I’ve been working on the Encounter tables for the current campaign, and it brought up something I’ve been doing for a while.
So, we all know the standard B/X/1e &c. tables run from 2-12, using 1d4 and 1d8 rolled together (and if you didn’t, you’re going to have a rather weird distribution – explanation after the break). These days, though, I just throw d4/d6/d8 at once. I originally started adding the d6 to do surprise for whatever I rolled, but then I started including non-combat “encounters” to
screw with my players — um, “enhance the atmosphere of the game”. I’ve found it adds considerably to the tension when the players find a smoldering campfire and don’t know whether they’ve just lost a Surprise roll, or only come across an irresponsible camper’s leavings. So I experimented with making larger tables at first, but it never felt right.
Now I use two to three tables, with the d6 determining which the other two dice are rolling on. Generally speaking, even numbers on the 6 are an “event” encounter – stuff like finding signs of another creature or party, roadside shrines, omens, what-have-you. Odds roll on the conventional Wandering Monster tables. To adjust to the changed odds, of course, I have to roll about twice as often as I normally would. I also skip the d6 and just roll directly on the WanMo table for “triggered” checks like smashing apart furniture or noisy fighting among the party. I also occasionally use the d6 for other things, or to adjust the odds of Weird Crap happening – a deserted road in the middle of a plain is going to have a lot more encounters with random ruins, foul omens, or water sources than with other people/monsters, etc., so I might up it to 1-4 or 1-5 being on the “events” table, and only a 6 triggering a true check, or using the 6 as a trigger for a table of one-time encounters replacing some of the “usual” ones. Unique encounters that you re-roll can be replaced with the normal one at your option. You can also have one or more entries on the Creature table drawing from a local lair or “pool”, with a fixed number possible to encounter, and after that ignore the result or replace it with the corresponding event.
Anyway, my current procedure looks like this:
Roll d6 for encounters -> comes up positive
Grab 4/6/8 and check the tables -> find results
(Note: even when it’s a complete non-combat encounter like a gust of wind, I still follow these next two)
Throw a d6 for Surprise, and having the party’s scout do the same
Roll 2d6 for Reactions or Morale based on the party’s actions (if neither, both, or the party got Surprise) and/or appearance (if the critters did). Note that even a person with Surprise and a profoundly negative Reaction roll might still run or hide if it seems like a good idea compared to taking on a pack of heavily-armed murderhobos.
Here’s a sample table for a road through a moderately-traveled forest, with some notes added below:
* For omens: Take a quick card reading/roll off a manifestation, or make one up. Can also represent signs of a “sacred” or “cursed” area, a feeling of being watched or protected, etc. Yes, evil omens are more common. Life sucks.
Good Omens: Any hirelings will experience a +1 on morale for the rest of the day and all party members gain an additional point of HP resting that night.
Ill Omens: next Encounter will be a Creature on 2+ instead, and Unique on a 5+. Morale check for any followers, if failed they will refuse to continue
+ Shrines: Use Surprise roll to determine the size or type and the Reaction roll to determine compatibility of the shrine’s faith. Surprise die: 1, 2 = abandoned/ruinous : 3,4 = Maintained roadside shrine : 5 = Single Hermit : 6 = Monastery or Church.
[Examples of Pagan shrines could be stone circles (for a church), roadside offerings, or even Fairy Rings and burial mounds. ]
** Roll again on the Creatures table, and invent an appropriate sign of the potential encounter. If a Creature comes up on the next Encounter roll, disregard the result and replace it with this creature.
So, not that you probably care to have it broken down, but the 2d6 method has a “normal” or bell-curve distribution: the number “7” pops up on one-sixth of your rolls, and “6” or “8” show up very nearly as often. This weights the results very strongly towards the middle, and makes the rare results rarer by comparison.
With the 4/8 method, it’s a flattened curve. That means the center 5 results (“5” to “9”) are all equally-likely, while the edge cases, “2” and “12”, are rolled one thirty-second of the time (as opposed to one thirty-sixth with the 2d6 method). It allows “common” results to happen without drastically weighting one event or creature type, and means that the rare results will still come up a bit more often (using 2d6, statistically, requires 108 WanMo checks to see every result on the table, while d4/d8 averages out to only 96. Not a huge difference, but it illustrates the point.).
If you’re designing your own dual-dice system, remember the following: No result can be more common than the number of sides of the smallest die. Two dice with the same number of faces will produce a Normal/bellcurve, and the wider the difference between them the flatter the curve will become. The same principle applies to larger rolls. For example, rolling stats with a d10 and a couple d4s would produce a much more varied population than 3d6. The curve would flatten out to only sixteen results (multiplying the two smaller dice together) for each of the most common numbers – instead of 36 – and a “3” or “18” would have 1:160 odds compared to 1:216 with the conventional method.
This ends today’s statistics lesson. Ite, missa est.