The 100-Hex Sandbox

As you've probably noticed, I've been spending some time using the tables found in Appendix B of the AD&D 1e DMG to populate my sandboxes, which I will definitely use in a real game at some point...

See also d66 Reasons Why the Castle is Totally Deserted, which is my attempt to answer a question that arose as a result of using the tables in the DMG, and Using Reaction Rolls to Determine Faction Relationships in the Sandbox, which is a nice companion to any of my posts about creating a sandbox.

Here is yet another take on using the AD&D 1e DMG to create a sandbox. First, allow me to remind you of this table:

I was looking at this table and thought about what it would be like to extrapolate the percentage chances of inhabitation per hex instead into a general rule. That is, instead of there being a 16% chance per hex of a settlement, castle, or ruin being present, maybe 16 in 100 hexes are populated with such locations. If you create a play area consisting of 100 hexes, you simply stock 16 of them with such locations, rather than rolling to determine their contents. 

Obviously this is statistically likely anyway, but this spares you a lot of rolling and creates a more standardized area of play. (Rolling is fun, of course, and standardization can be boring, but bear with me.)

You could of course instead prepare a play area of 50 or 25 hexes and stock 8 or 4 of them respectively for a more manageable map. This has a slight downside, however, as towns and cities are each 1-in-100. With a 100-hex play area, I feel comfortable saying yes, a town and a city both definitely exist in this space.

I present to you my 100 hexes:

You can generate the terrain however you see fit. As usual, I'm using some variation of Welsh Piper's Hex-Based Campaign Design. In this case, I rolled d3 to determine the climate and got 2 (temperate), then rolled d6 to determine the predominant terrain and got 6 (water). This area is a freshwater lake or sound or something. 

Using 100 hexes and the Welsh Piper method, I can also easily standardize the terrain as follows:

• 50 water hexes
• 25 plains hexes
• 16 to 17 forest hexes
• 2 to 3 each of desert, hill, and swamp hexes

(That part is optional and not super important, but it will make a later step easier.)

Now here is the list of inhabited hexes to add:

• 3 single dwellings
• 2 thorps
• 2 hamlets
• 2 villages
• 1 town
• 1 city
• 3 castles
• 2 ruins

There are four types of ruins, so you have to roll to determine which kind they are. I got a village and a shrine. If you instead want to have one of each type using this method, you will need to make a 200-hex sandbox!

Castles, of course, have their own subtables on another page:

There are technically 36 types of castles! There are 9 types on Castle Table I, but each of those types can in turn be totally deserted, deserted but inhabited by a monster, inhabited by humans, or inhabited by character-types. If you factor in the four types of humans on Castle Sub-Table II-A, the 11 character-types on Castle Sub-Table II-B, and all the monsters listed on the various outdoor random monster encounter tables, the possibilities explode beyond the scope of my comprehension. It is not worth it to try to make a play area that includes all of these!

Instead, we will have three castles. You could roll for all of them or, since 35% will be small, 45% will be medium, and 20% will be large, you could instead decide that one is small, one is medium, and the third is either small (5%), medium (35%), or large (60%). I roll and get a 5, so I have two small castles. 

The exact type of "castle" doesn't matter much to me, so I decide that small castles are towers, medium castles are keeps, and large castles are proper castles. I roll for the state of each and get a tower inhabited by brigands, a totally deserted keep, and a castle inhabited by a fighter.

Returning to my map, I start placing locations:

I place the city in the middle, because why not. I place the town and villages each a day's travel from the city, and randomly place everything else. I then draw in "roads" (although in this case they are merely a visual representation of how people get from one place to another, mostly by boat) connecting the town and city and going off the map, presumably to other towns and cities. Then I connect the other settlements by the shortest possible path to that main route. I draw in the "roads" at this stage because hexes through which they pass will count as inhabited, meaning they won't contain monsters' lairs.

Finally, I then place the monsters. Remember when I said that standardizing the terrain would come in handy? Well, since I know that my 100 hexes include 50 water hexes, 25 plains hexes, 17 forest hexes, 2 desert hexes, 3 hill hexes, and 3 swamp hexes, I can now subtract those which are inhabited and end up with the following uninhabited hexes:

• 23 water
• 18 plains
• 15 forest
• 1 desert
• 3 hill
• 3 swamp

Why does that matter? Because each type of terrain has a percentage chance of containing a monster lair:

• Water, plains, and desert: 10%
• Forest and hill: 20%
• Swamp: 40%

(I'm pulling these numbers from 2e, not 1e, and so am straying a bit from the spirit of this exercise. 1e instead uses population density to determine the chance of encountering a monster. However I'm also not using the 1e method of random terrain generation, so we are already beyond the glow of Gygax's light.)

Based on the percentages, the 23 water hexes will contain two monster lairs, with a 30% chance of a third. The plains hexes will contain 1 monster lair with an 80% chance of a third, the forest hexes will contain 3 monster lairs, and so on. I roll dice when necessary and get:

• Water: 2 monster lairs
• Plains: 2 monster lairs
• Forest: 3 monster lairs
• Desert: 0 monster lairs
• Hills: 1 monster lair
• Swamp: 1 monster lair

I tuck these in little pockets of wilderness, not too close to the city or the stronghold, but close enough to civilization to interact with it:

Voila! I think this is a pretty decent place from which to start a campaign, and it's much smaller than the other sandbox I've been posting about on this blog. There's a town, a city, plenty of smaller settlements and dwellings, a minor stronghold, a major stronghold, three ruins to explore (including a stronghold which can eventually be reclaimed by the player characters), and nine monster lairs to cause complications for the region.

The point of this exercise is to demonstrate that, rather than rolling on tables, you could instead pick a size for your sandbox and use the tables to extrapolate what all should be included in the region. Or, pick the things from the tables you want to be sure are included (in my case, at least one city) and determine the size of the sandbox from there. 

I won't go so far as to say this is the right way to do it, but it's another option in the toolbox. It presupposes that you care at all about what's written in the 1e DMG, which many don't. And that's totally valid! It describes a very particular type of game.

I am curious to about one other matter - how big would a sandbox need to be to "guarantee" the presence of a dragon? They are exceptionally rare in AD&D's encounter tables, such that in a hex containing a monster lair, there is only a 2% chance of it being the lair of a dragon in plains, desert, hills, and mountains, and only a 1% chance in forest and swamp.

Considering that roughly 60% of my hexes were uninhabited, and that there is roughly a 19% chance of any uninhabited hex containing a monster lair averaged across the different terrain types (again, using the 2e values), we can expect an average of 11 or 12 monster lairs per 100 hexes. Since there is either a 1 or 2% chance of such a lair belonging to a dragon, let's call it a 1.5% chance. That means that 1 in 66 or 67 monster lairs will be a dragon lair. 

That's roughly one dragon per 600 hexes, or one dragon per 6 cities! So if you want to "guarantee" a dragon in your sandbox, take the above map and multiply it by 6. Or, you know, just make however big of a map you want and plunk a dragon down somewhere. It doesn't matter. But I had fun with it as a thought exercise. Maybe you did too.