How often do you target "good saves" or "immunities" on players

[Witty Username]

For the people who say the DM shouldn't metagame at all, would that include encounter balance?

No value judgment, I am actually curious. And a value judgment would make me a hypocrite as I have tossed an Assassin at a level 5 party and Gnoll Flind at the same party when they were level 3.

[Tanarii]

For the people who say the DM shouldn't metagame at all, would that include encounter balance?

All DMs must metagaming. All players must metagame. It's impossible not to, because there's no such thing as total DM/Player-NPC/PC separation. Tailoring is in a similar boat when you run a single group that you can't help but know the character levels. But it's easy to use the "what if I didn't know" guideline as a starting place, unlike metagaming where doing that is usually just as bad as leaning into it.

Designing an area or adventure arc with specific levels in mind, but not specific characters or their specific levels, is what you would do if you didn't know what party levels there were, and then you'd Telegraph or directly communicate difficulty. Picking characters that roughly fit the expected level for an adventure, or going to areas they think they can handle, is what players would do.

Metagaming in general and tailoring for a single party do share one thing in common: acknowledging it can't be avoided entirely and dealing with it as best you can usually enhances the game, but leaning into it or actively trying to avoid it usually hurts the game. The difference is with tailoring, starting from "what if I didn't know" isn't actively trying to avoid it, because it's an knowable answer. Of course, as soon as you get to actual content included within a level bad, you end up right back at the same problem point. But leaning into tailoring isn't the solution either.

[False God]

For the people who say the DM shouldn't metagame at all, would that include encounter balance?

No value judgment, I am actually curious. And a value judgment would make me a hypocrite as I have tossed an Assassin at a level 5 party and Gnoll Flind at the same party when they were level 3.

I mean, I guess it ultimately depends on what sort of world you're designing.

If you're creating a theme park, encounters are balanced for the area, not a party. Once the players have outleveled it, they should move on. Doesn't mean they will, but that's the idea. If the party goes into a higher level area too early, then just like the lower level areas, the encounters are designed for the area not the party. Everything is designed around an estimated level range, and the party either beats the fight...or doesn't.

If you're designing a world where anything could be anywhere, be it random or having pre-set events and creatures, then, again the encounters are only "balanced" as far as "This is a level 15 challenge." Your party might be level 5 when they encounter it, or level 20. They could be highly synergistic, or not, they could he very OP, or not. The encounter remains independent of their qualifications.

EVERYONE should try not to metagame. As best as they can. The player/character dichotomy is not perfect, nor is the DM/NPC. My personal preference is to making worlds that exist independently from the players. So I don't believe encounters should be tailored for the party. The characters have reasonable in-world knowledge, and are able to obtain more via interacting with various elements of the game and choose where they want to go.

Which is to say: I don't like to tailor encounters for my party (except when I'm playing that specific style of game for people who want that). IMO: it's up to the party to tailor their experience within the world. It's not going to be perfect, and they'll sometimes hit harder encounters and sometimes easier ones, but ultimately the control rests with them.

[Mastikator]

For the people who say the DM shouldn't metagame at all, would that include encounter balance?

No value judgment, I am actually curious. And a value judgment would make me a hypocrite as I have tossed an Assassin at a level 5 party and Gnoll Flind at the same party when they were level 3.

IMO not only is it OK to have imbalanced encounters (both harder and easier) but it's better to have a variety of difficulty levels. I'd say it's more important to have encounters that make sense for the story than it is to have encounters that are balanced. The story should drive the encounters, the enemies motives should be what drives them to attack and it should inform their tactics.

[Quertus]

For the people who say the DM shouldn't metagame at all, would that include encounter balance?

No value judgment, I am actually curious. And a value judgment would make me a hypocrite as I have tossed an Assassin at a level 5 party and Gnoll Flind at the same party when they were level 3.

Well, that's not actually a question for me, then, as I believe, if the GM knows Bob is deathly afraid of spiders IRL, the GM should metagame, and not run Arachnophobia.

To answer the question I believe you intended…

I run on two basic modes: "sandbox" and "module".

Sandbox: if the 1st level party decides to agro the ancient red dragon, well, that's what they do. If the epic level BDH party decides to pown some goblins, well, that's what they do.

Module: regardless of whether I made the content (in ignorance of the party) or bought the module (clearly not made with this party in mind), I run a sample party through the module. Then the players can see what kind of characters the module "expects", and the players can metagame accordingly.

And, if the players choose to tell the GM that they want to run a DPS SA build, and the GM knows that approximately none of the foes in Necrophilia on Bone Hill are vulnerable to SA, they can inform the Rogue's player, and the player can make an informed metagame decision, armed with that information.

Or, if the SA Rogue's player chooses not to tell the GM their character, they can play the game honest.

One last bit of "tailoring" I'd like to discuss. If I know I've got someone, for example, into architecture (Roy), instead of just saying "columns", I might do the research to know ahead of time that they're Ionic columns. What they are doesn't change, but the level of detail that the GM gives does.

All DMs must metagaming. All players must metagame. It's impossible not to, because there's no such thing as total DM/Player-NPC/PC separation. Tailoring is in a similar boat when you run a single group that you can't help but know the character levels. But it's easy to use the "what if I didn't know" guideline as a starting place, unlike metagaming where doing that is usually just as bad as leaning into it.

Designing an area or adventure arc with specific levels in mind, but not specific characters or their specific levels, is what you would do if you didn't know what party levels there were, and then you'd Telegraph or directly communicate difficulty. Picking characters that roughly fit the expected level for an adventure, or going to areas they think they can handle, is what players would do.

Metagaming in general and tailoring for a single party do share one thing in common: acknowledging it can't be avoided entirely and dealing with it as best you can usually enhances the game, but leaning into it or actively trying to avoid it usually hurts the game. The difference is with tailoring, starting from "what if I didn't know" isn't actively trying to avoid it, because it's an knowable answer. Of course, as soon as you get to actual content included within a level bad, you end up right back at the same problem point. But leaning into tailoring isn't the solution either.

I mean, all you have to do is to create the content before you know the party. Then you've got 0% tailoring, right? And, if you don't even tell the players what level it's for (a strange practice for a module, but I've seen it happen in a sandbox run Honest, where not everything automatically perfectly telegraphed its difficulty), then you can have 0% metagaming on both ends, can't you?

But, yes, generally speaking, I find people enjoy the game more when they have some idea what kind of characters to bring.

[Tanarii]

I mean, all you have to do is to create the content before you know the party. Then you've got 0% tailoring, right? And, if you don't even tell the players what level it's for (a strange practice for a module, but I've seen it happen in a sandbox run Honest, where not everything automatically perfectly telegraphed its difficulty), then you can have 0% metagaming on both ends, can't you?

But, yes, generally speaking, I find people enjoy the game more when they have some idea what kind of characters to bring.

If you design content for a level band and number of characters, it's still a kind of tailoring, so to speak. Even if you don't know the actual party levels and numbers that will take it on.

[LordCdrMilitant]

I know there is a certain "metagame" going on within RPG systems. I am taking D&Desque approach because that is familiar to most of you. I want to know if you still target immune characters with stuff.

Example group:
Paladin is immune to mind affecting stuff and disease.
Druid is immune to poison and summons a fire elemental.
Tiefling Rogue has evasion and resists fire, cold and electricity.
The wizard has protection from missiles and shield up.

Would you target the wizard with magic missile or arrows? Would you use your succubus to try and seduce the paladin infight? Fireball the fire elemental and rogue? Try to poison the druid? Or would you go the opposite? Poison the wizard, fireball the wizard and paladin, seduce the rogue?

Wasting ones turns for the sake of RP seems feasible, but a bit pointless if only one side does it. On the other hand, I feel it should be obvious the zealous paladin does not easily succumb to a demon's lures, or the fast rogue is hit by a big explosion. On the other hand, it does make a fair few very good class features just pointless, when you're never targetted anyhow.

Just a quibble: if you're never targeted because of an ability, it's actually being quite reasonably valuably and isn't useless.

Anyway, onto the main point:
It depends on the intelligence (capability for logic, reasoning, and tactics) and the intelligence (information known about the party, their objectives, and their tactics) of the enemies involved in the encounters.
If an enemy is intelligent and sufficiently informed to know that a character has a strong defense against something [which might be pretty obvious from cursory observation even if they haven't met before. Most parties don't generally go out of their way to conceal what they can do.] the enemy will instead use methods that circumvent that defense.
If an enemy is intelligent, but doesn't know that a character is strong against something, they will act as if they player didn't have that defense, [at least until they find out]. This doesn't mean they'll go out of their way to use an ability that wouldn't affect a player, just that they have no inhibitions on using that ability if it would appear to be tactically advantageous to do so absent the information that a player is resistant to it.
If an enemy isn't capable of developing more advanced strategy and tactics than monsters and beasts, then it'll approach the situation in accordance with it's 'standard' strategy and tactics and won't adapt to the party's strengths and weaknesses, and will instead flee if things don't work.

Given that I generally don't feature encounters with monsters and beasts unless the situation specifically calls for it, [Several factors: 1: I think random encounters just to have a combat that have no contribution to the story are a waste of session time to run and my time to prep. 2: I assume that the vast majority of creatures that are just hunting or wandering will, like real predators, generally avoid humans. 3: Most travel-routes, unless there's some specific reason for it not to be, are safe enough for unprotected travelers that any armed travelers like a party would have a trivial time dealing with threats along the road], I don't often deliberately target a player with an ability I know they resist. However, when the players go out of their way to conduct deception and misinformation operations, they're rewarded appropriately by seeing enemies using tactics and weapons they think are effective that actually aren't.

[Quertus]

If you design content for a level band and number of characters, it's still a kind of tailoring, so to speak. Even if you don't know the actual party levels and numbers that will take it on.

But if you didn't *design* it for such, simply created it, and tested it, and found that it *happened* to be appropriate to such?

Like… if we started with the epic challenge of the locked door, moved on to the CR 1/10 "pile of papers" trap, continued with the toxic "what's gone bad in my fridge" demon, and finished with the monster under my bed, to retrieve the golden pillow. Then we could look at that adventure, and evaluate who it's appropriate for.

Yes, it's tailoring on the player end if the players choose a number of characters vaguely in that band. Which is probably for the best, in terms of what usually makes a fun game, we agree. But it technically isn't *required* - one could just grab a module at random for the party to run through, or the players could always say, "we know you said 'levels 10-12, but…" ["our epic level characters would, realistically, be interested in that plot line" / "we really want the challenge of trying it with 5th level characters"].

And, while I haven't seen that "go well" from a challenge perspective, I have seen players have fun with it (mostly in more sandboxy settings).

[Tanarii]

But if you didn't *design* it for such, simply created it, and tested it, and found that it *happened* to be appropriate to such?

Like… if we started with the epic challenge of the locked door, moved on to the CR 1/10 "pile of papers" trap, continued with the toxic "what's gone bad in my fridge" demon, and finished with the monster under my bed, to retrieve the golden pillow. Then we could look at that adventure, and evaluate who it's appropriate for.

Agreed, that wouldn't be tailoring at all.

[Spore]

If you design content for a level band and number of characters, it's still a kind of tailoring, so to speak. Even if you don't know the actual party levels and numbers that will take it on.

Tailoring is absolutely needed. I had a few NPCs walk into a low magic horror campaign, with their famed and favorite magic weapons still intact. As a consequence they tore the setting a new hole in the first few encounters, and only a display of uncertainty (aka not revealing the main monster they could've EASILY dispatched) did invoke some kind of fear in them.

As a result, I finally just doubled all the numbers of the game's statblocks, and suddenly the scary enforcer of the town's villain was an actual memorable victory rather than a cakewalk. If I had let them run amok with their weapons and gear, they could have easily stormed the final villain's lair at half the level they were supposed to, but metagaming assumptions on their part would have had them search the whole setting for a clear advantage before engaging the final boss. A long and gruelling boring travel.

But I agree here, that tailoring the boss to counter group's strengths or weaknesses was not the right call. I just made him powerful on all fronts, and they won because their strengths kicked in. My changes made the boss a demon, which powered up the paladin, and while his new form had damage resistances, the casters were not able to disable him, but weaken him enough so the paladin survived on single digit HP.

[Duff]

Plus it's an awesome moment when the succubus comes up and tries their seduction on the paladin player only for a big grin to cross the players face as they get to make some witty "I'm sorry, but you're not my type" style comeback (followed by repeated stabbings).

In design of an encounter, it's good to try and make these come up for all your characters at times. Give everyone their chance to shine. I guess maybe in 1/2 of sessions someone should be getting to be shiny?

Within the encounter (I'm not sure which level your question was at) play the monster. Make choices based on what the monster knows or can be expected to deduce - don't target the AC of the guy in sheet steel, don't target the mind of the monk and don't fireball the red dragonborn with flames on their shield.


Example group:

Paladin is immune to mind affecting stuff and disease.
How clearly marked as a Paladin rather than a fighter are they? Assuming it's "Pretty obvious" and a reasonably well informed opponent (ie, someone who knows what paladins are like) then this is a no unless "Taking the shot on the off chance it's just a fighter with a holy aesthetic" is appropriate for the character. OTOH, an opponent who doesn't know paladins but does know fighters should have a few goes at this. And an animal type thinker will simply try to hit whoever looks most threatening or whoever looks most likely to become lunch

Druid is immune to poison and summons a fire elemental.
Much the same - how obviously are they a druid? How much does their enemy know about druids?. I think anything which uses fire attacks probably knows not to target the fire elemental with those

Tiefling Rogue has evasion and resists fire, cold and electricity.
Do all Teiflings get those resists? if so, as above but for "Teifling" and "Rogue" but note that a rogue's look and a "mobile fighter" or "Budget Fighter" look can be more similar, making the bar a bit harder to know about the evasion. If you get the chance to fireball and include the teifling you should. Given the choice of them or the druid with the fire elemental, go the rogue unless you know teiflings are resistant

The wizard has protection from missiles and shield up.
That's pure guesswork. Unless your attacker knows this wizard, they won't make any assumptions about which spells the PC knows or has cast. Maybe make a specific exception if the attacker is a wizard who always precasts those specific spells

[Chauncymancer]

Is this a serious suggestion? Because I have never heard anyone saying "Hey, wouldn't it be a great idea to attack a being made of fire with fire?".

You stab the iron golem with a sword, do you not? Punching a man in the nose is attacking a being made of meat with meat.

[Max_Killjoy]

I know there is a certain "metagame" going on within RPG systems. I am taking D&Desque approach because that is familiar to most of you. I want to know if you still target immune characters with stuff.

Example group:
Paladin is immune to mind affecting stuff and disease.
Druid is immune to poison and summons a fire elemental.
Tiefling Rogue has evasion and resists fire, cold and electricity.
The wizard has protection from missiles and shield up.

Would you target the wizard with magic missile or arrows? Would you use your succubus to try and seduce the paladin infight? Fireball the fire elemental and rogue? Try to poison the druid? Or would you go the opposite? Poison the wizard, fireball the wizard and paladin, seduce the rogue?

Wasting ones turns for the sake of RP seems feasible, but a bit pointless if only one side does it. On the other hand, I feel it should be obvious the zealous paladin does not easily succumb to a demon's lures, or the fast rogue is hit by a big explosion. On the other hand, it does make a fair few very good class features just pointless, when you're never targetted anyhow.

Depends on what the character could reasonably know about the target and various levels of resistance to attacks.

This is one of the things that can be done to give weight to characters with a scholarly or tactical bent, or a real drawback to a character with low INT and no knowledge skills, without reducing it to raw marginal dice effects.

[LibraryOgre]

You stab the iron golem with a sword, do you not? Punching a man in the nose is attacking a being made of meat with meat.

One might be burning away the fuel the fire being needs, after all.

[HidesHisEyes]

I know there is a certain "metagame" going on within RPG systems. I am taking D&Desque approach because that is familiar to most of you. I want to know if you still target immune characters with stuff.

Example group:
Paladin is immune to mind affecting stuff and disease.
Druid is immune to poison and summons a fire elemental.
Tiefling Rogue has evasion and resists fire, cold and electricity.
The wizard has protection from missiles and shield up.

Would you target the wizard with magic missile or arrows? Would you use your succubus to try and seduce the paladin infight? Fireball the fire elemental and rogue? Try to poison the druid? Or would you go the opposite? Poison the wizard, fireball the wizard and paladin, seduce the rogue?

Wasting ones turns for the sake of RP seems feasible, but a bit pointless if only one side does it. On the other hand, I feel it should be obvious the zealous paladin does not easily succumb to a demon's lures, or the fast rogue is hit by a big explosion. On the other hand, it does make a fair few very good class features just pointless, when you're never targetted anyhow.

Haven’t had time to read the whole thread so sorry if I’m not saying anything new but…

I’d say it depends on the specifics of the enemy using the ability and the nature of the PC’s resistance. If the resistance is obvious and/or the enemy is reasonably smart, I think it’s fair to have them avoid using abilities that won’t work. But if a pc has a certain resistance then at least sometimes you want them to have the cool “nope” moment and leave the enemy shaking their fist in futile rage.

That said, I’d argue that even if you have the enemies never ever use abilities that you know won’t work, that doesn’t necessarily render the PCs’ resistances pointless. If you have an “immunity to charm” feature and because of that enemies never even attempt to charm you, you’ve still got the result of never being charmed. Sure, that enemy may decide to just punch you instead, but often enemies’ special attacks are more powerful than their more standard options, so when that enemy punches you instead of charming you you’re still getting off lightly because of the class feature. It’s like carrying an umbrella because it’s sod’s law that if you don’t have one it will rain.

[Duff]

One might be burning away the fuel the fire being needs, after all.

Good point. Though not a point that I've ever seen applied in D&D logic.
As far as I can think of, any case where the damage is coming significantly from the "element" of the attack, beings which are made of the element are immune. EG, fire, acid, electricity, cold.
Where the attack comes from what the element does, they might be affected.
I don't know if there is one, but imagine an ice creature which stabs it's enemy with Icey blade-hands. By D&D logic, this thing will be immune (or resistant) to ice damage but probably not to stab* damage. 2 of them could fight each other and poke holes but won't be doing as much damage to each other as to my warm fleshy PC

* OK, they might be resistant to stabs as the points skate off their hard slippery surface. But that's definitely optional.

[Rogan]

You stab the iron golem with a sword, do you not? Punching a man in the nose is attacking a being made of meat with meat.

But not an iron sword. Better use Adamantine, so this annoying DR doesn't stop me.
And everybody knows that unarmed combat is totally underpowered.

[Quertus]

That said, I’d argue that even if you have the enemies never ever use abilities that you know won’t work, that doesn’t necessarily render the PCs’ resistances pointless. If you have an “immunity to charm” feature and because of that enemies never even attempt to charm you, you’ve still got the result of never being charmed. Sure, that enemy may decide to just punch you instead, but often enemies’ special attacks are more powerful than their more standard options, so when that enemy punches you instead of charming you you’re still getting off lightly because of the class feature. It’s like carrying an umbrella because it’s sod’s law that if you don’t have one it will rain.

I'd say that, for the emotional impact, unless the PCs get to automatically know and avoid the resistances of the monsters, too, it's less like carrying an umbrella and never seeing rain, and not like bringing a raincoat, and never getting a date. Sure, part of that end result may have seemed life what you wanted, bit the path to get there makes all the difference.

[HidesHisEyes]

I'd say that, for the emotional impact, unless the PCs get to automatically know and avoid the resistances of the monsters, too, it's less like carrying an umbrella and never seeing rain, and not like bringing a raincoat, and never getting a date. Sure, part of that end result may have seemed life what you wanted, bit the path to get there makes all the difference.

Not sure I understand you, but if you mean it’s more satisfying to have the “no that doesn’t work on me” moment then yeah, as I said in my post. I was just adding that having enemies avoid resistance doesn’t render the resistance completely pointless, as op suggested.

[Tanarii]

One thing to keep in mind, if an enemy doesn't target you due to immunity or resistance, that doesn't necessarily mean it "worked". If they just target another ally, there may be no real net benefit to the group.

This also holds true for something like having an incredibly high AC or a truck-load of hit points.

[Quertus]

One thing to keep in mind, if an enemy doesn't target you due to immunity or resistance, that doesn't necessarily mean it "worked". If they just target another ally, there may be no real net benefit to the group.

This also holds true for something like having an incredibly high AC or a truck-load of hit points.

The Vampire goes to Dominate someone. He sees a floating man, covered in Arcane symbols; a man in simple robes, unarmed, in a trained fighting stance; a creature, less man than breast, lurking in the shadows; and a robot.

What if the robot, immune to mind control, had been something else?

If the Vampire is played with omniscience, then the odds of it targeting the weakest link are always 100%; the only way to change the outcome would be to run someone with a worse resistance to Domination.

If the Dominate effect is "at will", *and* completely undetectable, the only difference is between "robot" (obvious immunity) and "undetectable immunity". Which is just a difference in how long until the Vampire orders the others to kill you (or otherwise enacts whatever plan it was dominating them for). Or "not immune", in which case the replacement joins the thralls.

However, if the Vampire's just making a reasonable guess? If their defenses are as obvious as the alternate character's, then it's either the same case as omniscience, or move from ⅓ to ¼ odds of choosing correctly.

However, with the robot, the player knows that they will never have to worry about being controlled.

And, if they're robot Batman, they've already got contingency plans to handle the scenario of anyone else being controlled.

If they're most anyone else? Tactically, yeah, the party's still a man down. But if it's Max playing the robot, and me playing the dominated thrall? We're both happy.

So there's a little more to the calculus than just "always a man down". But, yeah, it's a good argument against (the ability having its proper value when used vs) omniscience.

[Tanarii]

I was purposely ignoring the debate on how they should determine the target for their ability. Point was, once they target someone else in the group instead of you, that may not be any net benefit to the group. As opposed to the individual.

Or maybe it will be a net benefit. But there seemed to be a lot of assumption going on in this thread that just because the individual doesn't get targeted once the enemy has some reason to believe their ability won't work, that the immunity did something that is positive. That it still "worked". That's not necessarily the case, on a party level.

Certainly for something that affects the group as a whole, like immunity to fire and a Fireball hitting the party, there's still benefit.

[HidesHisEyes]

I was purposely ignoring the debate on how they should determine the target for their ability. Point was, once they target someone else in the group instead of you, that may not be any net benefit to the group. As opposed to the individual.

Or maybe it will be a net benefit. But there seemed to be a lot of assumption going on in this thread that just because the individual doesn't get targeted once the enemy has some reason to believe their ability won't work, that the immunity did something that is positive. That it still "worked". That's not necessarily the case, on a party level.

Certainly for something that affects the group as a whole, like immunity to fire and a Fireball hitting the party, there's still benefit.

Yeah I see what you mean, and overall I’m still coming come down on the “don’t always have the enemies act with omniscience” side. But I did take issue with the idea that if the enemies don’t target your resistance then the resistances are completely meaningless. When it comes to calculating how much of an advantage such a resistance gives the party as a whole if the enemy just targets someone else… oof, seems complicated. But the fact is, that immunity to charm still lets that player say “I in particular never get charmed ever”, and that’s meaningful even if it might sometimes turn out to be a bad thing for the party. There’s a bit of tension between potential playstyles here, because if you’re approaching the game as a fully strategic/tactical exercise then of course you don’t want character features that create a net negative for the party - but there are games, playstyles and groups that value “impact on narrative” at least as highly as effectiveness for defeating enemies and overcoming obstacles.

[False God]

I was purposely ignoring the debate on how they should determine the target for their ability. Point was, once they target someone else in the group instead of you, that may not be any net benefit to the group. As opposed to the individual.

Or maybe it will be a net benefit. But there seemed to be a lot of assumption going on in this thread that just because the individual doesn't get targeted once the enemy has some reason to believe their ability won't work, that the immunity did something that is positive. That it still "worked". That's not necessarily the case, on a party level.

Certainly for something that affects the group as a whole, like immunity to fire and a Fireball hitting the party, there's still benefit.

I mean, this is a sound argument for more bounded limits on defenses. Total immunity (and AC to high as to effectively be total immunity) gets the boot in favor of improved forms of immunity. Anyone could get hit by the effect in question, some people are more likely going to pass their saves or fail, but it still leaves some minor percentile room for failure. Likewise, we should also eliminate the possibility of assured failure. No PC can have a a defense so low as to always fail.

It's very gamist to design a system this way though.

[Mastikator]

Yeah I see what you mean, and overall I’m still coming come down on the “don’t always have the enemies act with omniscience” side. But I did take issue with the idea that if the enemies don’t target your resistance then the resistances are completely meaningless. When it comes to calculating how much of an advantage such a resistance gives the party as a whole if the enemy just targets someone else… oof, seems complicated. But the fact is, that immunity to charm still lets that player say “I in particular never get charmed ever”, and that’s meaningful even if it might sometimes turn out to be a bad thing for the party. There’s a bit of tension between potential playstyles here, because if you’re approaching the game as a fully strategic/tactical exercise then of course you don’t want character features that create a net negative for the party - but there are games, playstyles and groups that value “impact on narrative” at least as highly as effectiveness for defeating enemies and overcoming obstacles.

It has to do with action economy, if a PC never targets an enemy that happens to be immune to the effect you're targeting them with then you're never "wasting" an action then they're always efficient, if they are able to do that by reading the monster manual or DM notes then that's tantamount to cheating in my opinion. Guessing that a firebreathing dragon is immune or resistant to fire is fine, knowing that a shambling mound is healed by lightning (without prior PC knowledge) is cheating.

It's exactly the same if the DM does it, if it's generally known that elves are resistant to enchantments then by all means have the enemy target someone else. But if it's not clear to the enemy that the monk is an elf and the paladin is a human and the enemy still targets the paladin and not the elf then that's cheating and the DM has no business doing it.

It's not a wargame and you should not treat it like a wargame, the goal of the game is not to win battles. Battles are a means to an end!

[Max_Killjoy]

I was purposely ignoring the debate on how they should determine the target for their ability. Point was, once they target someone else in the group instead of you, that may not be any net benefit to the group. As opposed to the individual.

Or maybe it will be a net benefit. But there seemed to be a lot of assumption going on in this thread that just because the individual doesn't get targeted once the enemy has some reason to believe their ability won't work, that the immunity did something that is positive. That it still "worked". That's not necessarily the case, on a party level.

Certainly for something that affects the group as a whole, like immunity to fire and a Fireball hitting the party, there's still benefit.

If there's a waste of time, actions, resources, etc in making the attempt on the here-to-fore unknown to be immune target, that's still of some benefit.

[Xervous]

I mean, this is a sound argument for more bounded limits on defenses. Total immunity (and AC to high as to effectively be total immunity) gets the boot in favor of improved forms of immunity. Anyone could get hit by the effect in question, some people are more likely going to pass their saves or fail, but it still leaves some minor percentile room for failure. Likewise, we should also eliminate the possibility of assured failure. No PC can have a a defense so low as to always fail.

It's very gamist to design a system this way though.

How long down the resolution chain does this need to hold true? If Joe uses X on Bob while condition Y is present is a 100% failure rate on Bob’s part acceptable, given that Y is not guaranteed to be present?

[kyoryu]

Enemies should player according to their intelligence, the knowledge they have, and the general knowledge available.

If wizards are common, it's not unreasonable for most people to know that wizards have some kind of magical shield thing that makes arrows miss. Of course, the response they have shouldn't be based on whether that particular wizard has the spell, as they have no way to know that.

Targeting solid defenses is going to be case by case. There's really three situations for smart critters.

1) the enemies have no info. Target whatever makes tactical sense without that info.
2) The enemies have generic info. Probably avoid targeting the strong defenses, unless it's tactically worth it. Do this whether or not they actually have those defenses. Generic info is stuff like "he's got big armor, he's probably hard to hit and take down physically, but should be weaker to mind stuff unless he does that weird holy stuff" kind of thing. Or, "wizards can block arrows". So, yes, a wizard without Shield can benefit if it's generally believed that wizards have the Shield spell.
3) The enemies have specific info. As above, but actually tailor it to the PC's actual defenses.

[HidesHisEyes]

It has to do with action economy, if a PC never targets an enemy that happens to be immune to the effect you're targeting them with then you're never "wasting" an action then they're always efficient, if they are able to do that by reading the monster manual or DM notes then that's tantamount to cheating in my opinion. Guessing that a firebreathing dragon is immune or resistant to fire is fine, knowing that a shambling mound is healed by lightning (without prior PC knowledge) is cheating.

It's exactly the same if the DM does it, if it's generally known that elves are resistant to enchantments then by all means have the enemy target someone else. But if it's not clear to the enemy that the monk is an elf and the paladin is a human and the enemy still targets the paladin and not the elf then that's cheating and the DM has no business doing it.

It's not a wargame and you should not treat it like a wargame, the goal of the game is not to win battles. Battles are a means to an end!

I see where you’re coming from. For me that opens out onto a larger discussion about to what extent D&D actually is a war game, which is complicated and I won’t go into it… but I’ll mention that I’ve largely stopped playing D&D and almost entirely stopped GMing it… because I wanted to play an RPG and not a war game.

[Weasel of Doom]

As a player I agree with the people saying monsters should be GMed as you would realistically expect them to act. That's how it goes for our group and I like it - some monsters will attack the weakest looking character because they're cowardly or just want the easiest meal available, others will attack the character they think is likely to be most threatening to knock them out early, others will just go for whoever's closest and others might go for whoever looks like they'll provide a challenging fight.
It's the same with specific attacks - monsters with reasonable intelligence or experience should know not to attack the wizard or the cleric with things that target a will save ... if they can work out who the wizard or cleric are (and that depends on the player's character descriptions, gear and, as the fight goes on, their actions).

It's important to play monsters to their level of INT/experience so that when the party are specifically targeted then they notice the contrast. A couple of sessions ago my party finally took out the assassin that had been dogging our heels for months IRL. He was particular challenge and a hated foe because he was the one hunting us and spent enough time spying on the party to specifically target weaknesses and not target "good saves" unless he had no choice. We could immediately feel the difference between stumbling upon a surprised monster in a dungeon vs being hunted by an intelligent and prepared foe.