Demographics in D&D – Another Look

Demographics in your role-playing games is one of those subjects that people either love or hate. Or rather, they tend to either find it to be either fairly useless (and probably rather boring) or a very engaging facet of world-building. If you’re reading this article, you’re probably one of the latter individuals.

In 3.X d20, demographics were baked right into the rules. Specifically, there were a series of tables in the 3.0 and 3.5 Dungeon Master’s Guide that allowed for settlements of various sizes to be created, with subsequent tables allowing for determining how many members of various classes were present, and what their levels were. Unfortunately, these rules weren’t added to either SRD, but we can still discuss them in general.

Personally, what I found frustrating about these rules was that you could only use them to generate populations – and, more importantly, the class-and-level breakdowns – for a single settlement, instead of a larger area such as a country. To my mind, what we got was a bottom-up level of world-building that stopped at the halfway point. But recently, I’ve hit upon an idea that’s made me reconsider the population tables in the DMG:

Specifically, look at Table 5-2: Random Town Generation on page 137 of the 3.5 DMG (Table 4-40 on page 137 of the 3.0 DMG). What if the percentage listing for each town on that table was the breakdown of how many settlements of each type were in a given country? For example, there’s a 10% chance of rolling a thorp on that table…so 10% of the country’s people live in thorps. The 1% chance of rolling a metropolis means that only 1% of the population live in metropolises, etc.

Now, that still doesn’t tell us how many people live in a given country. For that, we need to take a top-down approach, which means picking how many people we want there to be in total and then plugging that into the various percentages.

Once we do that, we just pick a suitable number of people per settlement size (using the ranges given on the table in the DMG), and divide that by the number of people who live in settlements that size, and voila! Now we know how many settlements of that type are found in a given country.

For example, we want to make a generic kingdom with a total population of 5,000,000 people. As such, 10% of them, or 500,000, will live in thorps. Since thorps are listed as having 20-80 people live in them, we’ll assign an average of 50 people to a thorp. Ergo, our generic kingdom has ten thousand thorps in it.

One thing to keep in mind is that these numbers are meant to provide a framework, rather than a mandate for how the population breakdown needs to be. If you’re using this method to make a campaign world, then once you’ve plugged in the various numbers at the various levels, go ahead and start making edits based around how you want various countries to look. What’s here is to help inspire, rather than present a straitjacket.

One question that often arises when looking at a game world’s population is who does it count among its people? Does it only look at the core races (e.g. humans, elves, dwarves, gnomes, halflings, half-elves, and half-orcs)? Or are all Humanoids part of the count?

Insofar as the tables in the DMG are concerned, this is decidedly the former. Only the core races are listed on the Racial Mix of Communities (p. 139 in the 3.5 DMG; Table 4-46 on p. 140 in the 3.0 DMG). As noted above, you should feel free to disregard or alter that as necessary to suit your world-building.

Let’s go ahead and take the “Generic Kingdom” from the previous example and flesh it out using this method. With five million people, its settlement breakdown will look like so:

• 10% of the population (500,000 people) live in thorps. Presuming about 50 people to a thorp, then the kingdom has 10,000 thorps within its borders.
• 20% of the population (1,000,000 people) live in hamlets. Presuming about 250 people to a hamlet, then the kingdom has 4,000 hamlets.
• 20% of the population (1,000,000 people) live in villages. Presuming about 800 people to a village, there are 1,250 villages in the kingdom.
• 20% of the population (1,000,000 people) live in small towns. Presuming about 1,600 people to a small town, then the kingdom has 625 small towns.
• 15% of the population (750,000 people) live in large towns. Presuming about 3,500 people to a large town, then there are 250 large towns within the kingdom.
• 10% of the population (500,000 people) live in small cities. Presuming about 10,000 people to a small city, then the kingdom has 50 small cities.
• 4% of the population (200,000 people) live in large cities. Presuming about 20,000 people to a large city, then there are 10 large cities within the kingdom.
• 1% of the population (50,000 people) live in a metropolis. Since there’s no upper limit on the population of a metropolis (minimum 25,001 people), it’s easiest to say that this will give us a single metropolis of 50,000 people.

Profession (cartographer)

A DC 10 Profession (cartographer) check will produce a map of the kingdom that has the metropolis and large cities marked on it. For each additional +5 to the DC, the map correctly notes all of the settlements one size smaller. For example, with a DC 25 Profession (cartographer) check, a map will accurately place the kingdom’s metropolis, large cities, small cities, large towns, and small towns, but will not have the villages, hamlets, or thorps.

One thing to note about the above aid for world-building is that this will cause problems if you have a kingdom with a population of 2,500,000 or fewer. That’s because, if you take 1% of the population for a metropolis, you won’t hit the necessary minimum for a settlement that size according to the DMG table (e.g. a metropolis has 25,001+ people living in it).

In this case, simple create the country using the method outlined above, and when you hit a point where the population assigned to a particular settlement size doesn’t meet that settlement’s size prerequisite, simply redistribute them among the next-lowest settlement.

For example, if you’re making a country with only 2,000,000 people, the 1% of the people that would go into a metropolis (20,000) are instead added to the 4% of the population (80,000) that would live in large cities. At that point, we decide to make five large cities of 20,000 each.

Overall, this method of fleshing out the number of settlements in an area helps to provide a basic overview of a particular country – though it should be noted that this doesn’t need to be for a country per se. Rather, this can function for any region that has multiple settlements within it, whether political, geographic, or whatnot.

At that point, if you want further inspiration, you can start using the supplementary tables, specifically the ones for Community Modifiers and Highest-Level Locals. Obviously, I don’t mean that you should use these to generate every single NPC in every single settlement! Rather, a few random rolls can help to generate ideas about who the power-players are in a given settlement (or larger area, if you think they’re high-enough level to warrant it).

Unsurprisingly, the DMG tables only deal with the Core PC and NPC classes. In the same vein as the rest of this article, we can presume that tells us something about the state of the rest of the game world. Specifically, that these classes are far and away the most common, with classes from other supplements being exceptionally rare, as per your needs.

Tags: demographics, Dungeon Master's Guide, population, settlements