This entry is part 1 in the series Creating ecology-based random encounters

Photo by Sara Moses (Sarame287)

Encounter tables seem to have gone out of fashion lately, and I’m not entirely sure why that is. Perhaps its a trend away from random or “wandering monster” encounters in a wilderness setting in favor of planned encounters – the latter can be defined as encounters that advance the plotline, while the latter are often viewed as meaningless, or worse yet, as sources of cheap XP for the characters. Perhaps its simply the spontaneity of the random encounter, which almost certainly means that they are less developed than a non-random encounter. Perhaps its that healing is thought to be too easily obtained and the old-school function of random encounters – that of wearing characters down to increase the drama and tension of the planned encounters. Perhaps it’s just that our lives have all become more crowded, and we have less time to game, and less tolerance for “non-contributing” encounters. Or perhaps its because the simulation aspects of gaming have become slightly more dominant in the last decade, and combats take longer to resolve as a consequence – perhaps too long to be tolerated in a climate of shortened attention spans. Or, finally, perhaps its because all-too-often, a random encounter table seems cobbled together at the last minute and detracts from the standards of verisimilitude that we seek in our gaming.

None of these objections are good enough. All these flaws can be overcome with surprisingly little judicious prep, and in the first sections of this three-part series, I’m going to show you how to do it – and why and when it’s worth the effort. Part two, “This eats that,” (next week) will focus on how to create better, more useful encounter tables. Part three, “Encounters with meaning,” (two weeks from now) will conclude the series by offering a technique for crafting better results from your encounter tables and integrating them into your plotlines in a meaningful way, and how to adapt the principles to urban and dungeon settings.

The Traditional Function

Here’s how it used to work, back in my AD&D days: The PCs would gather in an inn or tavern, all charged up and ready to go. They would learn the location of a dungeon, and decide that it was worth exploring. On the way, they would have various random encounters that exhausted or diminished their available resources – consuming healing potions and prepared spells and eroding the charges in wands and what-have-you. In compensation, they would receive a token amount of treasure. The degree of consumption would set the parameters for their dungeon crawl, because they could never be certain of uninterrupted rest to recharge, and the amount of quick healing from external sources that they had on hand had been eroded en route. When they reached the point where they were no longer confident of being able to make it back to home base, they would set out for the trip home. Now, the random encounters that they had experienced en route to the dungeon assumed a greater significance; not only were they are relatively low power, but they might also face additional encounters from “people” they had ticked off in the dungeon, at least for the first few days. Furthermore, they were relatively heavily encumbered, slowing their movement, giving more time for random encounters to occur. In compensation, they had the experience and knowledge that they had acquired on the trip out, enabling them to avoid some of the encounters, and of course, they had depleted the number of creatures out there to encounter on that trip out. Still, the number of times that I got them back to home base with only a hit point or two each made them properly cautious and respectful of the environment around them.

Then they started to get smart. Instead of being eager to explore the dungeon as quickly as possible, they decided that it was better to take their time on the outward leg, relying only on renewable resources. They increased their stockpile of non-renewable resources to the point where they were almost as heavily burdened on the trip out as they anticipated being on the trip back. Instead of trying to carry all the loot at once, they started burying caches on their way back, progressively lightening their load – the least-valuable items first. This meant that not only could they explore more of the dungeon per trip, but that they could expect greater profits per trip – easily sufficient to replenish their stockpiles, enabling them to make one or more trips to specifically to retrieve their caches. Slowly, they found ways of diminishing the risk.

At the same time, their power levels were rising, making them more self-reliant. Greater combat prowess made them able to defeat casual encounters more readily and at lower cost in hit points; they had more hit points at their disposal, anyway; and the party healers were better able to supply the demand for what healing was required. Their capabilities were expanding exponentially, and random encounters eventually became more trouble than they worth. At first, they were tolerated, but as the situation continued, the complaints became more frequent and more vocal.

I thought about an escalation in the difficulty of the random encounters, but a couple of experiments in that direction quickly convinced me that this was a dead end. The first was that these encounters provided still greater levels of treasure, exacerbating the general problem more quickly than they resolved it. The second was that they quickly became improbable to the point of being ridiculous. “Okay, so we somehow managed to avoid noticing or being noticed by four blue dragons? I don’t buy it.” The entire system was beginning to break down.

The final straw came when a party earned more experience and got more loot from the random encounters going to-and-from the dungeon than they did within it – and returned to home base with full hit points, leaving the wilderness because they had exhausted their capacity to carry more treasure. They started talking about buying wagons and recruiting lesser adventurers to guard them so that they could remain in the field for longer, and avoiding dungeons entirely – the risk-vs-reward wasn’t as favorable as dealing with random encounters.

2e and 3.x change the balance for the worse

Many years later, when I started my Fumanor Campaign, I was persuaded – somewhat reluctantly – to use 2e as the game system. The campaign was designed to be run using AD&D, a system for which I had worked out solutions to the imbalances that led to these problems, or so I thought. The combination of more experienced players, rules changes between the editions, and a GM who was completely inexperienced with the new rules system soon produced results that were even more drastically out-of-kilter. Things only became worse when – after an abortive attempt to translate the campaign into Rolemaster – we went to 3.x as the game system. The reason was that healing became more readily accessible. When combined with a lower play frequency – something I’ve addressed separately later in this article – tolerance for random encounters began to wear very thin once the PCs topped 12th level.

At low character levels, random encounters work as they always have. As character levels rise, the familiar problems I’ve described return, even more rapidly than they used to.

Phased Encounters

My solution was to phase random encounters out completely unless they were plot-significant – the chance of a roll being required 20 minus the average character level. Beyond this, encounters were dismissed as trivialities not worth mentioning. At best, I would include them in a daily/weekly synopsis of their travels, compressing the entire journey from A to B into a single descriptive passage. This had the unwanted effect of ensuring that the players knew that any encounter I did not hand-wave was significant, but that was a relatively small price to pay.

The XP Giveaway

One of the perceived problems with random encounters, especially at mid-to-high levels is that they are an XP giveaway. In pre-3.0 games, this perception is completely valid, a problem that has to be addressed. In theory, this is resolved by the 3.x system of awarding XP based on a theoretical risk-vs-reward scale – an encounter is worth less XP if the characters are a higher level, and eventually reach the point of being worth nothing. However, my experience is that the diminishment of xp is geometric in nature (or, more properly, the increase going in the other direction is geometric) while overall character capabilities increase on an exponential scale. To be truly balanced in this respect, characters gaining a level would have to choose between additional HP, improved combat capabilities, or dedicating the level towards an increase in class abilities. Spellcasters would have a fourth choice – to increase the number of spells they received.

But the game system is what it is, and my players would lynch me if I tried to implement anything so draconian.

The point – and the reason all this is relevant to the current discussion and not relegated to a sidebar – is that it still gets progressively easier for the characters to earn the XP on offer in the form of random encounters – far faster than rising character levels erodes the value of these encounters.

There are two solutions to this problem.

XP reflecting the encounter significance

In July two years ago, I proposed the concept of objective-oriented experience points. This approach solves the XP-Giveaway problem immediately – if the encounter doesn’t contribute to the characters achieving the plot objective, it’s worth nothing to them; at best, a random encounter is just another roadblock that the PCs have to circumvent or overcome in order to achieve their goals.

But even if a more traditional xp structure is in place, there is no reason why a partial solution can’t be adopted to this specific problem. The first solution is to make the XP reflect the significance of the encounter, and then adjust the difficulty of the encounter to conform. This recasts the XP problem as one of encounter planning, where – perhaps – it belongs.

For example, if you have an encounter indicated by a random encounter table that should be worth an estimated 500xp but the significance would be more commensurate with 100 xp, then you adjust the difficulty accordingly. Give the PCs a passing ally. Let the creatures being encountered be already wounded. Assume that there will be circumstances in the encounter that prevent the creatures from using one of their more dramatic racial abilities. Any one of these could be enough to halve the XP value of the encounter; all three in combination reduces it to something like an eighth, which is about 63 xp. Any two of them would drop the value to about 1/4, or 125 xp. Either of these values is close enough to the 100-point target, especially if you keep track of your overs-and-unders and add those to the next random encounter XP target.

Encounter significance reflecting the XP

The alternative is to ramp up the significance of the encounter – possibly in terms of the current adventure, possibly in terms of a more long-term development – until it is commensurate with the XP that will be earned. If the encounter has a book value of 500, and you think that as a purely random encounter it should only be worth 100xp, then you have 400xp worth of significance to load in. Perhaps the creatures are chasing/hunting an NPC who never appears in this adventure, but who will provide the hook to a future adventure. Perhaps the encounter itself is a harbinger of greater activity by that species or race within the campaign. Perhaps the creatures have attempted (or even partially succeeded) in looting the same dungeon the PCs are heading to, and have escaped with a book detailing some of the background of the place (which the GM would have to read anyway), or one of its treasures (one that’s worth about 400xp).

Either solution yields the same effect: the XP received becomes commensurate with the significance of the encounter.

Personally, I find the second solution to be less work and a greater spur to creativity, so I always attempt to employ it first; if that doesn’t work, then I look to reduce the XP value of the encounter.

Note that you don’t have to explain these adjustments to the players. In fact, the results are arguably better if you don’t – but have the reasons become self-evident later.

The Loot Giveaway

Most monsters have treasure – somewhere. Some have a trivial amount, others have lots. It does the GM no good to carefully moderate the treasure that he places in a dungeon if a random encounter table throws additional goodies into the mix.

Don’t let your campaign be held to ransom by the results of random encounters. Having fixed the XP problem, let’s look at ways to solve the Loot problem.

Enhancing the encounter significance

I’ve already touched on this solution when I suggested using the encounter as a “mule” to bring part of the background narrative or treasure from the dungeon to the PCs in advance of their arrival at the dungeon. But there is a further way to apply the concept: if the dungeon is going to require a magic item that the PCs don’t have, for example something that permits an attack against an ethereal foe, having a random encounter furnish just such an item can make the difference between a great game and an exercise in frustration.

Any of these three approaches provides loot in the random encounter that does not significantly unbalance or destabilize the game or the adventure – provided that sufficient care is taken in choosing the goodies in question.

Value reflecting the encounter significance

Another approach, one that works hand-in-hand with reducing the XP value of the encounter to some arbitrary value, is to reduce the loot that derives from it in a commensurate fashion. After all, if an encounter is only to be worth 100XP when the book says 500, and you have weakened the creatures encountered in order to achieve that 100XP target, it doesn’t seem at all unfair to reduce the treasure obtained by a similar ratio – while it does seem grossly unfair to have the loot have the full value that would go with a 500XP encounter, under those circumstances.

Remote Goodies

So your monster has a lair, and its supposed to be stuffed to the gills with goodies. Where is it? Is the monster smart enough to conceal it? Just because your players have won the encounter doesn’t mean they are entitled to the loot – at best they might get some of it and have bought a chance to look for the rest. And look, and look.

I once ran an encounter in which a dragon hid various bundles of hoard in 100GP units in various ways, then hired an adventuring party (the PCs) to try and find as many as they could. What they found, they got to keep – so they were well-motivated. It cost the dragon about 4,000GP – but at the end of the day, he knew which methods some very savvy PCs had not penetrated, and was able to hide the rest of his hoard accordingly. Another converted his entire fortune into statuary – then buried them upside down so that the statues looked like paving tiles. The PCs spent months searching while never realizing that the hoard was literally underfoot the whole time. A third found a hollow tree, stuffed his loot (in bags of holding) into it, then polymorphed the tree into a different variety of tree – one that have hollow spaces.

I’m not suggesting that every creature encountered will have a scheme to hide their loot that’s up to these standards – but they will all have been as clever as they can. After all, the lair is where the young are.

Potential vs. actual value

A favorite approach for random-encounter treasure is to furnish the PCs with a treasure that is worth quite a lot – to the right person, or if treated the right way. Sometimes it will also be grossly inconvenient or cumbersome in the meantime. Some goodies are naturally concealed, like a creature who uses gold flecks from creek water for roughage – it drinks the water, concentrates the gold into pellets, then passes these through its digestive system to help grind down the tough plants from which it derives its nourishment. The pellets are enveloped in the creature’s dung when expelled. When washed clean, these are worth about a silver piece each – but they are fragile and prone to returning to the gold fleck state when the dung holding them together is washed away. Over time, if the problem is properly tackled, you can earn quite a lot of money – but PCs never seem to do so. They are usually more get-rich-quick in orientation.

Little irritates a PC more than discovering that an undamaged pelt from a creature that they killed would have been worth $$$$ if properly preserved and not cut full of holes – but in the condition it’s now in, it’s only worth $. But the creature can still legitimately be said to have “loot” of $$$$ value – it’s not the creature’s fault that the PCs made it virtually worthless in the process of obtaining it.

While I describe this as a favorite approach, that is not to say that I use it all the time. I use it only when it makes sense. And I’m not above the PCs hearing false rumors from time to time – for example, that Bugbears have a pair of ivory molars at the back of their jaws for a brief period in adulthood.

Consumables and Irrelevancies

Finally, never forfeit to deduct the value of consumables and irrelevancies from the total before you start looking at how much actual wealth a creature has. A rare meat might be worth 10 GP per pound – and that comes to a lot when you’re talking about, say, a Bulette. But: How quickly does it spoil? How hard is it to preserve? How expensive is it to preserve? How hard is it to transport? What else will be lured to the vicinity by the smell of rotting meat?

I once gave a PC a “golden” berry, which – in fourty years – would grow into a tree whose leaves were real gold.

He ate it.

Random Relevance

By far the best solution to all the problems with random encounters is this: Make the a delivery system for something that matters. Information, Narrative, Background, Environment, Plot Development – whatever.


What do you want the PCs to know? If sentient, an encounter could parley that information for safe passage. Is there something the PCs want to know about? Ditto. If non-sentient, is there a way for the very presence of the encounter to provide the information? I once had some PCs trying to track down The Well Of Life to resurrect a fallen comrade. En route, they came across a pack of undead squirrels – and started suspecting that there was more going on than met the eye. Sure, I could have had a random NPC pop up with the information, or one of the Sages that they tried to consult – but it was a lot more effective when every encounter started having the word “undead” tacked onto the front of it. Undead squirrels. Undead Orcs. Undead trees. An undead Roc. Undead… mushrooms.

Exported narrative

There’s always a lot of descriptive narrative that you have to get across. Putting as much of it across as dialogue with some creature encountered is always preferable to simply reciting it from on high. Perhaps the Chimpanzee has been to the Temple Of Unmitigated Disaster and can describe it – if the PCs use an appropriate spell – instead?

Background information

At one point I had a lot of background information about a location to impart to the PCs, and was looking for a way to dress it up and make it interesting. When the random encounter table came up “Ghost” I found what I was looking for. One of the PCs found himself caught in the middle as two ghosts recreated the climatic (and unresolved) final battle between them. Instead of dry, third-person narrative, I was able to bring it to life for the PC.

On another occasion, I wanted a PC to find out what some elves were up to. I had a tree tell him. Why not? He was a Drow who had been converted to Corellan. It started him no end when the tree started talking to him, though. The next problem was getting them to be quiet – for a while, everywhere he went, the tree limbs would rustle in welcome and groan with gossip and innuendo – mostly about other trees and the wildlife around them. Even now, the character can be doing his best Stealth act when a tree will suddenly cry out in welcome.

Environmental awareness

If you create an ecology-based encounter table – and I’ll be talking about those extensively in part two of this trio of articles – encounters are the best way of bringing the ecology to the attention of the PC. Once they recognize the principles apon which your ecology is based, they will start anticipating, and a minor encounter (worth nothing) can serve as forewarning of a dangerous change of environment ahead. The more the PCs interact with their environment instead of simply passing it by without notice, the more real your world will be to them.

Plot development

I’ve touched on this one earlier. A single pair of Orcs – in war-paint and a long way from where they might be expected to be – can be the first signs of a new war of aggression, just a single raindrop can announce the arrival of a storm.

En route to a dungeon ruled by a necromantic sentient phase spider with umpteen levels of mage on the side, some PCs in one of my games began noticing that just before they were attacked by a wandering monster, there would be a peculiar light and a sense of spiderweb drifting through the air. Proceeding to investigate the next time it happened, they learned that the owner of the dungeon had scried their approach and put a bounty on their heads, arranging ambush after ambush as they approached. Suddenly, the “random encounters” had a purpose and a malevolence behind them.

A subplot dressed in random encounter clothing

This only works in urban settings where virtually every encounter is with something sentient. Take a simple story and divide it into small sections – five to ten of them. Have that story happen around the PCs, never involving them directly, but always connecting indirectly with them through random encounters. Or invent a narrative as you go, using each random encounter to advance that side-plot. Since the players won’t know that it’s an entirely separate plotline to the main plot, they will have a lot of fun spinning spiderwebs and conspiracy theories from moonbeams.

The imperative of play frequency

There’s a lot that you can do when you play weekly, or even fortnightly, that you simply can’t do when you only play monthly. There’s far more imperative to “get on with the story” with less frequent play. To some extent, the approaches to encounters that I have described here were developed out of the necessity to give random encounters a reason to be noteworthy. Of course, it helps to have cultivated a reputation for dropping obscure clues years ahead of their becoming relevant – and of playing it straight when the players try to make sense of these veiled hints. The advantage that the PCs receive as a result is fair compensation for the additional investment in thinking about the campaign that the PCs have put in.

The more often you play, the more often you can afford to lose a quarter- or a half-session of play to a meaningless encounter. Finding the level that’s right for your circumstances is all important – because these techniques are all the more powerful in moderation. Used all the time, they can lose some of their oomph – but that’s still a better choice than the alternatives.


UPDATE 5/4/13:

I tried, folks, but I simply wasn’t able to get part 2 finished in time. So for the second time in 5 years, I’m afraid there will be no post when one was scheduled. I could have taken time off from the article to run up something to appear this week but doing so would mean that it still wouldn’t be ready next week. So I decided to bite the bullet. Normal service will be resumed ASAP!

Related Posts with Thumbnails
Print Friendly