Thursday, June 6, 2013

The Great Odds Stacked Against Encounters

The more I think about monsters, and monster encounter tables, the more I want to be specific about how, or when, or where an encounter can possibly occur.  Who hasn't created a sea-going encounter table that included some monster who could not possibly interact with a party aboard a large merchant ship?  Or found that their subterranean encounter table is largely useless because the party is entering an ice cave?  In fact, the possibilities for different encounter 'areas' screwing with encounter tables fairly pushes a DM into establishing from the beginning what the monster ad hoc.  For myself, I've largely given up rolling for what the monster is, instead opting for a 2d6 scale for how aggressive whatever the monster might be that I suppose throwing.  Alternatively, I make sure the party sees the monster, or hears about the monster, in advance, so they can freely decide whether to approach.

I hate this.  I would rather have some sort of system that did hammer down what monsters could occur, and where, in a logical, orderly fashion, and thus remove myself a bit more from the random loop.  See, I believe that if I'm faced with the challenge - the die says this monster appears, deal with it as a DM - it makes it a less homogenous world.  I believe that when the DM chooses all the monsters, he or she will be inclined to choose certain monsters again and again, because those are familiar and the first to come to mind.  I don't feel that improves gameplay.

Nor do I think it rational to make sure that every monster gets a try, like a bored diner going through every dish from the menu on principle.  The distribution should not be based on the need for variety ... there should be some continuity, and that isn't helped by having every weird and bizarre monster show up for a campaign like they're all standing in line somewhere, off-plane, each holding a taken ticket.  We are not serving number 63.

So yes, the problem of nailing down the circumstances for an encounter ... more so than the fact that it is a temperate climate (mean yearly, no account taken for winter vs. summer) and that this is a forest.  That's pretty scant detail for a biological determination.

The basic aggression of the monster would seem to be relevant.  Is the monster the sort that sits and waits for prey, or one that deliberately hunts prey out?  What sort are you more likely to encounter?  It seems to me that something like rats are going to be found a lot more than something like a golem, that requires an elaborate creation process as compared to a brood of 20 offspring.  But are you really going to have rats that walk into the road, confront the party and start fighting?  Or is it more likely they'll pilfer food, that they'll flee the moment they're discovered and that they'll be an eternal pest without much likelihood of actual combat except when the rats have reason to be threatened?

If you're vigilant, you use a trusted water source, you use your exterminate cantrip to kill anything inside a bottle before taking a drink, are you going to get attacked by a throat leech from the Fiend Folio?  Or any other similar creature?  Even if there are such leeches in the brook where you fill your canteen, what really are the chances of collecting one?  Is the brook so thick with them?  Or are we talking about an extreme low chance, so low that the monster might occur one time in a two year span of regular playing?

A lot of the monsters are like this.  You have to really go out of your way to find most of those that live only in subterranean places; if you're not wandering the swamps of tropical regions, you're unlikely to be attacked by a crocodile or alligator.  And even if you're fifty yards from a pond infested with them, they're not likely to leave the water and come looking for you.  Unless you go down to the water and make yourself available, you're mostly safe.  Not guaranteed safe, but the % die roll has to be affected by the fact that you're camped in these rocks 75 yards from the river and not down on the beach.  Of course, there are no tables of any kind for such distinctions.

So, it has to be conceded that there is a difference in not just the aggression of the monster, but in the party's exact proximity to the monster's hunting pattern.  If the monster doesn't hunt at all, there may never be an opportunity for an encounter.  A stag may do nice damage with that rack, but that's not going to matter if the stag flees everytime its encountered (which they are want to do).  If you're hunting a deer, you're not doing it hand to hand - so does anything about it matter except its hit points?

There's a rare, rare chance that you'll encounter the deer when its in rut, whereupon it has a chance of going stupid nuts trying to kill you ... but that's only a danger if you happen to be fairly close by when you see it, or you try to get closer for reasons of your own.  If you come around a corner and surprise a large male deer in rut, watch out!  But how often is that going to happen?

I'd argue half the monsters in the books are pretty much that way - one offs that might be an encounter once in a blue moon, with the right circumstances and chance.  But wasted space, really, where it comes to a regular campaign.

Consider the difference between a huge spider, likened to a wolf spider, and a large spider, which spins webs.  Which are you more likely to encounter in a dungeon?  Which - if you're not an idiot and willing to go stumbling into a web you chance upon - is more dangerous?  A smart party of sufficient power may never, ever be attacked by a large spider (see web, burn web).  But a huge spider will still be a problem no matter what - because it doesn't broadcast its presence and it attacks by surprise.  It's out there to kill everything that moves ... not just very stupid things that don't know what a web is.

If you admit that not going into a cave filled with web is a good way not to get killed by a large spider, then obviously keeping to the road and not stumbling off into the wilderness has to be safer and more encounter free ... at least, of beasties.  You might increase the chances of bandits by staying on the road, but then that matters which road, right?  And how close to town you are, or how many people are on the road.  Too many people makes banditry difficult.  Too few people and it makes it meaningless.  The bandits aren't going to wait thirty days on this remote road in the hopes that something with money comes stumbling by.  They'll probably have someone in the town, and how well you advertise yourself will determine what sort of target your party is.

In any case, there has to be some sort of measure for where you're willing to go - off road, off trail, into caves or not ... and for how long you stay in a place.  If you're passing through a 2 mile hex in an hour, probably you won't meet the baddie that's been put there.  If you camp overnight, that increases the likelihood.  If you deliberately explore the hex, that's one better ... and if you hang out there night after night, sans weapons, running around naked and singing the same song at the top of your lungs, I guarantee the baddie is going to find you.  It's really a question of how badly you want to be found.

Is it even possible to come up with a monster generation system that addresses these things?  Along with the weather, how warm it is, what the weather is and so on?  Where do you draw the line?  Should the height, weight, number, race, amount of armor and weapons, magical signature and so on of the party have an effect on whether a monster will produce itself from the bushes?  Probably.  But how realistic is that?  How far can you go until the chances of actually encountering a monster willing to encounter the party drops so near to zero that the party can feel comfortable marching hither and yon unmolested?

I'm not sure.  There's a serious rethink needed, no question ... and it has to be a lot more complicated, I think, than the shit tables everyone tosses up on their blogs, where forty percent of all encounters in a desert is a lizard of some kind and therefore extraordinarily the same encounter over and over again.

Granted, for many gamesters, I'm asking the impossible.  I'm asking the undesirable.  Well, 'undesirable' until someone does all the damn work that makes it insanely rational and convenient for them.  The willingness of a recalcitrant DM to use a new, friendly mechanic is something akin to the amount of blood a possible victim is spewing while laying alone on a trail within sight of an orc.  The first question, is it a trap?  And later, while eating hearty with your friends, its all about how good a story you weave about your remarkable adventure in overcoming the great odds stacked against you.


Aaron Aldridge said...

Great post. I was considering the same thing recently. I may give it more thought later.

Homer2101 said...

One possibility for determining what the party encounters may be to use a dynamic table, rather than a static table. That's probably impossible to do using pen and paper, without making a book's worth of tables, but this is 2013 and we have things like Excel now.

If the party gets a random encounter, the DM would follow the usual decision tree of looking at the region, the climate, and the terrain type. That narrows down the possible menu of things to encounter fairly easily.

Once the DM picks a table, he could input the climate and terrain, the development index, the time of day, the weather, and the season, among other things. The table would then scale the encounter probabilities accordingly. For example, the probability of encountering a brown bear in winter should be near 0%, because brown bears hibernate.

Maybe the various things the party might encounter, like monsters, could have stats like weather and season sensitivity stored in a table or database. The encounter generator could pull those values based on DM input and produce a table on the fly.

That doesn't determine whether the party potentially encounters something in the first place, unfortunately. For that, a dynamic 'encounter index' to roll against might work. Maybe something long the following lines:

For every four miles the party travels, the DM rolls a d100 against the local encounter index. Rolling below the index nets the party a potential encounter. Say the index baseline is (for the sake of argument) 50, modified by the hex's type, development index, the weather, the type of road (if any), and what have you. Travelling across a well-developed hex would substantially reduce the encounter index; travelling across a wild hex might increase it, although realistically the odds of encountering anything substantial in a non-magical wilderness are usually not going to be very high anyway.

Alexis Smolensk said...

In fact Homer, what you describe is precisely what I have been working on.

Of course, it does require establishing that data base ... that is, making up a determination for all those things you describe for every monster.

Takes time.

Dave said...

A quick solution might be to rate your monsters as "Aggressive", "Neutral" or "Timid," then use 2d6 to determine if the beast is encountered, as follows:
Agressive, 5+
Neutral, 7+
Timid 11+

Add 1 for each hour spent in the hex after the first one.

Mic B said...

IF someone would be using a complex computer model, I think an exponential distribution would make sense. Each possible encounter would have a probability of occuring, which would be modified by any occuring encounter. This way, multiple type of encounters could occur at the same time (or preferably, with a slight delay).

for exemple, an encounter with a solitary predator would scare off most animals, but attract scavengers.

or if the party stumble upon an orcish army, the odds for different type of orcish "leaders" would drastically go up