Community

Character optimization is one of the most active sections of the D&D forums and I always wondered why there isn't a forum for encounter optimization.  If players can discuss crazy combos and maximized abilities why isn't there a section for dms to discuss effective, high-end encounter design and strategy?

The DM's job isn't to kill players (that's easy)

The DM's job is to make a fun world.  And there IS a forum for that. 

Before this turns into a 'optimization is bad'/'no it isn't' thread I'll throw in my 2 cents...

I think an optimized encounter gives the players a number of things to juggle while being defeatable. An example might be, fighting a bunch of guys mounted on birds, while the PCs are on a cumbling bridge across a large chasm that has a few harpoon turrets(with a pull feature) on the bridge supports, and the air floating up from the chasm is noxious...

This is the kind of stuff that makes combat more interesting for players, at least the players I know. And if we're looking to optimize something like that, by all means.

If the goal however, is kill the players, that's easily done as mentioned above.

I would argue that given the acknowledgement of the responsibility of the DM to be architect of the story would lead to the conclusion that "optimal" encounter design is a little more complicated than "rocks exploding minions fall, everybody dies."  "Optimal" for a DM means something different than it does for a player.  Players are encouraged to let their best shine through.  DMs are encouraged to let them.

Encounter optimization absolutely has a place, and can be a strong influence on the enjoyment of the group as a whole.  It's easy enough to say, hm, three elites and two standards, what exp budget is that?  It's a lot more to actually optimize the encounter.  I'd posit some starting questions:

1)  What is the intended level of danger for the encounter?  Should the players feel like they're about to die if they don't handle things properly?  Easy example is a BBEG battle at the end of a tier, say.  Your players should right well expect something challenging, and for the BBEG to kill them and continue on his BBEG ways should they fail.  This links in with dungeon and campaign design, as you may want several less-than-threatening encounters whittling away resources before a large fight, or one burn-all-your-dailies slugfest, or even a grueling gauntlet with nothing but anklebiters.  But the important point is that this is the first step.

2)  How likely is the difficulty of the encounter going to match the intended level of danger?  This is a trickier thing.  4e did it much better with having more streamlined stats and expected power values for monsters, so when you get a L+2 exp budget, you can be reasonably assured that it'll behave the way all the other L+2 exp budgets have.  But there are subtleties and tricks to using monsters properly, just as there are to using player characters properly.  How merciful or merciless you are within your experience budget can be adjusted on the fly as you play, so keep that in mind when using monsters with either potentially swingy or more or less fixed abilities.  As one example I played recently, two parties of about even competence resulted in completely different experiences.  One party absolutely crushed the encounter (as in the named boss didn't even get to make an attack), and the other party was TPK'd based on how ruthlessly the DM focus-fired the ghoul status effects involved (one hit immobilized (save ends), hitting an immobilized target stuns (save ends)).  Playing the monsters "smartest" isn't always the optimal choice, not it if gets them TPK'd on something that wasn't supposed to be TPK-level dangerous.

3)  How satisfying is the actual dispatching of the monsters relative to their threat?  People talk about risk-reward relationships, and things like "have good loot behind hard things" and such, but there's a satisfaction involved in solidly shutting down and dispatching a monster that does bad things to you.  Give your monsters weaknesses, and let your players know about them in knowledge checks or through subtle hints beyond that.  Vulnerability is an easy, boring example, but it works in a pinch.  Other examples might be skirmishers that have very specific lines of movement.  Let the players block their movement, and reduce the threat.  You can choose a monster out of the Monster Manual and drop it into a room, but it doesn't turn into a satisfying experience until you design the encounter around it.  What tricks does it have, when does it use them, how do the players find out about it and how do they shut it down?  These are the sort of questions that truly optimize encounters, turning from *yawn* we killed a bugbear to "Yeah, we totally smoked that bugbear!  But look what it did to the wizard first..."

EncOp isn't as simple a question as some might think, and it's distinct and separate from world design.  It may be covered in What's a DM to Do? but I don't frequent there much myself.