Starting from the Ground Up...Literally

Unlike many contemporary fantasy RPGs, the Adventurer Conqueror King System has been designed with the view that many gamemasters want to be able to simulate an ancient and/or medieval world in such a way that their campaign world makes sense. The income of peasants makes sense in relation to the income of kings. The cost of swords and the income of the swordsmiths who make them has some relation. Treasure exists in more forms than simply gold, gems, and magic.  The easiest and best way to achieve a world with versimilitude is to start with historical assumptions. Let's start with some of the assumptions I used in developing the economic systems for ACKS. Since virtually every fantasy world is pseudo-medieval or pseudo-classical, and those economies are always agrarian, I started by creating some simple, but historical, assumptions for the rural communities of ACKS. Start with the core assumption that 1 medieval English penny  = 1 D&D sp; 1 English shilling = 12 pennies = 12sp; and 1 English pound = 20 shillings = 240sp = 24gp. 1 penny / silver piece could historically buy:

  • A chicken
  • 3 dozen eggs
  • 3 gallons of beer
  • ½ gallon of wine
  • A night’s stay at a decent
  • A loaf of bread
    • Wastel Loaf: 1 penny = 4 pounds (wealthy eat white bread)
    • Maslin (“Coarse”) Loaf: 1 penny = 8 pounds (whole grain wheat and rye)
    • Horsebread: (rye, barley, oat, pea, bean): 1 penny = 12 pounds

A "quarter of wheat" is equal to 8 bushels of wheat, or 480lbs (because 1 bushel = 60lbs) According to the Assize of Bread and Ale, when 3 gallons of beer sell for 1 penny, a quarter of wheat sells for 3 shillings 4 pence. Given that 1sp=3 gallons of beer, 4gp=1 quarter of wheat. (The math: 3x12+4sp, or 40sp, or 4gp). So 1 quarter of wheat is 4gp. 1 square mile has 640 acres. Historically, a man could plow about an acre in a day, and there are 2 months for plowing & sowing (March and October), so one man can plow 10 acres (1 acre/day x 4 weeks x 5 days) each month, twice, or 20 acres. If each man is allotted 30 acres, he leaves one-third fallow, and farms the other 20, 10 in each growing month. Therefore, 1 man can farm 20 acres, and manage 30 acres. Historically, a medieval farmer would plant 2 bushels of wheat in each acre. From this he would hope to yield about 5 bushels of wheat for every bushel sown, or 10 bushels of wheat per 2 sown. But if each acre sown would yield 10 bushels, the farmer had to set aside aside 2 for seed, for next year. SO his final consumable yield would be 8 bushels per acre. 8 bushels equals 1 quarter of wheat. So 1 acre of land yields 1 quarter of wehat. 20 acres at 8 bushels yields 160 bushels of wheat, or 20 quarters of wheat. Remember that 1 bushel = 60lbs; 1 quarter = 480lbs (approximately quarter of a ton). As established above, 1 quarter of wheat sold for 4gp. So at the market, 20 quarters of wheat yielded 80gp. So 1 man = 20 acres = 20 quarters = 80gp. Each man thus produces 80gp of wheat. (remember 1 acre=1 quarter = 4gp; 1 man = 20 acres). Each man has a household, including a wife and 3 kids. A historical peasant ate 3lbs of bread per day (1 loaf). 3 pound loaves of horsebread cost 2.5cp. Assume a total of 1,000 loaves per year. That's enough to provide about 1 loaf for the man, 1 loaf for the woman, and .33 loaves for each of 3 children, per day. The cost is 25gp per year. In addition to bread, each day the man eats 2 whole eggs (1cp) and 3 pints of ale (1cp), for 2cp per day. The wife eats 1.25cp per day of food and the kids .75cp per day each, for a total of 5.5 per day. 5.5cp x 365 days = 20gp per year. Each household’s woman produces 80gp per year in domestic production (equal to the able-bodied male), and each of the children produces 10gp in value (1/12th the able-bodied male), collectively 110gp per year. Thus, each family of 5 produces 190gp of economic value while consuming 45gp of economic value. Most peasants lived at subsistence level, with excess money ending up in their lord's hands through various taxes, fees, and rents. Leaving aside the 45gp each family subsists on, then, each family is worth 145gp per year to the lord, or about 12gp per month, or about 12gp per 20 acres per month, or 7.2gp per acre per year. Thus in ACKS a peasant family produces around 12gp of economic value per month to their lord, and each peasant family has a subsistence income of around 3.75gper month.

First, I want to say that I love what you’re doing here–this is really interesting, and I also feel the need for a campaign setting that hangs together in a deeper way than traditional D&D economies really support. That said, I’m wondering whether this sort of back of the envelope calculation actually works. The 80 gp/male farmer (+80 gp per female farmer in domestic production) seems plausible, although that assumes a wheat crop, which I believe was a more expensive cash crop and not suitable for all areas (what about people who live in a sheep-herding area, or an area more suited to oats as a crop? It may actually produce more food, although it produces less money). But I’m much less sure about how this translates into baronial wealth. First, you assume that the only consumption that the peasants do is eating food. But surely some of the domestic production (i.e. making and repairing clothes, likely converting flour into bread (and thus adding value), etc.) also gets consumed by the peasants as part of their subsistence. Man does not live by bread alone, after all, nor even by food alone. Second, you assume that the lord extracts all of the surplus–but doesn’t some go to the miller, and to the blacksmith, and to the local priest, and so forth? I suppose you could argue that some of that is “baked in” to the cost of the bread: bread takes as inputs wheat, milling, and baking (mostly, plus a little other stuff). But if the bread has a value of 1 sp per 12 pounds, and the farmer’s 12 pounds of wheat is worth 1 sp, then the bread has the same value as the unprocessed wheat. (This is presumably in part because of assuming that the peasants are growing wheat but eating horsebread.) So the miller can’t be getting his living out of the difference in value between what the peasant grows and what the peasant eats. Maybe this is just a rounding error, in light of how much of the population is peasantry, but the division of wealth between the local manor lord and the relatively few prosperous non-nobles is an important question for these sorts of baronial calculations.

Anyway, I’m curious how you resolved those sorts of issues. My understanding is that constructing accurate representations of economies is a challenging modeling problem, and I salute you for attempting it, but wonder how accurately it works.

Adam, thanks so much for your detailed response! You raise good points. Some thoughts below…

You are correct that I’m assuming a wheat crop. I chose wheat because I was able to find the most available information about wheat prices, and it was a staple crop of both the ancient and medieval western world. In the game rules, to account for variability in production, I have a variable called “land value” that will vary the production per peasant family from 10gp per month to 15gp per month.

The non-agricultural peasant production and consumption is assumed to be handled by the other members of the family - the economic value of which ends up within their own family, and is therefore cancelled out. The simulation works at the manorial level, not at the peasant household level, so stuff that’s intra-peasant is largely ignored.

Likewise you are correct that I’ve oversimplified in that I’m assuming the lord extracts all the surplus. Complexity gets reintroduced because the lord has to pay out a percentage into tithing to the church, support for retainers, maintenance on his stronghold, support for his troops, and festivals and holidays.

Assume a lord with a 1,200-acre farm, with 60 peasant families (1 family/20 acres) generating 12gp/month for him, or 720gp/month total. Maintaining the manor home will cost him 100gp. Garrisons will cost 120gp per month. Tithes will cost 72gp/month. Scutage and taxes to his own lord will cost 144gp per month. Festivals and holidays will cost 75gp per month. The total is 511gp per month. So the lord nets about 209gp per month. At our historical conversation rate, that’s about 8.8 English pounds per month, or a little over 100 English pounds per year of income. According to Daily Life in Medieval Europe, that’s right within the range for a minor baron.

Using the rules in ACKS, I built an Excel model of medieval England, with a population of 2.3 million spread across 10,000 manors (which are roughly the historical values). Here are the values the model produces, compared to historical values:

Knights: 36L (ACKS), 20-30L (Daily Life in Medieval Europe)
Barons: 50 - 500L (ACKS), 100-200L (DLIME)
Earl: 500 - 1,400L (ACKS), 1,500L (DLIME)
Duke: 1,400 - 4,500 (ACKS), 4,000-5,000 (DLIME)

Hmm. Interesting, but untrue. This discussion might help you.

Realize that you cannot say things like “Each household’s woman produces 80gp per year in domestic production (equal to the able-bodied male)” or almost any of the above without adding something like “among the shepheards of the Blackforest region during the period of 1200 - 1275”

This kind of socioeconomic info will vary RADICALLY even between seemingly similar climate regions, cultural preferences, tech developments, social organization, land distribution, time periods, sun exposure, etc., let alone the range of differences one gets between mountainous, forested, arid, etc,

Daniel, thanks for your thoughts. Your comment has me slightly confused. You start by saying it’s “untrue” but then suggest maybe it’s true among e.g. “the shepherds of the Blackforest region.” I assume what you are saying is that they are untrue as broad generalizations but true under certain conditions.

If that’s the case, of course agree. Obviously the assumptions I’m providing represent a snapshot I’ve taken from a vast range of available climates, eras, and conditions. I selected a set of assumptions which I feel represent a workable and usable representation for a simulation at the baronial level for the sort of campaigns most GMs run. What I’m doing here is laying out the assumptions behind the system that ACKS uses, so it’s not a black box.

If you wanted to do a different sort of setting, you might change the assumptions. For instance, ACKS has a built-in maximum population for a domain at 100 peasants per square mile, because that was the highest recorded population I found in the areas I modeled the system after. But if you were going to model, say, feudal Japan or China, you’d want to tweak that. Or if you were doing a setting that was a scorched wasteland, you’d want to reduce the land value per peasant, and the population densities.

I have read the Dragonsfoot thread you linked but I think most of the commentators in the Dragonsfoot thread are probably obsessing at the wrong level of detail. The point of the simulation is to be usable at the level of the simulation. This is a common mistake game designers make, IMO. For example, if you read the memoirs of leaders like Von Mellenthin or Rommel, they will hardly distinguish between a Panzer IIIG and IIIJ tank, or the different types of PAK anti-tank guns. That’s because they are thinking about what’s important at the operational level. But all too often a game designer creating a wargame at the operational level will lavish his attention on the differences in the tanks, rather than focusing on the actual issues relevant at the operational level, such as fuel supplies or morale.

Anyway, I hope that makes sense. Certainly you get no argument from me that socioeconomic info varies radically, but one has to start with a baseline before you can start to customize it.

Ah, good. Didn’t mean to confuse, but yes I did mean not true under general but probably true in a specific case. As a game design issue, I do agree that throwing in too much realism won’t be helpful (such as knowing the sleep/work disparity between genders on northern Finland cattle farms - something I do actually know), but the key is the realization that there is and should be a great deal of regional, cultural, temporal and physiographic difference, even within a realtively small kingdom. Think of the variation of Switzerland or Ireland or Iceland. That’s the kind of thing you have to be able to model if the game is going to have any granularity and challenge at the domain level (operational, if you prefer - I have read Rommel and Von Manstein, but not Von Mellenthin)

Soooo. Maybe one way to do that, and maybe what you’re hinting at, would to be to create a baseline economic model, or better, a few baseline economic models based on different modes of subsistence, and then simply give procedures for variations from the “norm”. Could even be random or could be a combination of random events and set conditions. However it may be, my concern is that I would hope ACK will avoid a monolithic Befelstaktik approach to delineating economic conditions.

Daniel, first off - great to chat with another history buff. I assume you’re a professor or grad student? I was a BA in Ancient History with a focus on military, so my knowledge of, e.g. Macedonian tactics definitely exceeds my knowledge of Northern European farming…

Anyway, in ACKS, the way we model regional variation is as follows:

  1. Each 6-mile hex is classed as Civilized, Borderlands, or Wilderness. This determines the starting population in the hex. This classification is based on proximity to large cities.
  2. Each 6-mile hex has a Land Value associated with it that can vary from 3gp to 9gp per month per family.
  3. The Land Value then effects the Demand Modifiers for various trade goods.
  4. The climate, environment, and age of the settlement then effects the Demand Modifier for various goods. A 1,000 year old settlement has higher demand for wheat (because its exhausted its land). A coastal village has lower demand for fish. Etc.
  5. Large settlements adjust the Demand Modifiers for goods in smaller settlements through trade.
  6. The morale of the peasants (based on taxes, pop growth, etc) then adjusts overall productivity.

As a domain ruler, you are faced with choices like (a) securing a civilized domain and the red-tape associated with that versus establishing a domain in the wilderness; (b) finding an initial hex (or more) with good land values, potential trade routes, etc.; (c) investing to attract peasants; (d) maintaining a garrison large enough to defend the domain; etc. But you aren’t necessarily focused on whether the peasants are fishing or herding, except insofar as you get involved in mercantile trade.

As a merchant, you pay more granular attention to commodities and trade routes as you engage in arbitrage trading, shipping contracts, etc. Note also that lords get special bonuses for trading within their domains.

Thanks for taking an interest Alex. My background is an MA in Anthro and “all but dissertation” (more or less) Ph.D, on hiatus cause I ran out of money. I have worked as a professional archaeologist since '98, my theoretical approach of choice being historical ecology, symbolic ecology, chaos theory and environmental archaeology. ho hum.

Anywho, looks like a good start Alex, and its clear you guys have put a lot of thought into your model. In a couple cases I’d have to see some of the details of some of the modiiers you mention and give that a little thought before I could really comment on it.

The only questions/problems I can see from your summary mostly revolve around the question of “land value”. The impression I get - correct me if I’m wrong - is that Land value is a function of horticultural productivity and possibly resource extraction, as measured in 5 mile hexes.

The issue with that is that it presumes a settlement pattern that precludes most forms of transhumance and swidden socioeconomies, seasonal fishing economies, and fur trade economies (very big deals in medieval Russia for ex).

These sorts of social organization of production were very common in pre modern societies and I’m not sure how or if that fits into the land value equation, but in these sorts of socio economies the areas of heaviest population might have little actual production or land value even though the people there still generate a good deal of taxable wealth from “country” products, be they agricultural or other resource gathering.

Island economies can present a problem in this regard. Its not unusual for the areas of densest population to be poor places to farm and have little in the way resources - except the sea. Or if you look at a place like the Faroe islands where villages of a few hundred people crowded on the few places along the shore suitable for building while the sheep roam free in the interior. As medieval peasant life went, the faroese did pretty well between wool, fish and whales.

Of course I don’t know if the land value figure or some other measure takes these kinds of things into account or not. I’m guessing not, but maybe its an easy fix. Maybe its just a matter of assigning a “Tax Value” to a hex ranther than a land value, per se.

Hi Daniel! I have not heard of historical ecology or symbolic ecology, though when you throw chaos theory in there, it sounds cool.

As to your point, Land value is less a less a function of productivity and more a measure of productivity i.e. it’s not a function in that it’s not derived from any lower data point within the game. It’s just set by the GM and/or random die roll, within a range that seemed appropriate based on my historical research.

A 3gp land value might represent poor, unproductive land. A 9gp land value might represent rich, fertile soil, but it could also include a lake plentiful with fish, or a copper mine. You work out the specific details as outputs from the land value based on the terrain.

This is, of course, backwards from the real world. But it’s much, much easier on the GM, who can start off by assigning simple numerical land values to broad territories, and the only sharpen his focus when he needs to, such as if a player wants to trade in that region.

There are also “service values” and “tax values” associated with land. Service value is a function of the number of peasants and their morale. Tax values is a function of the number of peasants, their morale, and the tax rate. I did not model the range of other variables that could affect the service values and tax values, although this discussion is making me think maybe I should.

So then “land Value” as you explain it sounds more flexible than I thought it might be. I’d probably be happier if some kind of ranges were given depending how the land is exploited, with modes of exploitation depending on knowledge/expertise/subsistence types or something on these lines - categories like mining, fishing, horticulture, agriculture, textile production (wool cotton fur), with a minimum - maximum population for each. Just stream of conciousness thinking here, but for example a hex might have gold deposits that require x number of people to exploit x amount, to a maximum of x. This way you could tie certain industries, like say Irish style Booley sheepherding, across several hexes, as needed.

Symbolic ecology - nobody has heard of it :slight_smile: The term was coined by an old Prof of mine. Basically the idea is people view their landscape through the lens of culture.

Historical Ecology is a bit more common -