Let me tell you what happened to me the other day. Well, not the other day. A couple of weeks ago. But that doesn’t matter. It’s relevant to the thing I’m writing. At least as relevant as anything that ends up in these Long, Rambling Introductions™.

I had gone out to pick up this thing for someone. I wasn’t doing a favor for a friend. Someone was paying me to run out and grab this thing and bring it back. Kind of a delivery thing. They were paying me a decent price, but I was doing it more because I wanted to establish a good business relationship with them. So, I’m cruising along with the stuff in the back when, all of the sudden, there’s this alert. I’m getting a distress call.

Now, I was making good time. So, I figured I could check it out. I turned off the cruise control and popped out into normal space and found myself in the middle of a debris field. Looks like someone had had a big accident. But whatever had caused it was long gone. So, I started scanning around. There was some valuable cargo drifting around, but there were also some occupied life pods. The people on the ship had managed to evacuate and were adrift in these little floating coffins just waiting for someone to come along and pick them up.

I had some room in my hold, but not much. I did a quick count and I could grab about half the escape pods. The rest of the space in my hold was taken up by the cargo I was transporting. I could grab the pods I could fit and take them to the nearest station, but if I left the debris field, I knew I’d never find it again. Those things drift. They get caught in gravity wells. And they are too small to get a navigational fix on unless you fly right past them. It was only luck I found it at all. Anyone I didn’t take with me when I left was probably gone forever. And there was other stuff too. Personal effects.

Now, I could dump the cargo I was hauling and make some room, but it was a salvage job. The stuff couldn’t be replaced, and it’d be gone forever if I dumped it. So, I wouldn’t get paid. Now, that wasn’t a big deal. Because I knew the local station would reward me for bringing in the escape pods. Government search and rescue agents paid good money for rescue and salvage operations. That was to encourage people to actually help the survivors instead of just selling them to slavers. But the money wasn’t the only issue. I didn’t want to hurt my reputation with my new employer. I needed that reputation. I had big plans. And I’d been cultivating the relationship for weeks to get salvage jobs like this.

Still, these were innocent human lives. But no one would ever know if I just abandoned them. But I’d know.

On top of that, some of the cargo that was drifting around was valuable. If I grabbed some of that and sold it off, I could recoup the loss on my salvage job three times over. But I’d have to abandon the personal effects. The reward for that stuff isn’t great, but it would be a great comfort to the survivors to have their stuff. Especially irreplaceable, personal things. And if any of the personal effects belonged to people who didn’t survive, it might provide some comfort and closure to their families to know what happened.

Whatever I did, it wasn’t just going to be about money and business. It was going to say something about me as a person. And just thinking it through was saying a lot about me. The idea that I’d even consider abandoning people to die for a business opportunity didn’t make me proud. What was I becoming? Of course, I have to live too. Business isn’t just business. It keeps fuel in the tank and synthesized protein paste in the ships store. If I can’t pay for the ship and the insurance and the taxes and the docking fees and the occasional fine I incurred from doing these little side jobs, well, I might as well be floating in a space coffin. Because I’d starve and die.

It’s kind of funny how one lucky – or unlucky – die roll slammed me into a moral crisis. Stupid random encounters.

What IS a Random Encounter

The other day, I posted the first of two articles examining random encounter systems and belittling the GMs who dismiss them as useless. Because random encounters actually do useful things in adventure role-playing games. It’s just that they’ve evolved badly. So, they seem like crap. But with a little tweaking, they can be fixed. And with a little MORE tweaking, they can actually enhance your game in some very interesting ways.

In the first article, I explained that random encounters – used properly – create urgency and tension but also emphasize the players’ agency. That’s because they are a part of a game that is visibly outside the GM’s control and outside the planned game balance and the players’ actions directly impact the odds of their stumbling across a random encounter. And that’s not just good for a sense of challenge, it’s also good for a sense of narrative pacing, and it’s also good for setting up crucial choices that have lasting consequences. Sort of.

Then I explained how GMs and game systems overcomplicate the crap out of random encounters in useless ways. And finally, I gave you half of a good random encounter system which basically amounted to this: whenever the players spend time on something or whenever they do something reckless, roll a die. If it’s a one, they have a random encounter.

Okay, review done. Because that was only half the battle. Well, two thirds. Because the rest is about actually designing the damned encounters. And that still means a table of a dozen monsters to jump out from behind a curtain and yell boo, right?

Well, that’s OKAY if you’re running a bog-standard dungeon crawl. If the game is about fighting through hordes of monsters to see how deep you can get into the dungeon and how much stuff you can collect, that works fine. And that’s why random encounter tables are the way they are. That’s why they are lists of ten to twenty random monsters of varying difficulty levels. In a hostile dungeon where you have limited resources and you have to retreat to an exit before you can get any of those resources back, having to wade through a bunch of extra fights of random difficulty adds a lot of tension and urgency. Especially if the things you’re doing in the dungeon to make sure you get as far as you can and get the most stuff and spend the fewest resources increase the odds of having random fights. Searching for traps? Yeah, it keeps you safe if you guessed right and there IS a trap. But if you guessed wrong and there isn’t, you just got yourself an extra ticket in the random encounter lottery because of the time you wasted. Secret doors might hide shortcuts or extra treasure, but every search for a secret door is another crap shoot. You might find a door or you might get jumped by a level three monster your level one ass isn’t ready for.

Now, D&D and Pathfinder have evolved a lot. They aren’t just about delving into dungeons anymore. And they’ve shed a lot of the more finicky and complex rules to make themselves more approachable. But the random encounter tables in the GM’s Guide today are basically identical to the ones we were using back when we played ADVANCED Dungeons & Dragons and TSR wasn’t shy about putting the edition number on the cover.

Once upon a time, I wrote an article all about what an encounter actually IS. And I pointed out that an encounter – a good, modern encounter in a good, modern RPG – isn’t just a monster’s name and hit points. A good encounter presents a dramatic question and has a source of conflict. Moreover, a good encounter invites players to make decisions. And that’s because the game is actually ABOUT making decisions. That’s what role-playing is. It’s the act of deciding what your imaginary character does in response to a hypothetical situation while in the pursuit of a desirable goal. If you believe that – and you should, because I’m saying and I’m right – if you believe that, then how the hell could you possibly think a list of random monster names is good enough to define a random encounter. I mean, just because they are random doesn’t mean they get to NOT be encounters.

At the VERY LEAST, a random encounter table should list things like “a hungry giant spider thinks the party might be good eatin’” and “a goblin patrol ambushes any non-goblins it encounters and tries to run them out of the goblin’s territory” and “a sinister goblin surrenders to the party and then feeds them false information in the hopes of leading them into a deathtrap so it can take their stuff.” That, at least, presents a basic dramatic question and source of conflict. “Can the party escape or defeat the spider before it eats them?” “Does the party survive the ambush, and can they continue their exploration of the goblin caves?” “Do the heroes see through the goblin’s trick or do they fall prey to the deathtrap?”

And, hell, that’s still just assuming we’re on the most basic type of D&D adventure ever: the dungeon crawl. Even if all you’re doing is a crappy and very fun dungeon crawl – because they are fun, you stuck-up gaming snob – even if all you’re doing is a crappy dungeon crawl, your random encounters still need to be encounters. And that means the entries on your unnecessarily large table should be more than a monster name. But they don’t have to be much more.

But they can be. It just takes some planning.

Planning the Random

Allow me to blow your freaking mind here by pointing out something that apparently no one other than me has ever been smart enough to realize, random encounters don’t have to be random encounters.

“But Angry,” you say, “didn’t you explain in the last article that random encounters don’t do the things they are supposed to do if they aren’t random?”

First of all, I’m not doing that schtick again. No arguing with hypothetical readers in this article. My doctor says my blood pressure is high enough as it is and one of you hypothetical readers is going to give me a stroke if I don’t calm down. Second of all, what I said was that the OCCURRENCE of random encounters has to be random. The actual encounter itself doesn’t have to be. Because that part doesn’t face the players. And even if it did, it wouldn’t matter.

You don’t get anything – anything at all – that does any good for the game by having a list of potential encounters and randomly determining which one actually happens. And you actually lose some things if you do.

And no, the GM “being surprised” by random encounters and “not knowing what will happen next?” That’s not an actual, valuable thing. You already get that. You get that because you don’t know what the players will do, and you have to respond to whatever stupid things they do in whatever situations you create. Not knowing which situation will happen next doesn’t add anything. And if you think it does, you’re wrong. Be a player if you need to be surprised by the plot point. Your job is to run a good game that tells a good story. Authors don’t draw plot points out of hat because they understand things like narrative and story structure and crap like that.

Random encounter tables with dozens of entries from which a single encounter is selected? They make it harder to plan and run good games. Why? Imagine this: you have a short dungeon crawl. Six to eight potential encounters. You expect it to take one to two sessions. And you’re going to use the standard “1 in 6 of a random encounter every ten minutes” plan and then roll for which encounter happens on a table. How many possible random encounters do you need on that table?

Go ahead. Figure out your answer. I’ll wait.

If you’re like the average GM – and you probably aren’t, I have to grudgingly admit – if you’re like the average GM, you probably said something like eight or ten or twelve. Those are good numbers because you have random number generators in front of you that generate numbers up to eight or ten or twelve. And because that gives you about two or three times as many encounters as you’ll probably need. And that seems to be where GMs gravitate when they are coming up with options for a random table. Seriously. It’s weird, but it seems to be a pattern. Rumor tables? Random treasures? Random NPC names? GMs tend to always generate about three times the number they think they need.

Now, that seems smart. If you’re going to roll for something at random, you want to pick a thing from a big list of options. If you’re only going to use one thing, flipping a coin between two options just feels silly. You’ll want three. Or even four. And in a two-session dungeon-crawl with between six and eight potential encounters, you’ll want three or four random encounters. So, you’re going to want to select them from a list of nine to twelve. Or eight to twelve. Rounding off the to nearest die.

That’s also batshit insane.

Seriously.

Look, in that adventure of six to eight encounters, you’re going to want about three or four random encounters to POSSIBLY show up right? It might be only two. Or one. It’s random. But you really don’t want the adventure to be dominated by random encounters. So, even if there is no inbuilt limit on the number of encounters in your random encounter system, you’re going to stop flinging them at your players once you hit four or more. And that’s good GMing. After all, at that point, you’re probably adding another entire session to your adventure. Maybe two. Which means the adventure will drag on too long. Point is, you’re never going to use more than four random encounters.

WHY THE HELL WOULD YOU DESIGN TWELVE OF THE THINGS IF YOU ARE NEVER GOING TO NEED MORE THAN FOUR?!

Do you see what I’m saying?

And the thing is, the reason random encounters are just lists of monsters is that GMs need to populate lists of eight or ten or twelve of them. That’s a lot of work. Planning twelve encounters for a dungeon that has only eight possible planned encounters means you’re spending more time on random encounters than you are on fixed encounters.

Imagine if you only had to plan four encounters. You could spend more time on each one. You could make them more interesting. At the very least, you could provide sources of conflict and dramatic questions. You know: the bare minimum stuff you need for an actual, good encounter?

And that is the first trick to designing GOOD random encounters. It isn’t the encounter that’s random. What’s random is when they occur, where they occur, and if they occur at all. They are basically optional encounters that can happen anywhere at any time.

Roving Plot Points

Imagine you’re running this dungeon adventure. It’s a simple dungeon crawl. The ancient shrine has a treasure somewhere in it. The heroes are looking for the treasure. But there’s a twist. There’s someone else looking for the treasure too. They are also wandering around the dungeon. The party could encounter them anywhere. At any time. Or their paths might never cross.

Does that sound like something? Yeah, it sounds like what a random encounter really is. An optional encounter that might happen anywhere at any time. And it’s far more interesting than 1d4+2 goblins. Because it’s specific to the plot. In fact, it’s a plot point. Encountering the antagonist, you’re racing toward a goal? That sounds like exactly the sort of thing I’d called a plot point if I were writing another article about that sort of narrative BS.

Every adventure contains at least a few plot points – reversals, turning points, complications – that can happen at any time in any place. Or might never happen. And those are great fodder for the three or four random encounters you’ve got prepared for your adventure. And, the neat thing is, because random encounters provide a sort of natural tension curve and also connect the players’ action to the rise of tension, it makes sense that they trigger plot points. Plot points are supposed to happen when tension hits a high point.

And that is the first BETTER way to use random encounters. Fill them with optional plot points that can happen anywhere at any time and change the nature of the story. Hell, you can also have an entire plot arc that is told in roving plot point encounters.

Take that same adventure I just described. Here are your four random encounters:

Encounter one: the party encounters a sabotaged trap or a group of dead monsters that have been looted. They know someone else is in the dungeon. Or if they suspected someone might be trying to get to the treasure first, they now have confirmation.

Encounter two: a chance meeting with the antagonist and his party. There’s a tussle. The antagonist runs. Maybe it turns into a chase or a tracking situation. Or the party decides they are better off continuing their search and racing for the treasure.

Encounter three: a group of monsters ambushes the party. They discover the antagonist set them up. Maybe he bribed the monsters or convinced them to join his side. Whatever.

Encounter four: the heroes encounter the antagonist on his way out. He has the treasure, he’s running, and his monstrous allies jump in to kill the PCs to cover his escape.

Whoa. Holy crap. Do you know what you just did? You turned random encounters into a way of gradually advancing the plot. As the players waste time and do stupid things, victory gets farther and farther from their grasp and the antagonist becomes more dangerous. Wow. That’s not just interesting, that’s useful. It makes it easier to keep track of the failure state of the adventure. That’s always hard.

And what if the heroes hunt down the antagonist or kill him during encounter two? Well, then they can just explore at their leisure. Or you can have a backup encounter or two planned to replace three and four with a more generic guardian or monster or trap encounters. Yeah, you’re planning a few extra encounters in that case, but that’s okay because these encounters actually matter. They aren’t like random encounters at all. They have consequences. And their occurrence – and the ramping up of the plot – is tied directly to what the players do.

But roving plot points aren’t great for every adventure. And I did say that you can have encounters that serve the story without them having to serve the plot, right? Let’s talk about my recent adventures in space.

The Narrative Micro Cosmos

Back in the Long, Rambling Introduction™, I told a story from my real life. And because I like to keep the Long, Rambling Introductions™ option so I can make fun of the people who complain about them, I’ll review the relevant points. I was traveling through space and ferrying some goods for a business contact when I came upon the debris of a wrecked spaceship. There was valuable cargo in the debris. And there were also survivors in life pods. If I wanted to rescue the survivors or collect the valuable cargo, I’d have to dump the goods I was carrying for my business associate. I couldn’t save the survivors AND salvage the valuables AND complete my job. I had to pick.

And yes, that really did happen to me. It happened while I was playing Elite Dangerous a few weeks ago. I was briefly addicted to that game. Basically, the game is about you having a spaceship. And that’s it. You have a spaceship. Have fun. You can run cargo, get into space battles, salvage wrecks, explore the galaxy, get into wars, whatever. And while you’re tooling around the universe in your spaceship, the game generates random encounters and throws them at you.

Not long after, I found myself playing the Firefly board game. And that was a very different experience. Basically, that game is about you having a spaceship. You can run cargo, ferry passengers, commit crimes, salvage wreck, whatever. And while you’re tooling around the universe in your spaceship, you pull cards that represent random encounters.

It’s the cards that make the games different.

Now, the random encounters in Elite Dangerous get stale quickly. I admit that. They are all variations of “there’s a thing emitting a signal and it’s probably valuable, so you can drop out of warp and salvage it or whatever.” It’s mostly just finding random treasure lying around. But that situation I found myself in was more interesting because it presented me with a dilemma. I wanted to play the game as a “good person” and therefore rescue people in need, but I also wanted to advance my reputation and become a highly ranked smuggler. And I couldn’t have both. So, I had to choose which was a priority. And no matter what, the choice was going to cost me something. And that’s the important bit.

Now, let’s talk about the random encounters in Firefly. The random encounters in Firefly almost always present a choice between two options. And they almost always cost you something. Let me give you a couple of examples.

First, you might encounter a distress call. You can respond to the distress call and collect a reward, but it costs you extra fuel and ends your turn. And if you’re in the middle of a job or are low on fuel or are near a hostile ship, those can be very costly options. On the other hand, if you ignore the distress call, any of the members of your crew that have the “moral” trait become disgruntled. That’s a game state that makes your crew less effective and allows other players to hire members of your crew away. And if they become disgruntled twice, they abandon you.

There’s a similar encounter called Debris Field. Basically, you find the remains of a destroyed ship. If you have a skilled pilot, no big deal. Ignore it. Otherwise, if you want to keep moving, you have to spend extra fuel going around it. Otherwise, you have to end your turn. But, you can also see if there’s anything valuable to salvage. That requires you to make a skill check, but it also ends your turn. If the check fails, you found nothing. If the check succeeds, you found something useful.

Then, there’s one more. It’s called Family Dinner. That mirrors the first episode of the television show, Firefly, where the crew and their passengers eat together and get to know each other, and Friar Tuck breaks open his luggage and shares his fresh strawberries with everyone. Basically, the card presents you the option of expending some of your valuable cargo to have a nice dinner with your crew. The cargo – a useful resource – is spent, but any of your disgruntled crew become happy again.

Now, the thing about all of those encounters is that they actually tell miniature stories. They are Narratives in Microcosm. They present an inciting incident – you get a distress call or encounter a debris field or you have some quiet time to sit and eat – and then present several desirable things that each cost something. Answering the distress call offers a reward but slows you down and costs you extra fuel. Ignoring it angers your crew. If they care about such things. Salvaging the wreck offers a chance at a reward, but only a chance, and either way it costs you time and fuel. Dinner allows you to make your crew happy, but only at the cost of valuable supplies. Or you can just keep going. And like all good stories, the choices reveal something about the characters involved. Or they provide a payoff for the character’s traits. If you ignore the distress call, only you know the reason why for sure. But it does make you look like a heartless bastard. At the very least, it says you’re not prepared to take risks and make sacrifices to help people in danger. And if you skate on by and the crew doesn’t care, it reveals something about the people who work for you.

I’ve played several games now that offer these sort of “either/or” type random events. And the more I play them, the more I recognize that that’s actually a good way to plan a random encounter. In general, role-playing games are very open-ended. The idea of building encounters around that sort of binary “this or that” choices is pretty limiting. But, when it comes to random encounters, that’s just fine. Because random encounters are smaller story moments. They should be moments where the party just has to deal with a single choice that reveals something. Or allow the players to take a risk in return for a possible reward. If all the players do is take the choice at face value, then, they make an interesting choice and deal with the consequences. If they try to think outside the box, they might turn the encounter into a complicated scenario in which they try to find a way to “have it all.”

Imagine the sorts of simple choices you can set up as random encounters.

For example, the party is traveling across the countryside when they come along a stretch of parched, barren land stretching before them. Not a vast wasteland, but just a parched region having a really bad summer. There’s no food, little game, and little water. It also lies right across the most direct route to their destination. They can skirt the border of the parched land. But that will take them a few days out of their way – as far as they can tell from what they can see and their knowledge of the area – or they can go straight through. But, in the parched land, they won’t be able to forage and there’s very little cover if they do encounter anything dangerous because of the dead vegetation. A simple choice. A riskier, faster path that will force them to use the food they have on hand or a safer, longer path where they can forage for food but which will take them far out of their way and possibly cause more random encounters.

Now, the party might decide to spend a day foraging for food to build a good supply and then strike out across the parched land, compromising between their choices. That’s okay. That’s smart. At least the encounter provided an interesting challenge.

The party could encounter an injured creature in the dungeon. A creature that would normally be a foe, like a goblin, say. It is barely alive and in need of healing. They could ignore it. Or kill it. Or they could spend healing resources on it. Then, it might provide information. Or it might betray them. And good-aligned characters should wring their hands a bit about the whole thing. For that matter, it doesn’t have to be a foe. Imagine an injured animal caught in a hunter’s trap. The creature is badly hurt, but it is also crazed with pain. It might attack anyone who approaches. But can the party just leave it to die? And if the party frees it, can they also heal it or are they just ensuring it’s going to die in the wild?

The key is to start with a simple choice. Then, figure out why each option is attractive and desirable. Then, figure out what the party has to pay to make the choice. And that price has to matter. In point of fact, a random encounter should always – ALWAYS – involve a cost. A random encounter that has an overall good outcome isn’t doing its job. Because its job is to manage tension and make the players like the adventure gets more dangerous the less efficient, they are. Rewards and benefits from random encounters should be minor and they should always carry a risk and a cost. The salvage operation might pay off with some useful gear, but it also spent you a bunch of fuel and wasted a bunch of time. Going around the parched land means more potential for random encounters. Going through means you can’t live off the land and if you run out of food and water, you’re going to start to starve.

Making things Complicated

And that brings me around to one more possibility for random encounters. And the reason why I don’t call random encounters random encounters. Sometimes, random encounters are just sucky occurrences that make life harder. They use up resources or they impose difficulties. I’ve used random encounters in overland adventures like “oops, some of your food is infested with worms” or “you’ve got leeches” or even “you caught a cold.” Yes, I have inflicted illness and ruined rations with random encounters.

Because, to me, a random encounter IS a complication. It’s something that makes the adventure harder. Its rising tension personified. And the players can’t even call those things screwjobs because, if they’d hurry their asses up, they wouldn’t end up suffering so many of them. Stop dawdling in the swamp and you won’t get leeches or get sick from exposure. It’s no different than pickpockets in the streets of a crowded city, really. That’s just another random screwjob. The important thing, though, is that every complication has to happen directly to the party. Otherwise, you lose the vital connection between the players’ actions and the random consequences. It’s not fair to increase the size of an encounter that the party will have in an hour because a random encounter came up. How would the players even know the encounter was different? Random encounters are a very player-facing mechanic. That’s also why you don’t use them to advance plots in the background unless there’s a way the players can see the plot advancing. You can’t use random encounters to move the cult one step closer to completing their ritual unless the party can see the swirling vortex gate overheard suddenly swell in size. And even then, it should disgorge some demons that fly down to their location and attack them.

Now, if you feel bad about planning outright screwjobs as so-called random encounters, remember that you’re planning these encounters, not doling them from a random table. So, you can make sure if you’ve built in a “food goes bad” random encounter, you can build extra opportunities to find more food in the adventure. If the heroes can manage it. Because simply undoing the encounter also ruins the whole system. Giving the players the opportunity to spend resources and come up with clever plans to solve a complication is good. Inflicting a complication and then saying “JK guys, here’s some new food” is not.

Complications are also a great place for “a hungry monster shows up and wants to eat you.” That’s where I use those.

The Big Picture

Having now said all that, let me tell you how I use it my table. Most of the time. First, I figure out how many random encounters I need for a given adventure or session. Generally, I have half as many random encounters as planned encounters. But I have also increased the number of random encounters to build more random adventures. Especially if the adventure is a wilderness adventure with a big overland travel component. Also generally, I know I will get through about five or six encounters a session and I don’t want more than half of them to be random encounters, so I never need more than about three random encounters a session. Unless, of course, it’s an adventure of random encounters.

If the adventure is only going to be one or two sessions long, I use a mix of Complications and Binary Choice Random Encounters. Complications chew up resources relevant to the adventure. Using up food and water in a dungeon adventure, for example, is not useful. But using up hit points and healing resources is useful. As is breaking or ruining dungeon supplies. A simple one is a climb or a swing over a pit which requires the party to leave their rope behind because they can’t untie it without going back. Of course, parties often find ways to get the rope back. Fair enough. That’s the game. I might also work a Plot Point encounter in. But, no more than one per session.

If the adventure is going to be three sessions long, I will see if I can work a three-part Plot Point Arc into the list of random encounters. I spread those out to make sure they don’t all happen at once and usually make sure the last item on the list is the conclusion of the Plot Point Arc.

If the adventure is going to take place over several environments, I might have to come up with different random encounters for the different parts of an adventure. And then, I usually come up with one or two extras. For example, if the party will spend half a session traveling to a dungeon through the wilderness and half the session in the wilderness, I’ll generally have four encounters, two encounters that are good for the wilderness and two that are good for the dungeon.

I’ll compile those encounters into a big ole list and generally work through them in order. Which is why I make sure to carefully spread out Plot Points throughout the list. But I will adjust encounters slightly for logic. For example, if the party makes a loud noise and triggers a random encounter roll, that will result in a monster ambush type complication roll. And if I don’t have enough of those, for my stupid, reckless party, I will create a few more on the fly as needed.

Of course, sometimes, I don’t bother with all of that and just show up with a list of four very-designed “hungry monster wants to eat you” or “goblin patrol tries to run you off” type combats. Because it turns out, that crap is actually fun too. And sometimes, a good fight is all the game really needs.