Random Encounters


Yeah, kinda like that.

I’m not saying this sort of shit doesn’t happen in real life. This picture is testament to that. I’m saying there’s a difference between this and what we think random encounters are (or should be) in D&D.

Take the most basic example: a bear or pack of wolves. How long have they been living in this stretch of forest? This is their land; they know the hills and valleys, they know where the dead trees are, or where to find the hidden streams. We are the strangers here. We are their encounter. What’s the likelihood, if we look at it from that angle, that a pack of wolves will be caught unawares when a group of five to ten heavily armed and armored adventurers come tromping through the bush?

In other words, that bear and those wolves see you coming. If you encounter them, it’s because they wanted you to. There’s nothing random about it.

So how do we decide if the bear encounters the party (or vice versa)?

We start with a question: is there a bear? Better yet, what creatures are in the area? How big is the area and what’s inside it? And what motivates those creatures?

I won’t address the last question, not just yet. Others have tried and I know it’s a beast of a task to overcome. I feel I have a starting point, something I can draw upon to better understand the options, but I also realize that, as you go up the intelligence scale, the number of options available grows at a profound rate. It won’t be sufficient to have a single chart, or even a handful of charts, to account for all possibilities. So more effort has to go into addressing motivation.

The other questions though ~ I think I have answers for them. At least, I have answers that I feel are satisfactory for my game, and so I’m sharing them here; feel free to pick holes in them.

How big is the area? What’s inside it? How can we quantify it and how can we turn that into an estimation of what’s living there?

Let’s look at a few numbers. An acre is 43,560 square feet, approximately 208.7 feet on one side. How much biomass is in an acre? Or a square mile? (640 acres per square mile.) I’m sure you can see where I’m going with this: if we can estimate the biomass, we can estimate how many herbivores the hex (region) can potentially support. Then we can estimate the number of carnivores and omnivores and scavengers. And we can manipulate those figures further, based on whatever criteria we deem appropriate, such that the end result is a series of tables that can tell us, at a glance, which creatures (and how many) are in a given area.

Biomass. Not an easy thing to wrap our heads around. Simply put, we don’t know enough and there are too many variables involved to determine, with certainty, any fixed figure. But, as this is a game, we don’t need certainty. We just need a fairly good estimate that we can assume is a baseline for crunching numbers. (I can’t locate the webpage where I got this figure, but I know it was a university page, having published a paper on the topic.) I’m using the high-end number of 550 billion metric tons; further, I’m assuming the total biomass for the world includes insects and animals, so the 550 billion tons is reduced to 302.5 billion tons (55%), which represents the total plant biomass.

Our world has approximately 57.3 million square miles of land. This gives us an average biomass of 11.6 million pounds per square mile.

Remember, this is a baseline. It’s a figure we use for drawing a comparison. The comparison, at this point, is the relative plant biomass in a given climate. Or habitat, or biome, or whichever term suits your purposes.

Concerning biomes, there are some great resources out there, that provide all kinds of useful data. Most notably, we can learn the annual rainfall, temperature ranges, length of dry and rainy seasons, and similar details. All of this comes together in a calculation that lets us assign an average biomass to a given biome.

This is what I came up with: ((Annual Minimum Temp (Celsius) + 20.5) * (Annual Maximum Temp (Celsius) + 20.5) / 2) * (Annual Rainfall (mm)) * (Duration of the Wet Season (in months)) / (Duration of Summer (in months)) / 1,120).

That last number is two separate figures ~ 1.75 * 640, and I’ll be damned if I remember why I chose them. I know they have something to do with the research I did and the numbers I found, but I don’t have anything in my notes that explains this calculation. I only know that, when I run the calculation for a Mediterranean biome, I get 11.36 million pounds per square mile. Which means I trust my calculation to represent the approximate biomass of a given biome, assuming (of course) that the underlying data is sound (and it is, because we know the data points; we’ve only been measuring them for over a hundred years now).

Now, I get it, that’s a lot of number crunching and I probably didn’t do a great job of explaining it. However you arrive at a figure for the biomass of a given area, once you have it, you can estimate how many creatures that area can support.

But there’s a problem ~ and this is the real point of this post ~ eventually, you’re going to run into a snag in your research where you basically can’t get the information you need. Mostly, this is because there’s no easy way to guesstimate the calculation or the figures you’re looking for. So you’ll have to make a few assumptions. Or adjust your thinking.

In this case, I’m talking about the goal behind all these numbers. What we’re really after is an encounter table that makes sense, and for me, that means a table with a rational distribution of critters. I’m not concerned with precisely how many critters are in a given hex. Barring access to a powerful divination spell, it’s highly unlikely that the players will ever need to know that, so it’s information that I’ll never share. Which means I can work around “not-knowing-it.” And that workaround is this: I don’t need to know the exact number of critters, by species, in the region. I need to know the likelihood that the players will encounter one or the other (or both at the same time).

This is where the estimated values help us. We start with one estimate ~ the approximate total biomass of a given region ~ and we convert that to a potential maximum population. We do that conversion for each herbivore that could potentially live in our region. This is the point that must be understood: we’re not interested in knowing precisely how many deer are in found in a given square mile; we’re interested in knowing the likelihood of finding deer as opposed to moose as opposed to raccoons, or foxes, or wolves, etc.

Take the potential total population for a given species. Add it to the potential total population for all other species. Start with herbivores; the methodology works the same for carnivores and omnivores, but you need to compare apples to apples; so add up all the potential herbivores in a given region, and use it as the baseline to determine the chance (%) of a critter appearing. For example, if a given biome can support a maximum of 10,000 deer, and the total potential maximum population for all herbivores is 250,000, then deer have a 4% chance (10k / 250k) of appearing in a given region. (Since your regions will vary in size, based on your needs or the presence of civilization, that sort of thing, I suggest using more dice rolls to decide how many actual creatures are in the region; so for each square mile, roll the dice once, or something like that.) As you add more creatures to the pile, you’ll get smaller and smaller chances for each species, which is great because it’ll contribute to greater diversity.

When I tested this system, there were still some issues. Extremely small critters (anything less than five pounds on average) tend to appear in such large quantities that, although their likelihood of appearing is less than 5%, they tend to dominate the chart. Additionally, my bestiary database only accounts for creatures already published in a few sources; so I don’t, for example, have stats on rabbits. And normally, I wouldn’t. But if I want this system to work, I have to account for the potential presence of rabbits because their population will impact the potential population of every other critter in the region. Finally, I’ve only accounted for creatures with one of four dietary habits. There are other monsters with special requirements, or no requirements, like undead and constructs; when I use this method to populate a random hex on the map, I need another way to account for these creatures.

You can see from this screenshot how I envision the product will look. The calculations are complex; feel free to ask if you’d like to see them in detail.

Bestiary Encounters


3 thoughts on “Random Encounters

Add yours

  1. I came from your post in A. Smolensk ‘s group in facebook.

    I think an additional interesting point to consider is that, given that you know that X animals are present, how likely is an encounter. Of course this will heavily depend on the animal’s biology and ecology. For instance, is it actively seeking humans, or avoiding them. Is it a primarily nocturnal animal? The behaviour of the animal will also vary with its sex, age, and the season (think about stags during rut).
    All this could require the building of an extensive ecological data-base that would be the base for an animal encounter table (you’ll remember that Alexis had started such tables of course).

    This could answer your question about critters: there may be many more rabbit than deer, but your less likely to notice when you are in presence of one (it may be more cautious, and it is simply smaller and more easily hidden by the vegetation).

    Another interesting point would be to consider that some scientists asks themselves the reverse question : how can we estimate an animal’s population size (in an area) given a few observations?
    As an introduction you may look at : https://projects.ncsu.edu/cals/course/fw353/Estimate.htm
    These counting methods are based on mathematical models that may be reversed to tell us how we can observe the animals given that they are that many in the area.


    1. Brilliant. Thank you.

      I should probably add that I’m looking to accomplish something very similar: I want to present the players with something interesting. More often than not, though, I’m fielding questions like, “We go hunting, what do we find?” So I want to honestly say, “Nope, no deer in this area.” I’ll come up with a reason; but the randomness behind the answer is what I want.

      And I’ve been going crazy trying to find data on real world populations and distributions (across habitats) that I can easily port over to a database. Frustrating.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create a free website or blog at WordPress.com.

Up ↑

%d bloggers like this: