwhelm you. So you may have had the time to dismember their horde at your leisure.

[Jacobin]

The problem was indeed that I did not totally prioritise economic development from turn 1.

I've since tried an Oda campaign where I demolished everything that isn't a market or a sake den+upgrade, only focused on chi arts and spammed ashigaru.

I also exploited selling military access and the move your trade ships off the nodes before asking for a trade agreement trick.

Together these do indeed make a big difference.

I do however draw the line at using autosaves to try again and again and my PC doesn't like me switching windows out of STW and tends to crash the game so making a copy of each save is a laborious process...

The night fighter and ninja sabotage is indeed the way to go if this happens again.

[Jarmam]

I just went to Excel Spreadsheet and crunched some numbers. It is better to leave Markets in the first level, but only upgrade them to Rice Exchange and above in your richest provinces (very fertile soil) where you station a good Metsuke. The extra food consumption is not worth it.

I wholeheartedly disagree. Waiting for a very rich province requires you to actively force yourself out on the map as Chosokabe. If you rush a high lvl market on your starting island it will generate so much revenue and +town growth that it outclasses waiting. Theoretically the best thing to do is to never upgrade any markets at all (even in metsuke towns), but this is optimal at turn 200. You have already won at turn 200. Doesnt matter. What you want is a foundation for the middlegame as this is where the game is won or lost (except with Oda). Sure, you can rush the northern provinces, but there are still only 3 very rich provinces. There is no reason not to get markets asap on your own island to unlock Metsuke and start town growth and there is no real reason not to develop at the very least 2 of them as the absolute top priority (maybe 2nd priority without Christian battlecruis... trade ships).

It is however never a good idea to upgrade markets outside Metsuke-towns at any stage of the game, but there cannot be any doubt on this.

I actually think Ashigaru is great if you mass hire them during special events in the appropriate provinces. If you build a Castle (an upgrade of Fortress) in Satsuma, you can easily hire level 4 Yari Ashigaru with +5 armor during the "Ashigaru master" event. Worthy to mass produce them. Likewise you can easily get level 4 Bow Ashigaru with +25% accuracy (more practically, +15%) in Buzen during two events. They are all beasts! (drooling)

I have to question this. If you can get +5 armor on Ashigaru, you can get +4 melee/+2 armor on (insert naginata/katana samurai or better of your choice) or +5 armor monks from the same province. Satsuma is in a terrible position and works directly against the strength of Ashigaru; replenishing lost units from anywhere for a low initial fee. Ferrying them long distances will increase the punitive penalty of their horribly overpriced upkeep and lose the effectiveness of their super-fast reinforcing strength. Also Ashigaru gain experience wickedly fast in combat if led by a decent general (those that survive anyway). Hiring primarily Samurai will also intice the +2 xp for Samurai-units event, which is massively powerful if combined with high tier unit structures due to the exponential xp efficiency.

Ashigaru cost roughly 1/3 to 1/4 the initial price of a samurai or monk unit, but 1/2 to 1/3 of its upkeep. Fielding 3 Yari Ashis costs the same as 1 Katana, but having them around costs 1,5 Katanas per turn. That's not an insignificant difference, especially if you decide to pour resources into developing the units either via upgrades or good generals. This is of course utterly irrelevant earlygame as the cost of developing the infrastructure for Samurai will vastly outcost the price of just using Ashigaru (plus you're less likely to cap out 20 units early on, especially on the defense) but in Divide times you should be well prepared to ditch ineffective units. Especially if you're struggling with overpowering enemy hordes.

I havent tested Bow Ashis with +accuracy to test if they're any good compared to real archers so I cant comment on those.

[Maltz]

There could be a few common misconceptions about the tax system in STW2:

(1) All the Commerce money from Markets (200, 500, 1000...), Sake Den (100, 200...), etc. is NOT what what we actually get. We only tax a fraction of it. For example, the 500 extra commerce from Rice Exchange is only +150 at 30% tax rate before admin cost. So after admin cost it only generates 100 extra koku. Building a rice exchange only breaks even after 15 turns.

(2) Town growth is subject to reduction from tax rate, too. At 30% tax rate, the town growth is reduced by 30%.

I have seen (and probably made) many arguments that sound like "this works for me, so it must be true". I actually agree with it - whatever works. On the lowest difficulty level, maybe everything works. What I propose works on Legendary. But maybe even that is not an optimized strategy. So I use the Excel spreadsheet to really go through all the calculations. For the market upgrade argument, there are a few factors that changes the decision:

(1) Metsuke (effective tax rate). Having a high effective tax rate on the province favors having a high-level +income building.

(2) Number of provinces. Rice exchange gives 10 town growth. So it breaks even exactly at a province count of 10. If we have 22 provinces (the pause time before Realm Divide for a domination game), +1 growth for 22 provinces is better than +10 growth for 1 province. Naturally, a shorter campaign favors upgrading the markets because we pause at fewer provinces.

(3) Number of turns. In most scenarios, within the first 20 turns or so, not upgrading Market is always better (due to the initial cost of the Rice Exchange vs. nothing). Then upgrading market wins during 20-40 (it really varies in different scenarios) turns since the extra tax beats the extra town growth across all provinces. Then not upgrading market wins again. So playstyle matters here.

So my long-version conclusion is:

(1) Upgrade Markets if you:
- Use Metsuke in that province
- Play a short campaign
- Do not like rush, do not like to take things slowly

(2) Leave as Markets if you:
- Do not use Metsuke in that province
- Play a long campaign
- Likes to finish as quickly as possible, or likes to take things very slowly.

***

And the Ashigaru argument... I guess play style also matters. Let me try to make a similar comparison:

(1) Use more Ashigaru if you
- Like to finish the game as soon as possible
- Use agents to reduce enemy's morale
- Only fight a battle where you can win, avoid all battles you cannot win
- Exploit AI's battlefield weakness (bait and trap, etc.)

(2) Use more Samurai if you
- Like to take things as slowly as possible
- Do not use agents frequently
- Often fight battles at a disadvantage and hope for a heroic victory
- Charge AI head-on

But the bottom line is whatever works.

[Jarmam]

I think we pretty much agree on the market thing. Im being a bit vague in my argumentation - the reason I prefer merchant guilds fast (as Choso and only as Choso) is that rushing them on your island is 100% safe and 100% effortless as you dont need any samurai structures to take your island and commence building a fleet to secure the entire southwestern trade profits. That means they're ready to spit out their (commerce*tax%) right when you invade the Shimazu island and have to start developing military cities (Buzen and Satsuma usually) and the 3 rich provinces (if you invade "early"). The +town growth and super early metsute just mean that its less ineffective - obviously a richer province will spit out more revenue than a poor one as town growth only really takes off in merchant guild cities that arent being flattened by high taxes. Its a timing thing and what I wanted to throw in was that optimal development isnt necessarily the optimal choice, even discarding playstyles. I dont advocate doing it because its my style of play but because I honestly think its the most effective way of going about winning the game.

And the Ashigaru argument... I guess play style also matters. Let me try to make a similar comparison:

(1) Use more Ashigaru if you
- Like to finish the game as soon as possible
- Use agents to reduce enemy's morale
- Only fight a battle where you can win, avoid all battles you cannot win
- Exploit AI's battlefield weakness (bait and trap, etc.)

(2) Use more Samurai if you
- Like to take things as slowly as possible
- Do not use agents frequently
- Often fight battles at a disadvantage and hope for a heroic victory
- Charge AI head-on

But the bottom line is whatever works.

Absolutely. Whatever works works, this is also true on Legendary especially with an easy faction like Chosokabe. What I mean is the OP sounds like he's struggling with being overwhelmed by superior numbers of superior troop quality. If he starts fielding Ashigaru from distant provinces my claim is this will be counterproductive to combat this problem. Thats not to say that Ashigaru are total garbage. In fact you could argue that if you play an Ashigaru-heavy style (Oda is again the prime example) a good counter to hordes is to simply outmanouver them with twice the stacks of well-equipped Ashigaru and simply starve the enemy out by taking his provinces or isolating his stacks etc etc. Its a playstyle choice, but my questioning was related directly to the OPs concern and not so much to the effectiveness of Ashigaru in general (though I still claim that people grossly underestimate the punitive upkeep cost on Legendary when it comes to lategame stack-vs-stack races on the whole map). I made this mistake when playing as Ikko the first time and grossly underestimated the absurd value of Loan Swords and Yari Ronins. Playing as Hattori will teach you about upkeep the hard way (aka you pretty much have to rely on monks in the middlegame).
- Only fight a battle where you can win, avoid all battles you cannot win
Ashigaru's primary strength comes from A. Cheap initial cost and B. Fieldable from any province in the game. Dont be afraid to lose Ashigaru units or stacks - thats exactly what they're good at. Dying and being replenished.

I might sound like Im trying to get in the way, but what I am concerned with is fixation on simple concepts like "never upgrade markets outside provinces x" and "always rely on unit y". Its good to have a solid handbook, but with certain specific problems I believe in searching for solutions in any way possible. Even if it means +5 armor Ashigaru, I guess they make excellent cannon fodder ;)

[Jacobin]

I am now playing a successful legendary campaign with Oda am in the 1580s and have conquered the eastern half of the map without raising a single samurai or cavalry unit - upgraded Oda ashigaru stacks are beating any samurai-heavy stack that's thrown at them.

In the interests of verisimilitude I'd like to add a few samurai units to my offensive stacks but have simply never had the money or time to raise them - and with Oda ashigaru so cheap it makes no sense to raise one samurai unit when you can have 4 yari or 3 matchlock A's for the same price.

Having said this I've yet to experience the horde effect in this campaign.

Excellent cost-benefit analysis of the markets BTW.