Random Tables and Simulated Worlds: An Exercise
The thing about random tables – or “procedural generation” if you’re one of those midwits that thinks extra syllables makes you sound smarter than you really are – is that they are a shortcut method to operating the machinery of a game world in a way that makes you look smarter than you really are.
If you set them up properly, random tables can mimic the operations of a world without you having to track the minutiae. You don’t have to know how many goblins live in this forest, you just have to know there are twice as many goblins as talking spiders, and half as many goblins as elves, and so on.
Then you turn your players loose in the woods, and instead of checking six different patrol schedules, you just roll on a simple table, and it tells you which of the outcomes occurs.
When I mentioned this in the backrooms, the bros in the BROSR told me to take a look at the random tables in the back of Gary’s Monster Manual II. So let’s do that.
The simplest table to run the numbers is Dungeon Level 1. For this, we roll 2d10, add the dice, and that gives us nineteen different outcomes. To really understand what this table means, we have to also look at the percentages for each of these results.
As you can see, the most common result (10%) is an 11, which translates to the most common monster in the standard dungeon: a rival party of adventurers! Almost as common are bat swarms and orcs (at about 9%), with giant rats and bandits tying for fourth place (at about 8%). There are demons (1%) even this high in the mythic underworld, but you’re far more likely to encounter claim-jumpers than demons. This tells a story all in its own right, and the idea of the implicit world has been discussed often enough there’s no point in belaboring that here.
Instead, let me point to Gary’s advice in the AD&D DMG, that once you have rolled up a few encounters, those encounters should become the basis for your own personal random encounter tables.
So what happens in a campaign is that you as the DM can either create your own site-specific random tables, with percentages based on back-0f-the-envelop population numbers. Alternatively, if you’re a #ZeroPrep kind of guy, you can use the standard random tables to discover along with the players who the most common inhabitants are, and after you’ve encountered five or six locals, craft a new site-specific random table based on those organic and at the table results. This allows you as the DM to explore right along with your players – it’s brilliant design.
When I did this in the Wolflands, I made sure to include a low probability wild card – rolling snake-eyes on the random encounter table directs me to turn to the DMG tables. That adds the element of uncertainty, and allows for “new factions” to pop up when least expected. With just one little table, you can breathe whole worlds of life into your campaign, and you don’t need to overthink it.
You just need to trust Gary’s process, and trust the dice. They’ll give you the adventure of a lifetime, if you let them.