Data: Market versus Ninja

[Leptomeninges]

I'm never sure the five star metsuke argument is a good one to make for maximizing the market chain. But it's entirely possible I'm missing something. Here's how I look at it:

Assuming you have the tech and provines, to the best of my knowledge the highest you can start a metsuke is with three stars. In my experience I'm usually at mid game when I reach this point and most of my metsukes have spent the first third or half of the game in the 1-3 star range.

Unless you exploit a bit (chain revolutions and bribing the rebels) the leveling time in cities is prohibitive to go from three stars to five. As I recall it's ten game years to go from two to three stars, 15 years to go from three to four stars, and 25 years to go from four to five stars. So assuming you have the tech and provinces to launch a three star metsuke around 1550, you're still looking at 1590 before you hit five stars. In all honesty, I don't think I've ever had a campaign of any length or difficulty go past around 1575.

So yes, by end game your five star metsuke will be churning out tons of cash for you. But at that point (assuming you're still playing) I suspect the game is in sandbox mode and the cash is irrelevant except as a curiosity to demonstrate what can be done.

I think a better evaluation would be of a three star metsuke versus the food loss from building the market chain. I suspect that if you averaged the metsuke level over the course of the game it would come out somewhere in that ballpark. Or better yet, for some enterprising math whiz to show how fast the metsuke would need to reach five stars over the course of a game and how long the game would need to last, to make the market chain worthwhile in those five cities. The whole thing is obviously dependent on empire size, how fast you're expanding, and the value of the top five cities in your empire which further complicates the modeling.

[HopAlongBunny]

...you're still looking at 1590 before you hit five stars. In all honesty, I don't think I've ever had a campaign of any length or difficulty go past around 1575.

Largely agree, but it isn't that hard to get 4-star metsuke fairly early in the game. You have to be aggressive, and you need a co-operative opponent. Apprehending enemy agents can level them fast, but requires an AI that invests in agents.

Early game (if I remember) I march my metsuke all over my neighbours lands. Even agents that cower inside their forts are fair game.

Leveling them like this can lead to war

[RedKnight]

Haha Lepto, right you are... I think that any number of things are a challenge to consider, given the short time frame of this "TW lite" game. Not lite in being a great TW game, but lite in terms of forcing games to be shorter, which also means, forcing players to deeply consider their actions - like, the usefulness of levelling metsuke , or any other agents, versus the gain.

I and other CMs are working on improving the Org TW wikis so they can storehouse all the best stuff from these forums. I hope we can some day soon see a TWS2 page about "agent levelling versus clan growth versus realm divide".

[NinjaElk]

Your spreadsheet seems to assume a 100% tax rate. 100 wealth does not directly translate into 100 gold per turn. The wealth is multiplied by the tax rate to arrive at your gold per turn. So 100 wealth from a Sake Den at a 30% tax rate yields 30 gold per turn.

On the Food Surplus vs. Rice Exchange debate, I think the base mechanics are also being misunderstood. As far as I understand, this is how it works (please correct me if I'm wrong):

1 Food Surplus yields 1 growth clan-wide for all provinces. Each 1 growth increases the town wealth by 1 per turn. Town wealth is then added up from all various factors (farms, buildings, total current growth) then multiplied by the current tax rate to yield gold per turn. For example if my town has a total 1000 composite wealth, at 30% tax rate it generates 300 gold per turn. If it has a 10 growth, then next turn it will be at 1010 composite wealth, and generate 303 gold per turn at a 30% tax rate.

A Rice Exchange yields +5 additional growth over a market, and +300 static wealth to the province once built while costing one food. Whether or not this is worth it depends directly on how many provinces you own. If you have 10 provinces, 1 Food Surplus yields 10 total growth clan-wide, while the Rice Exchange yields only 5. Granted it'd take 10 provinces 60 turns to recoup the static +300 bonus the Rice Exchange gave you for existing. However, you also have to account for the initial 3,150 (before modifiers) initial cost of the Rice Exchange.

When you throw Metsuke into the mix it starts to get complicated. Let's assume a Metsuke grants you +15% tax rate, thus bringing a 30% tax rate to 45% tax rate in a single city, and as in the above example we still have 10 total provinces. This means we're comparing the worth of +300 initial static wealth and +5 growth per turn at a 45% tax rate, to +10 global growth at a largely 30% tax rate.

Now, this is where it gets complicated. 45% taxes produce 50% more gold per turn than 30% taxes (i.e. 100 wealth at 45% rate is 45 gold, 100 wealth at 30% rate is 30 gold. 45 gold is 50% more gold than 30 gold). So, we can assume that +5 growth at 45% is really worth +7.5 growth at 30% tax rate. Because of this phenomenon, the higher your base tax rate, the less effective the static 15% boost from the Metsuke would be.

What does this all mean? In general if you have a lot of provinces you're still probably going to be a lot better off with just Food Surplus. Especially once you get to the point where you have 15-20+ provinces (or more if you're playing Domination) the clan-wide growth just trumps what a Rice Exchange can give you, even with the Metsuke's boost. Also, that frees your Metsukes up to go wreak havoc elsewhere. However, there are situations where the Rice Exchange/Merchant Guild beats out a food surplus, but it'd take a pretty complicated spreadsheet to arrive at the exact Tax Rate, Metsuke skill at overseeing towns, current number of provinces, and bonuses to building cost.

tl;dr
~10+ provinces: Food Surplus > Rice Exchange
<10 provinces: Rice Exchange > Food Surplus

[RedKnight]

Thanks muchly NinjaElk, you've posted a very good analysis. I have learned a lot since I posted that spreadsheet in the distant past Some simple basics are what they are - but then are overlaid by many other things going on.

Another thing to consider is, Administration costs. There is a good thread listing Admin Cost vs number of provinces somewhere, but it escapes me at the moment. I am hoping for a simple rule of thumb of how the number of provinces increase versus the efficacy of metsuke and ninja buildings.

Would you agree that once you reach a certain number of provinces (aka no longer small and/or face Realm Divide), there's no way you would want a Market built into anything higher in the Market line, with the possible exception of, where you'll place maxxed metsuke? If true, it begs the question of whether you should build anything past Market, except for the metsuke provinces, if you know the province will ultimately not be in your top 5 income producers. (Otherwise, why invest the money?) You can go a step further than that and burn down anything higher than a Market in a captured province (and replace it with a Market), if it's not in your top 5 producers...

Related to this, would you agree that the Market line, even if kept to a simple Market, is better than the ninja line of buildings, assuming you only have one slot in a province? After having 5 of any type of agent you want or need.


It's all pretty complicated and overlaps a lot. But there are a lot of smart people here who see things that I haven't.

TW rocks. Thanks! - RK

[Stratege]

When I look at RotS it seems so the devs do not believe in the vanilla economic system concerning food as well any longer. In RotS there is a market line that gives additional food instead of costing food.

I only wish they would change this in vanilla. I do not like that apart from some provinces that serve as recruitment centres upgrading castles and markets just kills income you would otherwise get in the long run. And there is no way to downgrade castles if the AI has upgraded them too much.

[Sp4]

A friend of mine recently did some extensive maths on the subject, or wants to anyways. He says market chain is awesome for growth and if you can maintain it, give a very high income in the long run.