What is the purpose of random encounters?

[rel]

Random encounters were, when first introduced, absolutely part of larger minigames of resource management alongside daily resources, limited supplies and consistent danger.

likewise, I think its reasonable to say that a lot of systems now include those traditional design elements without considering their original purpose or what they might bring to the game but simply because they are common in other games and have a long history.

And when originally quite specific mechanics are blindly included in a game because of tradition, you will often run into problems, because said mechanics are unlikely to be a good fit for the system, and might be incomplete.

The OP asks what random encounters are actually for, given that they seem to actively reduce their fun at the table.
My answer is, that these days such mechanics often serve no purpose, vestigial elements of more complex gameplay loops that no longer exist.

Now you can certainly mod those old systems back into a game that doesn't have them, or adjust the existing mechanics or your gameplay until they fit one another, but a built for purpose mechanic will do a better job.

For you, perhaps.

I... don't think I understand this response.

As I pointed out, a formal encumbrance system is also in Blades in the Dark. (And it's a bit less fiddly than D&D's).

I'm sure it is. But what does the encumbrance system in some specific RPG have to do with the current discussion, or the OP's question for that matter?

[KorvinStarmast]

But what does the encumbrance system in some specific RPG have to do with the current discussion, or the OP's question for that matter?

You were complaining about old mechanics. The encumbrance is also an old mechanic/feature.

[Cactus]

Gbaji, I think Thane of Fife was presenting the thought process that leads to the tedium of narrating every hour in order to show that random encounters are not necessarily the cause. Thus this bad GMing technique (which I'm guilty of too) shouldn't be held up as a reason why random encounters are bad.

At no point did they say this was how they run or recommend anyone runs a game. You don't need to deconstruct why this is poor technique, everybody's in agreement on that point.

[gbaji]

Gbaji, I think Thane of Fife was presenting the thought process that leads to the tedium of narrating every hour in order to show that random encounters are not necessarily the cause. Thus this bad GMing technique (which I'm guilty of too) shouldn't be held up as a reason why random encounters are bad.

At no point did they say this was how they run or recommend anyone runs a game. You don't need to deconstruct why this is poor technique, everybody's in agreement on that point.

Ah. Ok. I think I misread that post a bit then. I think I got to the last bit about arriving in the same place, and interpreted that as saying "so this is just as good a method as any". Re-reading it, it's a lot more clear. I guess I was mentally following that right after the presentation of "sitting at the table rolling encounter tables while describing things, and dragging things out", and got them jumbled up a bit.

I will re-iterate my point that I'm not a fan of using random encounters as resource drains though. IMO, that's at least relevant.

[kyoryu]

I will re-iterate my point that I'm not a fan of using random encounters as resource drains though. IMO, that's at least relevant.

I think "deliberate resource drain" is a pretty minority position.

I think "pressure to keep moving" is more common, as well as "portraying possible but not definite risks".

[gbaji]

I think "deliberate resource drain" is a pretty minority position.

I think "pressure to keep moving" is more common, as well as "portraying possible but not definite risks".

Yeah. I can see that. Though, honestly, if I'm running the PCs though something where there is (or should be) some kind of "you need to keep moving" pressure, I'd prefer for that motivation to not be "or a random monster will show up and attack" and rather it be "or the denizens of the area you are exploring will report to their friends/allies/whatever that they saw you, organize their defenses, and come over and curb stomp you". Time pressures should be a natural reaction of the environment the PCs are in. If the PCs are assaulting the enemy stronghold, and they stop moving foward and attacking, then yeah, the defenders are going to organize, rally, and counter attack.

I does feel a bit artificial to me to use actual random encounters or wandering monsters for this though. Objectively speaking, there should be no greater odds (arguably lower odds) of encountering some random/wandering monster while spending X time sitting in a single location versus X time exploring. So the idea that "something showed up while you were sitting around in a single location, so you should hurry up and leave that location so as to avoid having such encounters", just seems counter intuitive to me.

I don't see the second part as actual random encounters (but can certainly be considered "wandering monsters"). That fits into one of the categories I think I listed previously as "guards/patrols/whatever of things that exist in the area". So, if they run into a group of NPCs roaming around, that clues them in that this particular kind of NPC is in the area, so when they encounter a larger group of, or location with, those same NPCs, it makes the area seem more dynamic.

I also think using actual random wandering monsters in this way may have a counter result to what the GM may intend. I would certainly assume (quite reasonably) as a player that there *should* be a fixed number of monsters in the area. And, if some of them are roaming around, I will almost certainly be better off encountering them while in a defensive position (like say where we holed up for the rest period), then while exploring the area itself. Thus, if the GM starts hitting us with random/wandering monsters, my inclination would be to think 'Ok, great. We can set up a kill zone here, where we are at our most defensive, and then wait until all the monsters that roam this area show up, and kill them one at a time using far fewer resources and at far lower risk then if we roam around ourselves". Then, having cleared the area of the roamers, we can safely explore the area without worrying about something randomly showing up while we're in the middle of doing something else.

To me, that sort of "actual random and unconnected monster" encounter is not going to incentivize to me move faster at all. It'll tell me "there are things wandering here, that you might want to take out before you risk encountering them at an inconvenient time". On the other hand, "connected monsters" (ie: things that are clearly part of a larger whole set of monsters), absolutely create a time pressure. Doubly so if I'm not really sure if we just killed all of them, or the group that just attacked us sent one of their number back to report on us. But yeah. I like to create whole areas as actual realistic eco systems that the party explores. So things aren't just there by chance. They're there because they "fit" in some way. That does not preclude some of the things in the area wandering around, but they will have a reason to be wandering around. They're hunting for food, or patrolling their borders/boundaries, or have some other purpose.

That's not to say that there can't be environments with wandering monsters that don't deplete (ghouls or other undead things poping up out of a doorway to the underworld in the area, and then roaming around, for example). But those are rare. Most of the time, if you kill something here, you wont run into the same thing later somewhere else in the area. So choosing where you kill something is often a pretty smart way to approach things.

[Pex]

Gbaji, I think Thane of Fife was presenting the thought process that leads to the tedium of narrating every hour in order to show that random encounters are not necessarily the cause. Thus this bad GMing technique (which I'm guilty of too) shouldn't be held up as a reason why random encounters are bad.

At no point did they say this was how they run or recommend anyone runs a game. You don't need to deconstruct why this is poor technique, everybody's in agreement on that point.

Correlation is not causation. I'm amenable to that. I can agree I've played in random encounters that proved fun in their own right, but too often random encounters are run via the "boredom method" even when the DM is not hiding he's using random encounters. The other issue of random encounters forcing players to waste needed resources is still a bother, but as a personal matter since I'm experienced in conserving resources knowing when to fire the big guns or not it's not that big a deal to me presuming the DM is not being adversarial. Not all random encounters are combat. The actual playing of one wouldn't be the issue. It's the boring part of DM monologuing until it actually happens, so the correlation is still an issue.

[Telok]

Many random encounters, at least the ones I run and the modules I adapt, are checked on a timer or per extremely noisy event near high traffic areas. I have one table with animals (hunting fleeing ambushing sleeping), local factions (patrol hunt trade), noises (animal people "natural"), a wandering crazy doctor looking for patients, a mercenary giant looking for booze, and an npc adventuring group. Roll 4d10, first is the 2/10 yes/no, next two are the table entry, last is what its doing.

The pcs holed up for a rest one time in an inaccessible place. There's a 2/10 chance per half hour or big noisy event. They missed meeting two neutral/friendly npcs, a patrol, and only had to deal with some trivial flying animals (aerial jellyfish, didn't even warrant combat, just sent the heavy armor & poison immune guy in to stab them each once). Pretty much every time they rip off a couple good long bursts from their guns I check for another wandering encounter. It only came up with another add-on encounter twice in the last six or eight sessions.

But recently they used full-auto firearms in a place where the locals don't have firearms. Ended up dealing with a 3-wave fight with a cabal of necromancers and blade zombies and when we return after xmas they'll have maybe a minute before the npc adventurers, mad doctor, and a double strength patrol of mutant clown-folk come down on them from different directions. That crap isn't random, its the result of ringing the "here's loot/dinner" bell a whole bunch.

[sandmote]

I just spent a bunch of time putting encounter tables together. Partially this was because it was for an area where I wasn't sure what monsters even could show up (and there definately aren't enough printed 5e stat blocks to fill it out), and since this I consider "what's the purpose" to be a necessary way to determine "how do I make them useful to me" I'm putting thoughts to paper here. Bit a stream of consciousness thing I guess, which I usually don't try.

Back to Random Encounters. Can some one explain how they make a GMs job easier, because I am not sure I see it. Can folks tell me more about how Random Encounters make things easier for a GM?

The first thing I get out of random encounter tables is that it narrows the scope of the question I want to answer. Sitting down and asking "what specific thing is the party encountering this time," is very broad and open ended. With a random encounter table, I an move on to more interesting questions.

---

Actually rolling dice for a table I have prepared: 5 on the d6. That particular d6 roll determines initial encounter type; so assuming the party is in the dismal delve on the plane of earth, that's a d20 roll. The d20 tells me which item of the rolled encounter table I'm looking at; I rolled a 9. That's an encounter with "2d4 mining slaves (veterans)"; I rolled 7 of them.

Now, instead of asking myself "what is the party encountering this time" I can skip to asking myself "what are these 7 mining slaves doing here." By setting up a table beforehand, I've skipped several questions and can move on to determining how to make the encounter interesting or useful or to say something about the world. This roll ended pretty rote (most of the results are supposed to make sense) but I managed it a lot faster than if I tried to make something up completely on the spot. It also helps when I roll for something less generic (running into a flail snail is on the same table) so I'm not always going with the most obvious option.

I will say, though, that the d100 tables a lot of books use (Xanathar's Guide has some examples) are far too broad to really work for an individual forest/mountain/plain. I broke my tables into four groups when working through what I think might be helpful. The numbers shown in front of each option are where rolls on the initial encounter table typically tell me to go:

• 1-2: social encounters with a die roll of d20 or d100 to choose which one shows up.
• 3: generic area wandering monsters with a die roll of d6 or d20 to choose which one shows up.
• 4-5: sub area wandering monsters with a die roll of d20 or 100 to choose which one shows up.
• 6: environmental encounters with a die roll of d12 or d20 to choose which one shows up.

Sometimes I know what I want the party to encounter. If I have news I want to remind them of or they've spent too long standing in one spot arguing, I skip to the social and sub area wander monsters respectively. I know what sort of encounter I need, so I skip the table picking what I don't need. Rolled a 5 for the former example, giving a result of 1 cult fanatic leading 11 cultists. Do I want them encountering that? If the party has been fighting a rival cult and these guys are willing to discuss targets to bring that cult down, then maybe. But that's a more specific question than "who do they encounter," so its easier for me to say I think it would go better if they encounter a pair of gahleb duhr. I rolled the party encounters two of them by the way; I just skipped to the result for gahleb duhr on the same table where I rolled the cultists. But "are the cultists what I want" is a much easier question than "what should the party encounter," so it helps keep me moving.

If I'm preparing stuff beforehand, having tables with social encounters, environmental encounters, and sub area wandering monsters (1-6 or 1-20 on the sub area wandering monster table is the generic area wandering monster table) just means I have a quick reference guide for what I think should show up in the area. But if the party is taking too long in game, it greatly speeds up my ability to get things moving if I roll on the appropriate table and throw 3 gargoyles at the party. I rolled 1 gargoyle, but that's probably underpowered, so I'll adjust the number. But as with all the previous examples, "how many gargoyles to use" is a more specific question than "what am I throwing at the party." A final roll for the number of creatures isn't always helpful either, but a 1 is going to bias my result downward, to an easy encounter rather than one that risks a PC's life.

I'd still probably try to use them the way gargoyles are supposed to be be used (hiding among inanimate statues to surprise people) if I can, but on the plane of earth its probably assumed all gargoyles are animate anyway, so a generic random encounter they can be.

With these tables I still need to DM, and knowing which creatures can appear in an area speeds me up, but assigning a die roll to the list of things that can show up still helps me come up with something that makes sense on the spot. It isn't a panacea to figuring out what's where (again, I still have to DM), but it helps.

So for a wider region with related environments, I have those four sets of tables, each helping me answer a different question:

1. Who is traveling through here
2. What are the most common threats
3. How are these areas different (with a separate table for each sub area)
4. What non-creature stuff is out here

And once I've got the tables set up, I know I have options to roll or choose when the party is going to encounter it. Presumably social encounters are friendlier and wandering monsters are less so, but not necessarily. The 7 slaves I rolled above could be working, they could be walking down the passage to a job site, they could be trying to escape. They could be hostile to being interrupted, they could be friendly and try to trade information, they could ignore the party completely. Questions I'd still have to answer for most encounters, but easier to answer for 7 miners than for all possible options of creatures showing up.

This also isn't the only place on the table something like them can appear: the social table has "A mining expedition of 1d12+1 veterans" and "1d4 sandmen with 1d12 slaves (bandits)" (bolding represents a stat block either in the books or which I've written). But the different prompts represent different things (a "mining expedition" might be dao slaves or not) so they still can pull me in some slightly different directions despite mostly making sure miners are more common on the plane than dao or dragons.

There's probably some additional ways to break up the tables to help with areas that contain a wider variety of living creatures or where the sub areas don't overlap enough to have united common encounters. But the work for making the table can be done before the session, saving me time during the session. And if I save more generic tables or the party comes back to an area, I don't need to repeat as much of the work.

---

I mentioned it above a bit, but I don't find those massive 1d100 tables with 50+ options very helpful. All the encounter types I break up above are mixed in, so I have a harder time when I know what sort of encounter I want bit still need to pick specifics. The Forest Encounters (levels 1-4) table in xanathar's guide is probably one of the most egregious? Given there's encounters that say X or Y (how do you end up writing "1 scout or 2d4 guards with 1d8 mastiffs?") there's about 90 examples on the table. What kind of forest simultaneously contains 2d4 woodcutters (commoners) or 1 yuan-ti abomination with an equal chance of stumbling on either? I think there's also 10 different encounters across the four forest tables that don't involve creatures, so the environment isn't playing much of a role in the...environment you're traveling through.

But we have that, because of course the tables are there to give the DM something to roll on, rather than a set of useful options for what the party might encounter. I'd rather have a d20 or d10 table for an "orc inhabited region (levels 1-4)" and another one for "goblinoid inhabited region (levels 1-4)" and use a different table to check the animals, but the listed environment types are things like "forest," so we're stuck with a "forest encounters (levels 1-4)" table where my chances of "an old tree with a wizened face carved into the trunk" are nill until they get higher level. But owls are as common as a CR7 creature which is (a) considered a "lethal" encounter for 4 PCs of the highest level the table is meant for and (b) has lore saying it should be travelling with a bunch of servants and lesser creatures.

Which I can still adjudicate as DM, but the "1d3 dire camels in a swamp" table doesn't do anything to help me skip questions I need to know about a off the cuff encounter the way focused tables broken down by encounter type do. And focused encounter tables don't necessarily limit what I can choose. The same way those 7 mining slaves can also be a social encounter, I can also grab something off a social encounter table and turn it into a fight. I've just given myself some guidance for the loot table based on whether the guys with the veteran stat block were rolled up as a social encounter or as a wondering monster, even if the table they originated from doesn't give strict instructions on how friendly they are.

Rolling the table doesn't necessarily make any individual encounter better, but having a proper set of tables with good options makes my encounters better on average; and I can generate them faster too.

[Pex]

Many random encounters, at least the ones I run and the modules I adapt, are checked on a timer or per extremely noisy event near high traffic areas. I have one table with animals (hunting fleeing ambushing sleeping), local factions (patrol hunt trade), noises (animal people "natural"), a wandering crazy doctor looking for patients, a mercenary giant looking for booze, and an npc adventuring group. Roll 4d10, first is the 2/10 yes/no, next two are the table entry, last is what its doing.

The pcs holed up for a rest one time in an inaccessible place. There's a 2/10 chance per half hour or big noisy event. They missed meeting two neutral/friendly npcs, a patrol, and only had to deal with some trivial flying animals (aerial jellyfish, didn't even warrant combat, just sent the heavy armor & poison immune guy in to stab them each once). Pretty much every time they rip off a couple good long bursts from their guns I check for another wandering encounter. It only came up with another add-on encounter twice in the last six or eight sessions.

But recently they used full-auto firearms in a place where the locals don't have firearms. Ended up dealing with a 3-wave fight with a cabal of necromancers and blade zombies and when we return after xmas they'll have maybe a minute before the npc adventurers, mad doctor, and a double strength patrol of mutant clown-folk come down on them from different directions. That crap isn't random, its the result of ringing the "here's loot/dinner" bell a whole bunch.

What game are you running? Serious inquiry. I know it's not your intention, but my inner cynic argh is tingling due to rolling for a random encounter when they use a weapon you said they are allowed to use. I know it's not every instance, but the possibility itself triggers my cynicism. It gives the appearance of punishing the players for using the weapon by possibly drawing more enemies they can't handle right now. Is this normal for the game or are you thinking it's a logical consequence and the players really aren't caring about the noise they make? As the saying goes, don't balance something by making it annoying to use.

[Lord Torath]

What game are you running? Serious inquiry.

Dungeons the Dragoning 40k 7th Edition. Check his sig.

[Telok]

I know it's not your intention, but my inner cynic argh is tingling due to rolling for a random encounter when they use a weapon you said they are allowed to use.

DtD40k7e is not balanced. It is not gentle. It is not fair. You can start out as a gnome Atlantean pyromancer starship captain with tamed riding T-Rex, wearing power armor and dual weilding ak-47s, on top of the starship you own. Legit starting character that.

They are in a long lost whacko dungeon of weirdness that a corporation tried to monetize a couple thousand years ago. Something went wrong, the place sealed up, massive explosions, etc., etc. The pcs opened it up again and found the descendents of the corp employees and original inhabitants who have been warped a mutated by the dungeon. They're basically in a d&d style dungeon with random technological artifacts and some SCP style anomalies.

The locals, in the levels they've explored are between stone age and early iron age technology. That are not, with the exception of a few unfortunate mutations, stupid. They hear gunfire and find the blasted corpses the party leaves behind, largely ascribing it to a new kind on noisy combat spell. Undead respond by their programming or hunger, if unintelligent, and as intelligent creatures otherwise.The animals are either hunter/scavengers or herbivore/scavengers and will cautiously approach noises if hunting or without a current food source.

Everything down to the giant pillbugs has been mutated or changed by the environment to be more aggressive, carnivorous, and hostile. And pretty much nothing is as noisy and attention getting as gunfire. So there's 3000 years of no guns, weird magic, and near starvation conditions. A new noise might be a new food. It doesn't help that the party can't effectively communicate with the natives or read much of the written stuff they find. They have enough recovered documents they could unravel the mystery, but the language barrier and their... disregard... for stuff that isn't precious metals, gems, or magic items, has been seriously hindering them.

[Pex]

Ok, a logical consequence of action based on game setting and rules used as designed. Telling my inner cynic to go to bed.

Thanks.

[Telok]

Ok, a logical consequence of action based on game setting and rules used as designed. Telling my inner cynic to go to bed.

Thanks.

No worries. If they were using laser guns, silenced pistols, stealth, or trying to talk to people then it would be different. And easier. But its rock & roll with gunpowder heavy weapons as their first reaction to almost everything. Not even assault rifles, actual "no rate of fire selector" machineguns.

They're combat monsters, and one has great intelligence skills, but talking to npcs is a big weak point for them. Murderizing stands them in good stead when they're just hacking & blasting primitives with stone spears and basic vodoun traditions. But the next assassin and bounty hunter (two characters took enemy & hunted flaws which certain game rules have triggered) have normal social skills & stats and are going to try getting chummy with their targets before the plasma grenades and two weapon frenzy chainswords come out.

[kyoryu]

I would certainly assume (quite reasonably) as a player that there *should* be a fixed number of monsters in the area.

That makes sense for a more limited dungeon, where "clearing" it is a reasonable expectation.

It doesn't make as much sense in the more original context of a megadungeon, where the dungeon is basically large enough that the idea of a known, finite number of critters doesn't make as much sense, and the dungeon "replenishes" itself over time between delves.

[KorvinStarmast]

That makes sense for a more limited dungeon, where "clearing" it is a reasonable expectation.

It doesn't make as much sense in the more original context of a megadungeon, where the dungeon is basically large enough that the idea of a known, finite number of critters doesn't make as much sense, and the dungeon "replenishes" itself over time between delves.

Also, how time is tracked, and how time passes (particularly when you are recovering one hit point per day unless magical healing is available) allows time for the "world" to change and for other monsters to happen by, occupy empty areas, etc.

[gbaji]

That makes sense for a more limited dungeon, where "clearing" it is a reasonable expectation.

It doesn't make as much sense in the more original context of a megadungeon, where the dungeon is basically large enough that the idea of a known, finite number of critters doesn't make as much sense, and the dungeon "replenishes" itself over time between delves.

Sure. But there must (should?) be allowances for the "area". Yes, in a large enough area (like a mega dungeon), we can make those assumptions. But at any given time period the "area" we are concerned about and actively encountering is much smaller than that. I was specifically addressing the idea of the GM having wandering monsters come along and attack the party while they are resting in an otherwise cleared space, in an otherwise cleared "area" around that space, and the GM doing so to "hurry the players up" (implication being that the GM doesn't want them to take time to recover fullly before continuing to explore new areas).

Sure. Given enough time (like if we leave the dungeon and come back weeks later), it's reasonable to assume some new things have migrated into areas previously cleared out. But it's pretty darn absurd to assume that during a single 8 hour rest period X number of monsters unrelated to what we've already been fighting in this area just happen to stumble upon where we are resting. If that was the rate of "monster roam", then why didn't they run into and attack the prevous residents of the space we're in now? Presumably, the kobolds that were here before we killed them would also have been attacked by any ransom slimes, or gelatonous cubes, or cave trolls, or whatever, if that's what wanders around here. The "immediate area" should have reached some sort of balance point in terms of "things settled in spaces versus things wandering around/among them" long before the PCs ever arrived.

If the actual rate of wandering monster was that high, then they should have wiped out whatever was settled here long ago *or* whatever was settled here should have wiped out the wandering monsters and reduced them to a reasonable rate that they could maintain and hold their area from. And sure. Maybe we don't need to think this hard if we're just going for "toss fun encounters at the players". I tend to prefer to make the spaces seem at least somewhat realistic. When running PCs though any part of my game world, I strive to make that part make sense from a "how did it look and work before the PCs arrived?" pov. I don't toss random encounters at the PCs that, if they were tossed at whatever else is "static" and in the area, would have wiped them out. This rule of thumb is the same whether I'm generating encounters while they are traveling along a road between two towns, or they are exploring a dungeon. To me, the same "rule" needs to apply.

But yeah. If you and your players don't care about this, and are ok with a "the dungeon generates monsters and we fight them" style game, then that's perfectly fine. That's certainly how the old school dungeons were structured. The PCs explored until they could not maintain the rate of encounters any more, and then left with their loot (or over extended and died horrible deaths!). And yeah, I do put things in like this, but they are very specifically structured and presnted as some sort of magically enchanted area/thing that generates monsters for some reason (the tomb that eternally generates a variety of undead, the ancient war tower in the center of the area that creates construct opponents eternally for the particpants to fight, some plane the PCs travel to which a bored god created as a playground to test people in, etc) . But for most "normal" adventure locales in my game setting? I wont do this. What is there is what is there, and it wont ever exceed some kind of logical balance limit, and it will only recover from losses below that limit over a period of time far longer than "one or two rest periods". That's just how I prefer to run my games.

[kyoryu]

Sure. But there must (should?) be allowances for the "area". Yes, in a large enough area (like a mega dungeon), we can make those assumptions. But at any given time period the "area" we are concerned about and actively encountering is much smaller than that. I was specifically addressing the idea of the GM having wandering monsters come along and attack the party while they are resting in an otherwise cleared space, in an otherwise cleared "area" around that space, and the GM doing so to "hurry the players up" (implication being that the GM doesn't want them to take time to recover fullly before continuing to explore new areas).

In a megadungeon, no area is "cleared". That's, to a certain extent, the point. You don't clear a megadungeon room by room - you go in there, try to get what you can, and get out before you die. The presumption is that the monster population is high enough that it is, effectively, unknowable.

And it's not about recovery - it's about doing eighteen search for traps checks or look for secret doors every step. To force some balance of danger and caution.

And, to reiterate, for most modern scenarios I agree with you! Most people aren't playing '70s and '80s style megadungeons, and the why and how of random encounters needs to adjust based on the play and scenario assumptions.

(Interestingly in the reboot of XCom, they ended up using a different mechanic to also counter overly-defensive play - people were playing the game very cautiously originally (move + overwatch), so in the expansion they added a resource that would dissipate in a few turns, to force you to not just sit back and hold a defensive stance.)

[KorvinStarmast]

(Interestingly in the reboot of XCom, they ended up using a different mechanic to also counter overly-defensive play - people were playing the game very cautiously originally (move + overwatch), so in the expansion they added a resource that would dissipate in a few turns, to force you to not just sit back and hold a defensive stance.)

I remember the discussions in Star Craft on how to deal with "turtling" as a strategy ... that goes back a ways ...

[sandmote]

My previous comment was way too long and really focusing on the tiny details on account of my trying to put together some encounter tables for a plane in general, so I think I'll back out a bit and try to collect my thoughts of the overall way random encounters work:

---

General
First, a lot of encounter tables are just badly put together. A gelatinous cube that the inhabitants let stay for extra protection and janitorial work might make sense, but you shouldn't be seeing goblins, orcs, and kobolds in the same dungeon. Maybe in different areas of the same megadungeon, but not all right next each other taking turns to attack the party one group at a time.

Similarly, not all forests are going to (or should) contain all the same threats, but we tend to grab one "forest encounter table" with everything you could maybe see encountering in that type of environment on it. So with a couple of bad rolls you could end up having to justify a bunch of factions that should have driven each other away. Further, in trying to force these tables bigger, they start to include things that's don't make much sense, like the 1d3 Dire Camels on a swamp encounter table.

Trying to figure out exactly is in every particular spot of the world is impossible, and can take a lot of time. One thing I can do to save time as DM during the session is to have basic answers prepared when something unexpected come up. And since no plan survives contact with the party, that's pretty beneficial.

---

Immediate Area
For the immediate area, I do think that that random encounters should be forcing the party to consider the environment, how the enemies act, and not to try topping off their rest based resources after every fight.

To that end, I think an encounter table for an individual dungeon or part of a megadungeon should be small, and certain types of targets should be struck off of it once beaten. The types of units should be relegated to ones the inhabitants would field, known auxiliaries, and occasional creatures that make sense to that group.

Gnolls warbands are pretty heterogenous, so for a level 4 5e party fighting them, maybe something following this general concept:
d12 Encounter
1-4: 1d6+1 gnolls
5-6: 4d4 gnoll witherlings
7-8: 1d4 maw demons
9-10: 1d4 flesh gnawers
11: 1 leucrottta with 1 gnoll (4 pairs total)
12: 1 troll (2 total)
For more organized groups or ones with fewer special creatures, this should probably include some more variety in units: guards making the rounds, sure, but also archers or skirmishers wandering around, people looking for a private place to duel, and individuals who run off to try to raise the alarm.

The party therefore is necessarily facing slightly different problems every single time you need to keep them moving or they attract attention. The fact you have multiple options for what shows up helps keep things fresh. They might be able to hold person one flesh gnawer long enough to kill the other next time, to they might have to focus on clearing out a bunch of witherlings, and that's going to be a better session than if every group of patrols is 6 guards with the same stat block. Not necessarily a massive difference, but better.

---

Region or Megadungeon
For a larger region, random encounters against help flesh out the area, but there should still be a clear set of creature types you end up finding for a particular forest, that isn't necessarily the same as those of every other forest. One forest could be infested with gnolls, another with orcs, and a third with yuan-ti, and should have something to reflect that. The other bit is that in a larger area you're going to want a larger variety of encounters, and listing them out helps make sure the list is big enough.

For a forest region, I'd want several tables: mundane animals that fill out the world and/or can be hunted, native beasts strong enough to be a threat, friendly encounters in case the party needs help (merchants, druids, pixies, couatls, ect.) and whatever the main local threat is. I want this prepared when I arrive to the session, so I'm not spending 5 minutes flipping through pages every time and having to settle on the first thing I find that works. And then if I have an idea of what sort of encounter I want, I can roll to see which table I'm looking at, instead of having everything stuck on the same d100 table with all different types of encounters mixed together. It doesn't help to roll to see what the party attracted with their loud noises and land on "1d6 web cocoons hanging from the branches, holding withered carcasses." But if I want to mark a particular part of the forest the party will be able to recognize in the future, saying its next to the trees with those cocoons is actually helpful, and the party running into a phase spider isn't. A table of options here also helps if the party ends up interested in multiple spots in the forest, because I've got them prepared.

In this case the encounter table helps you tailor local encounters to the local area, where not only are the encounters different in a forest than in a desert, but also the encounters in this forest are different from the other forest.

---

Large Scale Tables
Individual d100 tables like the ones they like to put in first party materials are probably only useful for pulling ideas from. Functionally just a list where if you're not sure you can let a die decide, more useful for ideas to add to a table for a particular area. A big list of everything that can show up can make it easier to decide what does show up.

This is the sort of situation where you're going to expect patrols of the kingdom throughout the kingdom, but the plains on the northern border have different problems than at the empire's core, and the separated encounter tables can help represent both. Its harder to forget to have various elementals show up on an elemental plane if they're 60% of random encounters compared to trying to remember which elementals you have stat blocks for off the top of your head. After all, I've got them written down on a little list next to me, and if I'm not sure which one to use I can toss a die before the party gets too bored and off track.

Here, the encounter table takes over the effort of making sure I know what's around the party and that the party is regularly reminded what this region is like without having to spend a lot of time at the table to make this happen.

[RedWarlock]

Can I just ask, what the heck is it with some of you and being so hyper-specific about what constitutes a "Random Encounter" as to create a strawman about it MUST be entirely dice-generated (no connection to the plot even as a post-facto justification), not regional-specific or CR-selective, randomly-rolled-out-of-the-entire-MM, must be utterly boring, badly run, and resource-draining, or else it's only "sparkling" railroading?

[Pex]

Can I just ask, what the heck is it with some of you and being so hyper-specific about what constitutes a "Random Encounter" as to create a strawman about it MUST be entirely dice-generated (no connection to the plot even as a post-facto justification), not regional-specific or CR-selective, randomly-rolled-out-of-the-entire-MM, must be utterly boring, badly run, and resource-draining, or else it's only "sparkling" railroading?

It's important to know the perspective of the individual of why they like or not like random encounters. In my case it's been more the correlation of random encounters being involved in wasting time listening to the DM monologue about nothing happening when the encounter doesn't even happen as the cause of my displeasure of it. When we understand and agree what the discussion is about we gain insight to others' point of view as well as clarification of our own. We come to understand the different degrees of randomness people enjoy from true random of everything determined by dice to randomly chosen based on a specific list of possibilities to randomness of one defined event but random on the time of adventuring day it will occur. The strawman that exists is only the one you brought in yourself.

[thirdkingdom]

My background is with BX and similar-style games, and less with AD&D, although there are similarities. I think that, as originally introduced, the idea of dungeon delving was to get in and out as quickly as possible with as much loot as possible. This is a system where gold is far more valuable as an XP metacurrency than killing creatures.

Random encounters are meant to speed things along and encourage the PCs to strike a balance between full investigation of each chamber and rushing through (plus, if you're doing a good job mapping, you should be able to tell where a number of secret doors/chambers are based on the map); random encounters do not typically result in meaningful treasure, so you're risking resource drain in exchange for, theoretically, very little.

Plus, the morale and reaction rolls in BX ensure that not all encounters end in combat; in fact, the incentive is to minimize combat. Surprise rolls give the PCs the chance to potentially avoid combat altogether, and most monsters are automatically hostile only on a roll of 5 or less on 2d6 (modified by Charisma if the characters attempt to parley). Morale checks are made, and many monsters will attempt to flee or surrender instead of fighting to the death.

A great Actual Play podcast that does a good job highlighting the old school style of play, and the use of random encounters, is 3d6 Down the Line, especially their current delve into the Arden Vul megadungeon.

[KorvinStarmast]

My background is with BX and similar-style games, and less with AD&D, although there are similarities. I think that, as originally introduced, the idea of dungeon delving was to get in and out as quickly as possible with as much loot as possible. This is a system where gold is far more valuable as an XP metacurrency than killing creatures.

Random encounters are meant to speed things along and encourage the PCs to strike a balance between full investigation of each chamber and rushing through (plus, if you're doing a good job mapping, you should be able to tell where a number of secret doors/chambers are based on the map); random encounters do not typically result in meaningful treasure, so you're risking resource drain in exchange for, theoretically, very little.

Plus, the morale and reaction rolls in BX ensure that not all encounters end in combat; in fact, the incentive is to minimize combat. Surprise rolls give the PCs the chance to potentially avoid combat altogether, and most monsters are automatically hostile only on a roll of 5 or less on 2d6 (modified by Charisma if the characters attempt to parley). Morale checks are made, and many monsters will attempt to flee or surrender instead of fighting to the death.

Yes, that style of play incentivized more than combat.
The other key feature of that was "Treasure in Lair" - if you weren't in the lair, there wasn't any treasure, beyond the usual claw, tooth, or maybe body part one might be interested in for making potions or trading to an NPC wizard for something ...

[gbaji]

It's important to know the perspective of the individual of why they like or not like random encounters. In my case it's been more the correlation of random encounters being involved in wasting time listening to the DM monologue about nothing happening when the encounter doesn't even happen as the cause of my displeasure of it. When we understand and agree what the discussion is about we gain insight to others' point of view as well as clarification of our own. We come to understand the different degrees of randomness people enjoy from true random of everything determined by dice to randomly chosen based on a specific list of possibilities to randomness of one defined event but random on the time of adventuring day it will occur. The strawman that exists is only the one you brought in yourself.

Pretty much this. If we don't ask for details about what someone's talking about, then the statement "Random encounters are bad!" is meaningless. Similarly, a counter argument of "No. Random encounters are good!" is similarly pointless. The term can mean a lot of things, and be used in a lot of different ways, many of which are "good" when used well/properly, and many of which may be "bad" if not.

I personally, find that the realization as a player that I may randomly run into the same begger, street urchin, or prostitute, in every single town we travel to in an adventure setting to be "bad" use of random encounters. Same deal if it becomes obvious that the same "traveling in plains" or "traveling in a forest", or "traveling in moutains" tables are being employed by the GM runnning the game (and yes, it really doesn't take that long to figure this out). But that's just me. So yeah, asking questions like "how customized are the tables to the specific area/place I'm adventuring in" is super relevant.

And yeah, as you pointed out, asking questions like "do you just sit at the table monologuing while rolling to see if we have an encounter or not", also becomes relevant. I would vastly prefer that the GM predetermine that we're going to run into gnolls while traveling through the hills, and then engage their own brains and decide "this would be a good spot to have the party encounter those gnolls I randomly determined they'd run into", then have the GM rolling on encounter charts while we're at the table, theoretically traveling through the hills. Again though, that's my preference.

Other players may absolutely love the idea that the GM is never actually applying any of their own thought process into anything, and truely just determining everything based on random tables and die rolls. IMPO though, if you distrust your GM that much, maybe you should not be playing in their game in the first place. But again, that's just me. And yes, some posters have expressed enjoymnent in games where everything is random, so it may not even be about trust at all. I personally find such games extremely tedious and boring, but that is, again, my own opinion. I suppose there's still a social element there, so that's a bit better than just playing a CRPG, or just rolling up encounters on my own and having my characters fight things for exp and loot (which yeah, seems silly).

Different players get different things out of playing RPGs. There's no "wrong" way of playing. And how random stuff is managed is absolutely going to tie into that.

[JusticeZero]

Random encounters fill a clock. It keeps people from going "Okay, we wait eight days" without thinking about it in terms of "Ugh, we have to stand around for a week." Is it a good system for doing that? No, it's awful. But there's a purpose to it, so I respect it for being a primitive solution to an existing problem.

[rel]

Random encounters fill a clock. It keeps people from going "Okay, we wait eight days" without thinking about it in terms of "Ugh, we have to stand around for a week." Is it a good system for doing that? No, it's awful. But there's a purpose to it, so I respect it for being a primitive solution to an existing problem.

If that's the design goal I don't think it's going to deliver the desired results without some very careful balancing.
Unless the random encounters are so dangerous there's a legitimate chance of the party not making it home or not making it back, it's not that much harder for the Players to say 'we retreat to a safe place where there are no random encounters, wait 8 days, then go back to where we were before.'

[JusticeZero]

If that's the design goal I don't think it's going to deliver the desired results without some very careful balancing.'

I mean... it was the 80's or whatever. Blue box stuff. You can't expect the ideas from then to actually be great ideas, they were more... valiant rough drafts.

[Mutazoia]

Random encounters have been a part of the hobby for fifty years, and I am still not sure if I see the point of them. I sometimes include them, and they seem to always piss the players off.

Furthermore, many random encounter tables include a few things that are wildly out of line with the rest in terms of difficulty. For example, most D&D encounter tables have an adult dragon as the maximum result, despite the fact that the vast majority of PCs will have no chance against one in a fight.

So what actual purpose do random encounters, particularly powerful ones, serve on either a mechanical or narrative level?



My players thought of a few reasons, none of which I really buy and all seem overly cynical:

1: They are there to add a horror element of the game to placate the small minority of players who enjoy horror at the expense of normal players who are there for fantasies or power and control.
2: The original designers are trolls, and later designers are just aping them.
3: They are there as a crutch for new GMs who don't actually know how to build or balance an encounter on their own.
4: They are included as a warning system for players; any GM who uses random encounter tables by the book is clearly incompetent and you should leave their game.

Monsters don't sit idly by in their assigned room in a dungeon, waiting for some random group of adventurers to wander in. They move. The less intelligent ones hunt for food, and the more intelligent ones investigate noises and anticipate threats. Think about the last time you were walking outside and a squirrel ran by. Technically, that squirrel was a random encounter for you. Good thing they can't cast fireball....

Whether you are clanking around in a dungeon, or spending days trekking across the wilderness, you are eventually going to encounter another living thing that you were not expecting to encounter. Random encounter tables are a method to aid a DM in replicating this in their games, giving the less experienced DMs a tool to generate a random encounter on the fly. Tables like this are not perfect, they are merely a starting point that can (and should) be modified by the individual DM as they see fit and gain more experience in their system of choice.