What level of monster optimization and monster tactical skill do people play at?

[zlefin]

I'm just wondering about the distribution of optimization levels for the monsters/foes (both the regular ones, and more boss-style ones) and of how smart the enemies fight. I know there's quite a bit of variety of playstyles, I just have no sense of the average.
Do the enemies tend to play at the level expected for their int/wis/training? Do they tend to all play fairly smart? All pretty dumb? To what extent will beings retreat when it would be appropriate? How cautious are they about attacking?

[Eldariel]

Builds about on par with the general power level of the game world (mostly influenced by what's agreed on beforehand; monster spells, feats, etc. get similarly improved if necessary). Strategies, tactics, etc. determined by their organisationand its mental aptitude and martial inclination. Obviously a breeding pair of dragons acts very different from a non-involved one and obviously an Orc clan lead by an illithid behind the curtains is quite more canny than one lead by a standard Orc chieftain. Well, that's how I run my games anyways and that's how I'd prefer for my DMs to run theirs. Obviously goals (every creature in the world wants something, after all) always influence things as well.

[PunBlake]

I usually run enemies based on their mental stat blocks. Are they smart enough to do <optimal thing>? Depends.
Enemy selection is based on party composition; you choose based on what you think the party can handle while being challenging, and to a lesser degree what you as GM want to do and what plot makes available. You get better at gauging party strength with more GMing experience.

Often, once a fight has started, enemies don't have time or ability to retreat unless they can Teleport away or move in a way a party is reluctant or unable to emulate or follow (burrowing, swimming, flying at great speed, a Gate back to a home plane, etc.)... but if able and intelligence/planning dictates, retreats do happen. In my experience, with mid-op parties or lower, this is rare; with higher op-fu, enemies become more capable of shenanigans.

What determines how cautious enemies are, and basically answers most components of these questions, is usually three-fold. In no particular order:
1. Alignment/motivation. (Ex. Following orders.)
2. Mental stats. (Is planning possible? How thorough? Leaders in groups matter a lot here.)
3. Plot. (Are you, as GM, wanting to make these enemies cautious? This may require additional effort, but is generally worth it.)

[arkangel111]

Generally speaking I play to their mental stats.

This of course gets adjusted based on playstyles, plot, tropes (goblins are fun when played crazy, even if smart), and how badly I need the PC's to suffer in a battle. If I am building suspense I might hit lightly several times without giving a chance to rest. I'll also sometimes adjust strategies based on what I am trying to get my group to learn. For instance, If they are facing a mind-affecting BBEG that will cream them soon I start bringing in some baddies with similar abilities but gradually increasing difficulty to get them thinking about that aspect. Last thing I want is a TPK because they felt cheated and never considered a tactic. Granted my group is mostly new players.

[Thurbane]

As a DM, I try to play creatures by their Int, but have sometimes been accused by my players if running all monsters as tacticians.

[Particle_Man]

I simplify. Dumb monsters are dumb. Smart monsters get one sneaky trick and after that are dumb. Maybe that is because I am dumb. :smallsmile:

[Lapak]

I try to play monsters as intelligently as their stats allow... but that's a low bar because for anything above animal intelligence basic tactics like 'choose good ground to fight on' and 'use surprise and positioning to improve your chances' are really, really basic to survival and anyone with an INT of 4+ can be expected to hit on them.

Balancing that somewhat, one thing that surprised me above is the person who said their monsters don't retreat unless they are likely to succeed. From my view, people/monsters don't flee combat because they think it's likely they'll escape - they flee combat because they think they are going to die if they don't, so any chance is better than none. Depending on their nature/intelligence/alignment/etc., they may flee in a coordinated or random fashion, may try to do a group escape or every-monster-for-itself scatter to improve individual chances, or whatever, but once the outcome of a fight is clear most of my monsters are going to flee if they are physically able and attempt to surrender if not (unless/until the party has established that this is also certain death via reputation.)

[PunBlake]

Balancing that somewhat, one thing that surprised me above is the person who said their monsters don't retreat unless they are likely to succeed.

It's not that my monsters don't retreat; it's that they often don't have a chance with a combination of good crowd control, positioning/tactics, and timing in combat (ie, mid-op play).
I guess I'm more likely to use fewer, challenging monsters than swarms of things that use guerrilla tactics because my players enjoy slug-fests? Hmm. This is why forums are good; my perspective is challenged. Maybe I should make a "Tucker's Kobolds"-alike situation and see how they handle it at some point.

[Karl Aegis]

Monsters play to their strengths and try to win. Goblins will use ranged attacks and try to keep their distance because they have low strength and high mobility. Minotaurs will try to close to melee as soon as possible to bring their long reach and high strength into play. Giants will throw rocks until they think they can finish things with one blow unless they run into a dedicated ranged character. Mounted units will try to chip away at health until a lance charge becomes feasible. Things like that.

[Uncle Pine]

I generally play dumb monsters less intelligently than smart opponents, sometime even glaringly so. I remember at least two occasions (one against skeleton opponents, another against ogres) in which one of the most strategy-oriented players was surprised and commented about how a creature charged in or positioned itself in particularly suboptimal ways - to which I replied "it's a skeleton, it doesn't care!"

I usually optimize opponents and NPCs for fun or spectacular results rather than higher numbers. This is also the way my players (or I for them) build their characters, so both sides end up with about the same optimization level. On the other hand, I'll mention that there are times when upping the optimization level a bit can drastically decrease the amount of preparation necessary for a few encounters or even an entire portion of an adventure or just make for more dramatic climaxes, so I'd put myself at medium monster optimization with an occasional mid-high power spike.

[Necroticplague]

My world is very unforgiving for all involved. Food chains extend upwards a very long ways, and survival isn’t always easy. As a result, all combat is as cutthroat and optimal as the monster’s resources allow for. Those without a good sense of tactics have died off long ago. The only fools who may still remain are those locked away in heavily protected cities, shielded from the cruelties of the world. Even a mindless beast is subject to natural selection proliferating those with the sharpest instincts on the matter.

However, the harshness of the world stems from the rules of the world. Thus, retreat is a rarity, because it is difficult. AoOs, charges, grapples, trips and ranged weapons can combine to mean that, even without magic entering the picture, disengaging from combat can be as lethal as continuing it. So most things will avoid direct confrontation if possible, but if pressed, will fight on to the bitter end unless they have some exceptional ability to flee.

[King of Nowhere]

I'd like to say that I play enemies to their mental statblocks, but I actually only have two speeds there: enraged animal (attacks the most convenient target) and smart (I play it to my best skill). This only afffects tactical decisions, like "full attack the barbarian in melee, or take an attack of opportunity to go charge the wizard behind?". As for strategical options (ambushes and the like), I tend to play them as they make sense. It doesn't take a genius to set up an ambush or use fortification, given time, and even a genius can only do as much as his resources allow.
Regarding retreat, it depends case by case. Monsters will generally fight to the death if they didn't retreat early, but intelligent opponents will escape or surrender when possible. high level npcs will try to save equipment over their lives, as they are quite confident their allies will resurrect them, but all the gear that will be looted will be much harder to replace.

[Goaty14]

I play my monsters dumb. If I played them smart then my players would complain when "kill a couple of goblins" becomes "kill all of the goblins in the countryside before they all rally up and erase you from the face of the material plane" :smallfrown:. Even the lowly kobolds become significant threats 1-2 CR higher than their stat blocks would suggest if played as such.

[Bucky]

I try to run encounters with a degree of tactical sophistication. The monster's tactical skill varies with int, but I try to make the encounter tactically interesting even in mindless encounters. Some examples...

6 mindless zombies attack the party. They have one tactic. Each one picks a party member and fights to the death, ignoring the other party members (except to take AoOs) unless its target drops. Any coordination is coincidental. No zombie ever makes a choice for tactical reasons - if it needs to, it chooses randomly. If the tactic becomes inappropriate, such as if their target kites them, they continue it anyway. The tactical depth of the encounter comes from deciphering and abusing the zombies' simple intentions.

6 wolves attack the party. They have a strategy - "surround and destroy" - and a couple of tactics - "flank" and "focus fire". They switch between tactics based on opportunity. Given a choice of target, a wolf attacks the one it thinks is most vulnerable (e.g. wounded or in an exposed position). If their strategy becomes inappropriate, such as if they no longer have the numbers to surround anyone, the remaining wolves retreat. The tactical depth of the encounter comes from the wolves' tactics.

6 level-1 Elven Fighters attack the party. They have several available strategies, but choose to fight from a formation. 4 of them set up across the map from the party and use bows; the other two take a position somewhat closer and wait with readied actions and reach weapons. Each of them fights with a degree of tactical expertise - for example, the vanguard fights defensively to buy the archers more time. If their strategy becomes inappropriate, they adjust it - e.g. if the vanguard is overwhelmed in melee, one of the archers charges in with a sword, and if the party doesn't approach and demonstrates superior archery the vanguard charges the archers.

[Kelb_Panthera]

Generally speaking; most monsters are not adjusted from their basic stat blocks for optimization and I try to run them in line with their mental stats and the descriptions in their MM entries.

Boss monsters and classed NPCs are another matter.

If a monster is playing a special role in the adventure/ campaign then I'll rebuild it to be substantially more optimized than normal most of the time. True dragons get absolutely cut-throat.

NPCs, since they must be built, tend towards a lowish-medium level of optimization, just a little lower than our group tends to optimize our characters. I like to experiment with weird ideas from time to time and it doesn't always work out. Most NPCs get so little screen time before being permanently removed from play that it doesn't really hurt when an idea that seemed good gets them all killed horribly. Tactically it varies pretty widely. Dumb, low level characters usually only get one good tactic and fall apart quickly if it doesn't work. Smarter and higher level characters get several and characters that are both smart and higher level get pretty vicious. Bosses tend toward the higher end of both optimization and tactics, regardless of stats and level though never substantially higher than the group's acceptable levels.

[Uncle Pine]

Speaking of monster tactics, I remember reading a post a while ago about having your players state what actions they are going to take at the beginning of the round before a hyperintelligent opponent (i.e. an illithid BBEG, or an old dragon) decided how to act on his turn. The concept fascinated me and I plan to try it out in-game eventually, but haven't had the chance to do so yet.

[Darth Ultron]

I play everyone and everything in the world to their mental stats, though that does not say much. The average ''10'' mental person can think of ''average" things like locking doors or setting up a guard.

The typical RPG world is a LOT more ''life and death'' then the world of a typical player or DM. So really even the ''low mental stat folks" in a ''life and death" world be very tactical just to survive. Even the dumbest orc in the world knows enough to keep guard and watch their back.

So compared to most players, that take a very laid back casual way of things, I run NPCs and monsters very tactically. And even beyond: I typical do things like have a guard watching the guards or doors with two locks.

[Mlmiii]

My monsters, especially animals, usually rely more on type than stats for their strategies. For example, a giant mantis will, unless hit by a ranged attack, spam Stealth out of combat, pick a good spot to fight on, pick an aggro range that it won't let the party enter without a fight, and not move from that spot until it dies or the party leaves. This is because that's how irl mantises work--similarly, wolves have basic flanking and herding tactics, but don't know enough about adventurers to plan out who to focus beyond who's the scariest at that moment. Most of my animals do run away if they're at low health, though a pack hunter may stay behind to assure its allies' victory if it thinks its death will win the fight.

Sometimes to spice things up I have animals working together, most notably one time I had a flock of hawks acting as seeing-eye birds for a giant scorpion, since arachnids have garbage eyesight and excellent tremorsense irl. The birds would designate a target and provide flanking bonuses while the scorpion did the usual arachnid combat routine of "jump from ambush, claw or grapple into submission, poison if needed", leading the birds in retreat once it got below 25% or collected a dead/unconscious meal of 200+ pounds--it's here for food, after all, and if the players bought a sacrificial pig to bring on the trail this particular scorpion travels then they still get XP for "winning" the encounter if they toss the pig to the scorpion instead of fighting it.

[Lapak]

It's not that my monsters don't retreat; it's that they often don't have a chance with a combination of good crowd control, positioning/tactics, and timing in combat (ie, mid-op play).
I guess I'm more likely to use fewer, challenging monsters than swarms of things that use guerrilla tactics because my players enjoy slug-fests? Hmm. This is why forums are good; my perspective is challenged. Maybe I should make a "Tucker's Kobolds"-alike situation and see how they handle it at some point.

I like both slugfests and crowds in their proper place; slugfests are good for boss fights, but I find I get more of the expected mileage for regular encounters by using crowds of weaker monsters as often as is reasonable. Still legitimate threats for the level, mind you, but monsters where an individual poses little threat to a given PC.

The more dice you're rolling, the closer the end result hews to the average. A bunch of little monsters are unlikely to roll a massive crit that takes out a player, but also unlikely to have bad luck and miss every attack.
Players tend to feel pressure to whittle down the numbers rapidly, especially if good tactics are in play, so they are more likely to burn some limited resources to do so. If they don't, the dice averages will mean that they burn hit point attrition instead.
Balancing that, plowing through a crowd of lesser threats tends to give the players a moment of feeling badass.
Also, it's way easier to adjust the actual threat level in play. You can throw an absolute mob at the party knowing that only some of them can effectively fight at once; if that proves to be too few so be it, but too big a threat is less likely - the crowd can have morale issues at any point after the danger that the PCs pose is clear so they aren't going to have to kill every single one of them to survive. Once both sides have been bruised a little the remainder can flee.