How does the community of DM's calculate threat assessment in battle?

[MBControl]

One area of DMing that I would like to improve is my in game threat assessment on the side of the Bad Guys.

I have no real rules, other than basic motivations, for example, that barbarian just hit me for a bum ton of damage, so I'm hitting him back. I also use story motivations to filter my strikes. An archenemy will often blindly focus on his/her nemesis.

But looking back I find that my battle plans have been biased one way or another.

Do you fine people have tips, tables, or rules that help you?

[gfishfunk]

I have just some general approaches, nothing systematic:

1. When I create a threat, I determine how smart that threat is, and a general idea of how it will go after people.

2. When I multi-attack, I generally split it up if I am able. EDIT: I do this mostly for balance, so my players don't get frustrated. Its totally Meta and might break reality.

3. If there is a leader type, I will generally give it a bonus action that lets it command others, providing slight movement advantage (all enemies move 5', Attacks of Opportunity still apply) or something similar; I use this type of thing to show that the enemy has a tactician. While this guy is alive, the enemies generally attack smart (going after certain targets, coordinating, etc.

4. Absent anything else, I talk it out while giving my enemies actions and attack as it seems obvious based on the last combat round.

[Kryx]

I typically have enemies try to play to their intelligence. Pack creatures like hobgoblins will focus squishies. Big monsters, while dumb, may avoid plate armor still as they aren't stupid. Dragons will dive in and out, trying to eliminate threats.

Sometimes I'll have them return fire on a tank if that makes sense.

Multiattack is almost always against the same target. It doesn't make much flavor sense to split it in most cases imo.

[Tanarii]

Pack creatures like hobgoblins will focus squishies. Big monsters, while dumb, may avoid plate armor still as they aren't stupid.This doesn't really equate to "intelligence" when armor != least squishy in D&D worlds. Being squishy or non-squishy doesn't necessarily have a visible tell.

At least, in recent editions.


Edit: Immediately after I hit post, I realized that may be true for PCs, but its not necessarily true for the world. ie PCs may run around with Light Armor + max Dex, or Medium + Dex 14, but when the rest of the humanoid world has Dex 10 as their average, better Armor does equal less squishy.

Which is something a naked barbarian tank can take huge advantage of psychologically to keep the enemies focused on him. :smallbiggrin:

[Kryx]

This doesn't really equate to "intelligence" when armor != least squishy in D&D worlds. Being squishy or non-squishy doesn't necessarily have a visible tell.
Ya, I definitely don't expect most monsters to understand that a barbarian or a monk (to a lesser extent) will be hard to hit. In general I may have them attack for a round and then focus on the guy in robes after that doesn't work. It depends. I do my best to RP the monsters focusing threats and not to metagame PC's ACs.

[pwykersotz]

Smart villains will use a certain amount of GM knowledge to assess danger, focus fire, and do other brutal tactics.

Less intelligent creatures and general mooks attack based on predefined and generic rules. Ranged creatures target party members in the rear or focus fire the one on point if he becomes a direct threat or is blocking . Melee creatures target the closest threat first. If there is ambiguity, I count the number of possible targets and roll a die to determine the target.

[Tanarii]

I do my best to RP the monsters focusing threats and not to metagame PC's ACs.
My initial thinking was it wouldn't be meta-gaming in a universe where un or lightly armored characters were regularly as tough to hit as armored characters. (Sans shield at least.)

But of course the mistake I was making was thinking PCs were normal for their universe. :smallconfused:

[FaradayCage]

If you were playing the encounter to the best of your ability as a DM with knowledge of combat tactics and keeping in line with the enemies' goal (which by default is to kill all the PCs), then you would probably focus fire the squishiest target and then work your way up the ladder. (This is a simplification of the best tactical choice, but you get the idea).

Of course, that's kind of boring and frustrating for players (and we forgive them for employing the same tactics because there's always more enemies), so most DMs probably don't do that.

This came up in a game where the PCs encountered an intelligent aberration controlling four really nasty undead. In the first round the undead wailed on the fighter (who was closest to them) and dropped him down to about 20% health. The PCs each took their rounds and didn't pose much of a threat, except for the psion in the rear that gave a sound trashing to each enemy with an AoE type attack. I wondered whether I had been fair to gang up on the fighter type, and sent the undead toward the psion in a sort of borg-like threat analysis.

The psion quipped that the undead "were acting awfully intelligent for undead." I replied "if they were acting awfully intelligent, the fighter would be dead by now" with a laugh. (Another player correctly hypothesized that they were being controlled by the aberration.)

This led to a post-game discussion about enemy "AI" in D&D. I had a bit of an epiphany in realizing that there is a slight different in "DM vs PC" and "enemies vs PC". If a DM ever wanted to "beat" the players, he would probably use the ganging-up tactic, the goal being to kill all the players and "win". :wink:. (Or try to kill at least one player and whittle them down to a TPK over the course of a few encounters). But for most enemies the goal is to win, but also to survive. Most enemies would not be in their graves thinking "well at least we killed one!"

So the epiphany I had was that being a killer DM and role-playing killer enemies are distinctly different.

The best way to "AI" enemies is probably just to RP it. Even if it's Bandit 1, Bandit 2, Bandit 3, they're all different people. Maybe Bandit 1 and 2 are close friends and like to stick together in a fight, while Bandit 3 is terrified of dwarves but gets a sick thrill out of killing mages. Conveying this information in some way during combat might help to spruce up what can easily devolve into a zero-RP situation.

[MaxWilson]

So the epiphany I had was that being a killer DM and role-playing killer enemies are distinctly different.

The best way to "AI" enemies is probably just to RP it. Even if it's Bandit 1, Bandit 2, Bandit 3, they're all different people. Maybe Bandit 1 and 2 are close friends and like to stick together in a fight, while Bandit 3 is terrified of dwarves but gets a sick thrill out of killing mages. Conveying this information in some way during combat might help to spruce up what can easily devolve into a zero-RP situation.

So much this. Players stumble upon a Giant Poisonous rattlesnake? Its default action should not be "attack--everyone roll initiative!" It should be "rattle, rattle, rattle, hiss! What do you guys do?"

Although, it's fun to break this rule on occasion too. One of the fun things about playing zombies and skeletons, for me as a DM, is that they really do just want to kill without caring if they are killed back. It's simpler to run and it scares the players.

[ClintACK]

... But for most enemies the goal is to win, but also to survive. Most enemies would not be in their graves thinking "well at least we killed one!"

So the epiphany I had was that being a killer DM and role-playing killer enemies are distinctly different.


This.

I'm always amazed by the number of NPCs who are willing to fight to the death, without trying to surrender or flee.

[Tanarii]

This.

I'm always amazed by the number of NPCs who are willing to fight to the death, without trying to surrender or flee.

DMG pg 273, Combat Options, Morale

An optional rule for the DM who doesn't like, or is apparently incapable of, deciding without a die roll on when it's reasonable for the enemies to break and retreat.

Of course, part of the problem is that in D&D effective retreat is really damn hard. Especially if you wait too long, which both DMs and Players tend to do. Even the DMG check points of half creatures dead (group) or half hp lost (creature) often means no chance of escape before death in the case of NPCs/Monsters.

[MBControl]

I was going to reply with some quotes of the most useful replies, but I decided not to, as everybody was extremely helpful. This is the best advice I've been given since being on this site.

Thanks team.

I will use a lot of all our thoughts, but there are a couple key take aways, that help me specifically. Play as the NPC, not the DM. It does clear up a couple of grey areas for me. On top of that, if the NPC is a character that is trying to survive, he is likely going to be quicker to run away, when over-matched.

That's just a couple points that helped, again, thanks.

[Scaleybob]

I also like to try to give my Monsters a bit of personality on the battle field - if you can make the PCs see them more than numbers then you're doing well.

A few tricks: Foes who run away when badly hurt, leaving others in the lurch. This can lead to the abandoned ones then being more keen to take on the cowards as much as the PCs- and can lead to some good interaction later.

Have foes bear a grudge in combat - it the ranger pincushions someone full of arrows on round one, then have that person try to chase down the ranger, despite whats going on the rest of the battlefield.

Have particularly evil or stupid enemies catch their allies in AOEs. PCs love to see this, and again can lead to people switching sides.

Play foes as intelligent as they are meant to be, and realize most foes generally don't want to die. It can be a good idea to have a recurring type of opponent who this doesn't apply to .In my last campaign this was gnolls and Foulspawn - neither understood the idea of surrender.

[Regitnui]

DMG pg 273, Combat Options, Morale

An optional rule for the DM who doesn't like, or is apparently incapable of, deciding without a die roll on when it's reasonable for the enemies to break and retreat.

Of course, part of the problem is that in D&D effective retreat is really damn hard. Especially if you wait too long, which both DMs and Players tend to do. Even the DMG check points of half creatures dead (group) or half hp lost (creature) often means no chance of escape before death in the case of NPCs/Monsters.

Most wild or unintelligent animals will retreat when they face serious resistance; most Beasts and unintelligent Monstrosities would fall under this. However, if the opponents are intelligent enough to start using advanced tactics, they're msart enough to fight on despite fear and morale checks anyway...

[Tanarii]

Most wild or unintelligent animals will retreat when they face serious resistance; most Beasts and unintelligent Monstrosities would fall under this. However, if the opponents are intelligent enough to start using advanced tactics, they're msart enough to fight on despite fear and morale checks anyway...Given that morale checks are an option rule, "despite morale checks" would be pointlessly making checks then ignoring them. Why not only use the optional rule for not-very-intelligent creatures in the first place?

Edit: despite the name, they don't have to represent broken morale and fear. They can easily represent "is it smart enough to realize it can't win this fight" checks instead.

[Regitnui]

Given that morale checks are an option rule, "despite morale checks" would be pointlessly making checks then ignoring them. Why not only use the optional rule for not-very-intelligent creatures in the first place?

Edit: despite the name, they don't have to represent broken morale and fear. They can easily represent "is it smart enough to realize it can't win this fight" checks instead.

Apologies, I meant it more as a dig at the mechanic, which is a holdover from wargaming when a single miniature represents a squad. A morale check is useful there because a squad that disbands is just as dead as a squad that has been killed. For individual creatures, where the DM can use their judgment, it's really not necessary. I think a look at a creature's Intelligence score will give you a good idea whether it's smart enough to weigh up risk vs reward; Is the score 3+? Then yes, it's going to realize at some point that the adventurers aren't worth its life. If it doesn't have a life, then this is a moot question.

[Tanarii]

Oh, definitely it's a holdover from wargaming. And it's still useful for a DM that prefers the 'neutral referee' type of DMing that arose from wargaming.

It's a resolution method for the DM not to have to introduce his personal bias into whether or not the enemy breaks and retreats. Not all "using your judgement" is a positive thing.

Similarly, many DMs will randomly determine who an enemy targets among PCs, when there's no obvious choice. The idea is: random resolution is a way to make things fair for the PCs, when using it won't break verisimilitude.

Obviously YMMV. But the point is: it is an abstract resolution mechanic so you don't have to think about all the variables that go into if a creature retreats. And in the process risk making effectively arbitrary decisions based on just a subset of those variables, or risk making the same decision every time, or risk bias for or against the PCs.

[Hrugner]

I usually just try to keep an eye on what the enemies want, feel and understand. Some general rules though. For defense I'll have the ranged focus on anyone casting spells and other ranged, then move on to the weakest targets while melee try and engage anyone attacking the ranged. For offense I decide if they are animal, berserk or skirmishers; animals go for opportunity targets and pack attacks, berserk go for biggest or most dangerous target, skirmishers attack ranged and light targets then retreat. Armored targets end up pretty low on the list unless they manage to engage the weaker npcs, have some ranged ability, or if they're on a mount.

If the NPC group has a commander or a spell caster, then the tactics are a bit more encounter specific and often have tells through commands given or uniforms worn.

[Regitnui]

Oh, definitely it's a holdover from wargaming. And it's still useful for a DM that prefers the 'neutral referee' type of DMing that arose from wargaming.

It's a resolution method for the DM not to have to introduce his personal bias into whether or not the enemy breaks and retreats. Not all "using your judgement" is a positive thing.

Similarly, many DMs will randomly determine who an enemy targets among PCs, when there's no obvious choice. The idea is: random resolution is a way to make things fair for the PCs, when using it won't break verisimilitude.

Obviously YMMV. But the point is: it is an abstract resolution mechanic so you don't have to think about all the variables that go into if a creature retreats. And in the process risk making effectively arbitrary decisions based on just a subset of those variables, or risk making the same decision every time, or risk bias for or against the PCs.

I can certainly see the utility of it, and I agree with a lot of what you're saying. It's just not something I'd use. It also raises the question; do the party also need a Morale check after being hit by Frightful Presence and the like?

[Tanarii]

I can certainly see the utility of it, and I agree with a lot of what you're saying. It's just not something I'd use.Fair enough. Personally I don't think a DM should use it 100% of the time. She shouldn't feel like her monsters 'have' to retreat when it'd be stupid. I'm sure that's why it's an optional rule in the first place. D&D isn't intended to be a board game on the DMs side.


It also raises the question; do the party also need a Morale check after being hit by Frightful Presence and the like?Nah, they should be in control of their own role playing.

If players choose to make morale checks to decide if their PC breaks and runs, that's on them. I'm not going to stop them. :smallamused:

[Logosloki]

For laziness I use four categories of thinking for combat threats. Ambush, Offense, Defense and Rout.

Additionally I have recently started using group initiative (and am looking at other systems) as well so I ask the players to roll their initiative and compare it to the enemy initiative. In an ambush the first round is Initiative order and then Ambusher then Ambushee. In Offensive the side with the highest initiative goes first. In Defensive the Defenders go then Attackers go. In a rout it is initiative order but the defenders are legging it so it's all rearguard, tarpitting or full on run.

In an ambush the enemy has time to plan out and so will try and secure dominance by splitting the group up or if they are cocksure then they go for the softer looking targets. In an offense situation the enemy is going to manoeuvre to weak points but will be cautious around any sort of frontline, same in defense. Rout the enemy is going to leg it. I may designate some enemy as brave or browbeaten enough to take a stand and try and force the pursuers to slow down but for the most part they are running, hiding, burning spells, anything to misdirect or get out safely (whether that is themselves or their group is down to ecology).

As for what they target. I tend to do mixed groups of enemies already so not only do the party have ranged options but their opponent does as well. This means that in my games the way I play the enemy is to shutdown the frontline so the mid and back lines have to either burn resources to defend the frontline or find a way to reposition so that they are less exposed.

[Safety Sword]

I wanted to chime in and say that I just play the monsters as intelligently as possible.

It's up the the characters to use their resources to avoid damage or position themselves advantageously.

Just take the kid gloves off and try to kill them. Your players will surprise you. And they'll appreciate it more when they survive truly deadly ambushes or hard battles.

Edit: On morale, the bad guys are usually more scared of the BBEG than the PCs or are actually defending their interests or even their homes. Only the most cunning and powerful NPCs retreat, and only if they have a reason to abandon their current endeavour and still reach their nefarious goals.

[MrFahrenheit]

I'll break down how I calculate a threat from a mechanical perspective. Let's take a look at an encounter with 10,000 XP.

One BBEG, CR = party level, sitting at 10,000? Easy unless the party has an incredibly severe terrain handicap and the BBEG gets a surprise round.
A BBEG at 7,500 and one or two "advisors" making up the rest? Easy if terrain is neutral or in the party's favor, or medium if against the party.
A BBEG at 5,000, an "advisor" at 2,000, and six grunts for the last 3,000? Difficult no matter the terrain, unless the party gets a surprise round.

More is better. And not just from a deadly aspect, but the combat becomes more fun and more intense, too. The average party can lock down any individual opponent, but distract the caster and archer with a couple mooks, and suddenly focusing fire becomes a lot more difficult.

[MaxWilson]

I just eyeball it. If the PCs are outnumbered by a factor of two or more, and the enemies are anything but 10-HP mooks, that's probably a tough fight. If there's a spellcasting dragon on the other side, that's probably a deadly fight (in the true sense of the word, not the D&D-ism) and one I should telegraph to the players instead of throwing at them blind. If there's as many monsters as players, and the monsters are about each about the same CR as the players, that is likewise a fight that the players could lose and I should telegraph to them first.

When in doubt, telegraph and/or foreshadow everything. I'll throw in a random crocodile attack in a swamp, because it's easy and won't kill anybody, but it's still better if I somehow leave signs which in retrospect point to "crocodile must be here."

[mephnick]

Of course, part of the problem is that in D&D effective retreat is really damn hard. Especially if you wait too long, which both DMs and Players tend to do. Even the DMG check points of half creatures dead (group) or half hp lost (creature) often means no chance of escape before death in the case of NPCs/Monsters.

That's my main problem, even though I have enemies run all the time if they're outmatched. It's virtually impossible for humanoid NPCs to escape PCs and fairly tough (without magic) for PCs to escape monsters.

So I can see why a lot of DMs just have everyone fight to the death. Why run when escaping is basically a death sentence?

[Tanarii]

It helps if the monsters (or whatever) have somewhere to retreat *to*. Such as nearby groups of allies, in the hope they can overwhelm the PCs if they all act together. That encourages PCs to stop and think before just charging off after them. But even then they have to make a break for it fairly early.

If the PCs have superior firepower (which is the default assumption of 5e), and the ability to pursue their enemies indefinitely, then every battle is going to be to the enemy's death. In that case, the enemies are about as effective as a mounted knight against a mongol horseman in an open field. They're dead meat no matter what.

[MaxWilson]

That's my main problem, even though I have enemies run all the time if they're outmatched. It's virtually impossible for humanoid NPCs to escape PCs and fairly tough (without magic) for PCs to escape monsters.

So I can see why a lot of DMs just have everyone fight to the death. Why run when escaping is basically a death sentence?

It shouldn't be, especially if the PCs planned ahead. Indoors, you spike a door and/or Hide. Outdoors, you scramble over to your horses and shout "Giddyup!" Because of the way the mounted combat rules work, once you're on the horse you're pretty much home free against any normal monster.

Not planning an escape route is a choice too--just a difference kind of choice. Cortez did that and it worked out okay for him. "On death ground, fight!"

-Max