In the early days of Dungeons & Dragons, the rulebooks gave various methods of determining what monster was randomly encountered. The earliest tables used a flat roll – roll a d6. On a 1 it is an orc. On a 2 it is a goblin. And so on and so forth.

Advanced Dungeons & Dragons gave a more nuanced approach: Percentile tables, with different probabilities for each monster. If the dice show 01-10 it is an orc, if they show 11-30 it is a goblin. Thus, goblins appeared twice as often as orcs on that table.

A third type of table arrived with Monster Manual II, the 1d8+1d12 table. If you roll the dice, you get a number from 2 to 20. But why this? Why not 2d10? Doesn’t that give the same probabilities?

It took me a while to understand that it did not!

The roll of 1d8 + 1d12 makes all rolls of 9 to 13 equally likely, making a flat “top” of the graph. This is compared to the sharp top of the 2d10 roll.

This construction is not always used, but you can see it in the fifth edition Dungeon Master’s Guide in Chapter 3, Creating Adventures, where it discusses Creating Random Encounter Tables (pages 66-7 of my copy).

Rare, Uncommon or Common?

Every monster in AD&D was rated as Very Rare, Rare, Uncommon or Common, depending on how frequently it was encountered in the land. This could be used when constructing the random encounter tables.

In Monster Manual II the tables give:

2: Very Rare
3: Very Rare
4: Very Rare or Rare
5: Rare
6: Rare
7: Uncommon
8: Uncommon
9: Common
10: Common
11: Common
12: Common
13: Common
14: Uncommon
15: Uncommon
16: Rare
17: Rare
18: Very Rare or Rare
19: Very Rare
20: Very Rare

To mix things up, you could make an Uncommon entry instead a 50/50 choice between two Very Rare monsters, and likewise for Common and Rare monsters.

Probability Graphs

Consider the difference between the d8+d12 graph, and the 2d10 graph. Unusual, are they not?

d8+d12:

2d10:

Graphs by AnyDice.com

Regional Encounter Tables

One of the tricks that the designers used in the World of Greyhawk setting was to create regional tables that provided special encounters for the area; however, certain results sent you back to the regular encounter tables in the Dungeon Master’s Guide. This was a useful technique of providing flavourful encounters while not reducing the potential variety of encounters.

Here is the table for the Dim Forest, which is where my Greyhawk campaign is currently located:

01-03: Demi-humans
04-12: Elves, Sylvan
13-18: Humanoids
19-22: Men, Bandits
23-25: Men, Brigands
26-27: Men, Patrol, Light
28-31: Men, Raiders
32-35: Men, Tribesmen (woodsmen)
36-00: Use Standard Encounter Tables

I find this table quite odd, because the description of the Dim Forest says that “terrible creatures live elsewhere within its bounds”. Well, I suppose men of all sorts are “terrible”, but I was hoping for more of that sort to be called out!

However, the idea of patrols and the most common inhabitants being called out as more common helps give ideas to the DM.

I am now creating my own – more monstrous – version of the table. My players will appreciate it; at least, I hope they will!

Other Styles of Tables

While the probabilities of these tables are interesting, other designers have taken the opportunity to do more than just give what creatures are encountered, but also their motivations and the terrain that they are encountered in.

I am fascinated by Zzarchov Kowalski’s method used in his Neo-Geek Revival supplements such as Lost in the Wilderness (pictured right). You roll three dice: one indicates the monster encounter, a second the terrain, and the third some strange things. And then doubles, triples or other combinations may alter the encounter some more.

There are aspects of the tables I do not like so much (such as the small selection of monsters in the basic version), but creating new ones would not be particularly hard: it is a good example of how a few random elements combined together can give a large variation in results!