Party optimisation philosophy

[MinotaurWarrior]

I think there is such a thing as winning more.

Worst: losing/dying/TPK

Barely winning: pyrrhic victory, winning at a permanent cost.

Winning: winning at the cost of significant limited resources like spell slots and HP.

Winning more: winning with little or no risk or resource expenditure (e.g. caltrops and some arrows).

Winning even more: winning in a way that makes you even stronger than you were going in, e.g. extracting poison from a dead Purple Worm, Polymorphing a Neothelid into a frog to make a Neothelid grenade that you can throw at a lich to maybe Feeblemind or kill it, taking control of a Mummy Lord via Command Undead, claiming an Orthon's infernal crossbow as your own.

That's true encounter by encounter, but what's the point of all of those things you mentioned?

Barely winning: the permanent cost makes you worse at dealing with all future challenges

Winning: you can't use those limited resources for the next challenge in the day

Winning more: you have more resources for the day

Even more: you can use that poison, "neothelid grenade", mummy thrall, etc to have even more resources for future encounters than you started with.


For a more general example, consider suggestion. I think that having the ability to cast suggestion can have a huge effect on your party's overall success chance, because you can use it to win one encounter (this high CR enemy with a shared language is attacking!) and then use the results of that encounter to win the next (that high CR enemy killed the monsters in the next room).

Concretely, consider the following adventure for a level 3 lore bard with suggestion:

Bandits have captured a noble's child and are holding them hostage. The goal is to rescue the children.

Encounter 1: A bandit captain comes, anticipating the ransom be delivered to him. Bard with +4 to initiative has a 55% of going first, bandit has a 45% chance of failing its save "Why don't you work for the nobility?". Deadly encounter won.

Encounter 2: A bandit patrol of two bandits outside of their camp. The bandit captain kills them all on its own.

Encounter 3: Four bandits in the camp. The bandit captain and the bard have to work together, but they get through it without another 2nd level spell slot being used.

Encounter 4: The other bandit captain is in his tent with the child. Suggestion is used again. The child is rescued.

Hypothetically, suggestion allowed the bard to "win more" by leveraging the 1st bandit captain to more easily deal with that very challenging (450xp over budget) day. But at the end of the day, all that actually matters is that the child was rescued successfully.

Compared to, say, a law firm, where at the end of the year you might have made five million dollars or six million dollars depending on how well the firm did (caused by the contribution of the various staff members collaborating).

Your version of "winning more" folds neatly into overall chance of victory.

[KorvinStarmast]

Bard Bandit Encounter example, nicely done

But if the save isn't made, it's a TPK of a one person party. YOLO situation, rather than "avoid disaster" situation.
Interestingly, I think you bard example makes one of your earlier points obsolete.

Party optimization needs to rely on a lot of the same abstractions as character optimization, e.g. DPR and Effective HP as proxies for %victory. Because they work well enough when most of what you're trying to do is reduce the other side to 0HP while keeping your HP above 0.

Your bard versus bandits (Hey, wait, what a great name for a game!) scenario isn't a DPR fest.

[MinotaurWarrior]

But if the save isn't made, it's a TPK of a one person party. YOLO situation, rather than "avoid disaster" situation.
Interestingly, I think you bard example makes one of your earlier points obsolete.
Your bard versus bandits (Hey, wait, what a great name for a game!) scenario isn't a DPR fest.

Sorry I could have been clearer there. DPR is just an example of one of those standard measures we use when optimizing characters. I don't mean to come across like everything is a DPR fest.

For example here, on the solo bard party I might say something like, "The bard's DC 13 Wis save beats X% of level appropriate monsters". Or, "The Diviner with Portent has a 9.75% of rolling a one which can be used to cause a monster to auto-fail its save". Whatever abstractions are typical when discussing spell-casting ability in single-character build discussions.

[MaxWilson]

That's true encounter by encounter, but what's the point of all of those things you mentioned?

...

Hypothetically, suggestion allowed the bard to "win more" by leveraging the 1st bandit captain to more easily deal with that very challenging (450xp over budget) day. But at the end of the day, all that actually matters is that the child was rescued successfully.

Compared to, say, a law firm, where at the end of the year you might have made five million dollars or six million dollars depending on how well the firm did (caused by the contribution of the various staff members collaborating).

Your version of "winning more" folds neatly into overall chance of victory.

It's the same as the law firm: if you can accumulate assets you're better off at the end of the year. "Winning most" in this context means you've now got a bandit captain working for you as a hireling, unless/until you get him killed off on a different pyrrhic victory somewhere.

[KorvinStarmast]

For example here, on the solo bard party I might say something like, "The bard's DC 13 Wis save beats X% of level appropriate monsters". Or, "The Diviner with Portent has a 9.75% of rolling a one which can be used to cause a monster to auto-fail its save". Whatever abstractions are typical when discussing spell-casting ability in single-character build discussions.

Got it, I see the analytical method there, makes sense.

[MinotaurWarrior]

It's the same as the law firm: if you can accumulate assets you're better off at the end of the year. "Winning most" in this context means you've now got a bandit captain working for you as a hireling, unless/until you get him killed off on a different pyrrhic victory somewhere.

Which is better:

The campaign ends. You stopped Tiamat (or whatever). You have two million GP, a giant zombie army, 500 wands of magic missile, and extraplanar servants.

The campaign ends. You stopped Tiamat (or whatever). You have none of that.

My answer: they're the same. You won. The campaign is over. The fictional world disappears after you're done playing in it. Unlike the Law Firm's assets which survive even after real-life human death.

You're right that results from one encounter carry over into the next (and to a lesser extent, from one day to the next), but that should be factored into your overall chance of victory. The party that uses every spell slot at goes down to 1HP a person after the first deadly encounter has a lower chance of victory than the one that gains resources after the first encounter.

[MaxWilson]

Which is better:

The campaign ends. You stopped Tiamat (or whatever). You have two million GP, a giant zombie army, 500 wands of magic missile, and extraplanar servants.

The campaign ends. You stopped Tiamat (or whatever). You have none of that.

My answer: they're the same. You won. The campaign is over. The fictional world disappears after you're done playing in it. Unlike the Law Firm's assets which survive even after real-life human death.

You're right that results from one encounter carry over into the next (and to a lesser extent, from one day to the next), but that should be factored into your overall chance of victory. The party that uses every spell slot at goes down to 1HP a person after the first deadly encounter has a lower chance of victory than the one that gains resources after the first encounter.

Which is better: dying with a million dollars in the bank, or dying with a hundred billion dollars in the bank?

Law firm is the same as the campaign. In both cases, you either have uses for the extra resources during the epilogue, or you don't. If you don't then both endings are the same. If you do, they're not.

P.S. This might be a playstyle difference too. It sounds like maybe you're assuming the whole campaign is built around a single conflict. To me that sounds like a campaign arc (possibly one of many overlapping arcs, some player-driven and some DM-created), not a whole campaign. Why would you stop playing and retire those PCs just because Tiamat was defeated? IME you wouldn't. You'd keep going until Real Life killed the campaign or a different set of PCs became more interesting.

[MinotaurWarrior]

Which is better: dying with a million dollars in the bank, or dying with a hundred billion dollars in the bank?

Dying having made a hundred billion (not necessarily in the bank) because the world continues to exist after you die. For example, you leave it to your kids, or you donate it to charity, and the beneficiaries continue to benefit down the line until we all die out.

P.S. This might be a playstyle difference too. It sounds like maybe you're assuming the whole campaign is built around a single conflict. To me that sounds like a campaign arc (possibly one of many overlapping arcs, some player-driven and some DM-created), not a whole campaign. Why would you stop playing and retire those PCs just because Tiamat was defeated? IME you wouldn't. You'd keep going until Real Life killed the campaign or a different set of PCs became more interesting.

Regardless of why the campaign actually ends (I think it's fair to say RL kills most campaigns for most people) campaigns are bounded. At the end, either you won or you lost, and then *poof* everything you built up is gone.

I guess the exception would be if you were theoretically willing to play one campaign indefinitely. Theoretically there are optional rules for epic boons and all that, and I will say that as a matter of personal preference I'd never want to play an "epic level" game.

But even in that case, because of the principle that "You've got to get there first" the impact of theoretical session 217, 180 sessions in to being level 20, doesn't actually matter that much.

[MaxWilson]

Regardless of why the campaign actually ends (I think it's fair to say RL kills most campaigns for most people) campaigns are bounded. At the end, either you won or you lost, and then *poof* everything you built up is gone.

See, that's the thing: "winning more" on a tactical or adventure level increases the chances you'll get to something that feels like a gameworld/campaign-level win before Real Life kills it off. You can tackle the bigger problems now instead of six months from now.

[Frogreaver]

Good Thread.

Party Optimization - It depends on the campaign you are going to be in. It depends what level it goes to. It depends on the pace of encounters and short rests throughout the day. It depends on the types of encounters you will be facing. It depends on your DM. It depends on what goal you are optimizing toward.

That said I think there are some principles that will help guide us. All enemies can be defeated with damage. Thus, damage is the foremost consideration. The next consideration ultimately boils down ensuring you are able to keep doing damage. That may be through control, defensive buffing, healing, self defenses, counterspells, etc. Thirdly, there are the abilities that fundamentally change the nature of encounters, such as scouting, surprise, avoiding surprise, disarming/bypassing traps, bypassing encounters entirely, using social skills/suggestion/charm/dominate to gain allies while taking away enemies, etc.

What I find is that abilities in that third category are often the highest impact but least reliable - at some point you'll fail stealth checks, same with perception checks, same with suggestion and saving throws, even persuasion tends to have DM limitations put on it for what it can ultimately accomplish, teleportation may get you past the encounter now, but there's always the possibility that whatever you teleported from may seek you out later.

Bards, Wizards, Druids tend to be fairly good at this role. Rogues and Clerics aren't terrible at it.

Control abilities are also not the most reliable, but are some of the most important abilities for turning an otherwise hard encounter easy. I would place opportunity attacks under this section as well, though they are a poor form of control typically.

Bards, Wizards, Sorcerers and even Warlocks fair well here. Clerics have a small bit of control.

Defensive abilities are primary about party defense/healing - although you want to have enough durability to take a few hits. This is also where the role of a tank can come in (although tanking is highly dependent on how the DM runs enemies). These abilities help you keep the worst case from happening and tend to be fairly reliable.

Clerics, Bards, Druids, Wizards, Sorcerers are all great here. Paladins deserve honorable mention.

Finally we get to Damage. There's lots of ways to do good damage. Generally SS or GWM builds work the best. However, druids can do good damage, clerics do well to if they burn slots on spirit guardians + spiritual weapon in a fight. Sorcerers can twin buff allies to add alot of extra damage. Animate objects does good damage but that's moving.

If you cover these bases well, you'll have a fairly optimized party. I wouldn't go for a solely gimmick party, because at some point the gimmick will fail.

[Skylivedk]

@Frogreaver:
Is it on purpose that you ignore the Social and Exploration pillars? Is it a reflection of how your table plays?

It's extremely far from my play experience for the last 20 years or so. Most of my group's most successful "win more"-strategies have been focused on circumventing dealing damage and instead convert hostiles into allies/tools.

Example from a 3.x. campaign:
We accidentally murder hobo'd a kinda friendly Minotaur Prince (one character hadn't been there as we used illusions and persuasion to make the Prince friendly).

That was at least a minor annoyance. Being in the Prince's good graces would have led to a bunch of benefits and cool areas.

Nevertheless, you know what to do, when you accidentally murder someone? Blame someone else. So we did. There was a kingdom of elves that we thought needed a change of leadership. By using some of the insignia from the Minotaur Prince (and I think some polymorphing), we successfully started a full scale war between the minotaurs and the elves, which weakened both and allowed us greater freedom to sneak in and unalive the pesky Elf king.

The killing was way easier and less important than all the subterfuge. I would also say the killing is less fun than all the scheming and world building. I would not say a party is optimised if all it is good at is increasing the amount of corpses in a given area.

[Frogreaver]

@Frogreaver:
Is it on purpose that you ignore the Social and Exploration pillars? Is it a reflection of how your table plays?

It's extremely far from my play experience for the last 20 years or so. Most of my group's most successful "win more"-strategies have been focused on circumventing dealing damage and instead convert hostiles into allies/tools.

Example from a 3.x. campaign:
We accidentally murder hobo'd a kinda friendly Minotaur Prince (one character hadn't been there as we used illusions and persuasion to make the Prince friendly).

That was at least a minor annoyance. Being in the Prince's good graces would have led to a bunch of benefits and cool areas.

Nevertheless, you know what to do, when you accidentally murder someone? Blame someone else. So we did. There was a kingdom of elves that we thought needed a change of leadership. By using some of the insignia from the Minotaur Prince (and I think some polymorphing), we successfully started a full scale war between the minotaurs and the elves, which weakened both and allowed us greater freedom to sneak in and unalive the pesky Elf king.

The killing was way easier and less important than all the subterfuge. I would also say the killing is less fun than all the scheming and world building. I would not say a party is optimised if all it is good at is increasing the amount of corpses in a given area.

Social and Exploration were talked about in my post...

[KorvinStarmast]

That said I think there are some principles that will help guide us. All enemies can be defeated with damage. Thus, damage is the foremost consideration.

In a CRPG/ARPG like Diablo III? Yes. In D&D? Maybe not.

Sun Tzu: To win one hundred victories in one hundred battles is not the acme of skill. To subdue the enemy without fighting is the acme of skill.

Three pillars, not one. Some of the published adventures have some interesting chances to get a monster/NPC on your side (STK being one of them), and surprisingly, Tomb of Annihilation was another.

Control abilities are also not the most reliable, but are some of the most important abilities for turning an otherwise hard encounter easy.

Yes. A dozen times yes. They also let you sometimes evade or avoid an encounter that consumes resources you need for a much harder one a few rooms/encounters down stream.

I wouldn't go for a solely gimmick party, because at some point the gimmick will fail.

Amen, deacon. Preach it!

[KorvinStarmast]

I would not say a party is optimised if all it is good at is increasing the amount of corpses in a given area.

I tend to agree, with the caveat that sometimes, one does need to leave the field of battle littered with the corpses of one's enemies. It can improve one's leverage with the next set of enemies.

Party: let's make a deal; we'd like to offer you this for that -
NPC: I am not interested in making a deal with you. Buzz off.
Party; You remember that (refer to that bloody battle) group? They didn't want to make a deal.
NPC: Hmm, what shape shall we make the table as we work out a deal to our mutual benefit.

It will depend somewhat on the campaign. I don't think I can recall ever making a deal with gnolls, but over the years the various parties I've been in have made deals with kobolds, pirates, orcs, giants, goblins, elves(wild elves who were xenophobes), bandit princes, and even dragons.

[Skylivedk]

Not trying to be antagonistic, just puzzled by how different our parameters are when it comes to defining a good party.

All enemies can be defeated with damage. Thus, damage is the foremost consideration. The next consideration ultimately boils down ensuring you are able to keep doing damage. That may be through control, defensive buffing, healing, self defenses, counterspells, etc. Thirdly, there are the abilities that fundamentally change the nature of encounters, such as scouting, surprise, avoiding surprise, disarming/bypassing traps, bypassing encounters entirely, using social skills/suggestion/charm/dominate to gain allies while taking away enemies, etc.

What I find is that abilities in that third category are often the highest impact but least reliable - at some point you'll fail stealth checks, same with perception checks, same with suggestion and saving throws, even persuasion tends to have DM limitations put on it for what it can ultimately accomplish, teleportation may get you past the encounter now, but there's always the possibility that whatever you teleported from may seek you out later.

The above quote shows a pretty clear focus, IMO, on damage. I didn't see this as having a connection to the Exploration and Social pillar due to the phrasing pointing pretty much straight to encounters again. Again, I'm curious. It's very far from my play experience. Using the two other pillars actively often make damage dealt by characters a side note in many of my campaigns. Even in combat I wouldn't use damage as the most important. Getting through the encounter with the biggest win is the most important. That can just as easily mean cutting the bridge or destroying the building. You know the solutions you don't have in most computer games and in Gloomhaven etc.

Agreed on the skill system being kinda unreliable. Luckily, there's ways around that and I do think optimised parties should include some of those ways.

I tend to agree, with the caveat that sometimes, one does need to leave the field of battle littered with the corpses of one's enemies. It can improve one's leverage with the next set of enemies.

Party: let's make a deal; we'd like to offer you this for that -
NPC: I am not interested in making a deal with you. Buzz off.
Party; You remember that (refer to that bloody battle) group? They didn't want to make a deal.
NPC: Hmm, what shape shall we make the table as we work out a deal to our mutual benefit.

It will depend somewhat on the campaign. I don't think I can recall ever making a deal with gnolls, but over the years the various parties I've been in have made deals with kobolds, pirates, orcs, giants, goblins, elves(wild elves who were xenophobes), bandit princes, and even dragons.

Just wanted to include this, because
1) agreed.
2) I don't want to come across as not being a fan of my favourite debuff ;-)
3) I think there's a separate underlying point here: if you take the violent route, make sure that either nobody knows or the bards are singing about it 😁 deterrence is a way of winning combats without rolling initiative.

[KorvinStarmast]

Agreed on the skill system being kinda unreliable. Luckily, there's ways around that and I do think optimised parties should include some of those ways.

Yes. Between Guidance, Enhance Abilities, Expertise, and Bardic Inspiration, the chances for succcess can be boosted. (I'll leave Reliable Talent for another sub discussion since one first needs to define how long the campaign will last).

Just wanted to include this, because
1) agreed.
2) I don't want to come across as not being a fan of my favourite debuff ;-)
3) I think there's a separate underlying point here: if you take the violent route, make sure that either nobody knows or the bards are singing about it 😁 deterrence is a way of winning combats without rolling initiative.

In re (3): this takes us back to my quote from Sun Tzu.

[Frogreaver]

Good Thread.

Party Optimization - It depends on the campaign you are going to be in. It depends what level it goes to. It depends on the pace of encounters and short rests throughout the day. It depends on the types of encounters you will be facing. It depends on your DM. It depends on what goal you are optimizing toward.

That said I think there are some principles that will help guide us. All enemies can be defeated with damage. Thus, damage is the foremost consideration. The next consideration ultimately boils down ensuring you are able to keep doing damage. That may be through control, defensive buffing, healing, self defenses, counterspells, etc. Thirdly, there are the abilities that fundamentally change the nature of encounters, such as scouting, surprise, avoiding surprise, disarming/bypassing traps, bypassing encounters entirely, using social skills/suggestion/charm/dominate to gain allies while taking away enemies, etc.

What I find is that abilities in that third category are often the highest impact but least reliable - at some point you'll fail stealth checks, same with perception checks, same with suggestion and saving throws, even persuasion tends to have DM limitations put on it for what it can ultimately accomplish, teleportation may get you past the encounter now, but there's always the possibility that whatever you teleported from may seek you out later.

Bards, Wizards, Druids tend to be fairly good at this role. Rogues and Clerics aren't terrible at it.

Control abilities are also not the most reliable, but are some of the most important abilities for turning an otherwise hard encounter easy. I would place opportunity attacks under this section as well, though they are a poor form of control typically.

Bards, Wizards, Sorcerers and even Warlocks fair well here. Clerics have a small bit of control.

Defensive abilities are primary about party defense/healing - although you want to have enough durability to take a few hits. This is also where the role of a tank can come in (although tanking is highly dependent on how the DM runs enemies). These abilities help you keep the worst case from happening and tend to be fairly reliable.

Clerics, Bards, Druids, Wizards, Sorcerers are all great here. Paladins deserve honorable mention.

Finally we get to Damage. There's lots of ways to do good damage. Generally SS or GWM builds work the best. However, druids can do good damage, clerics do well to if they burn slots on spirit guardians + spiritual weapon in a fight. Sorcerers can twin buff allies to add alot of extra damage. Animate objects does good damage but that's moving.

If you cover these bases well, you'll have a fairly optimized party. I wouldn't go for a solely gimmick party, because at some point the gimmick will fail.

I've bolded all the content in my post not dealing with damage...

Not trying to be antagonistic, just puzzled by how different our parameters are when it comes to defining a good party.


The above quote shows a pretty clear focus, IMO, on damage.

You inevitably will face encounters where you have to fight. The only way to win those encounters is to deal damage. So yes, damage is the cornerstone of an optimized party because it's inevitable that you will have to result to it at some point.

But that doesn't mean only damage, which is how you seem to be taking it.

I didn't see this as having a connection to the Exploration and Social pillar due to the phrasing pointing pretty much straight to encounters again.

I specifically call out a number of non-combat abilities and your accusation is that I'm ignoring exploration and social. I don't understand how you can even entertain for a moment that your extreme caricature of my position in any way realistically reflects it.

Again, I'm curious. It's very far from my play experience. Using the two other pillars actively often make damage dealt by characters a side note in many of my campaigns. Even in combat I wouldn't use damage as the most important. Getting through the encounter with the biggest win is the most important. That can just as easily mean cutting the bridge or destroying the building. You know the solutions you don't have in most computer games and in Gloomhaven etc.

Yes, cut the bridge and destroy the building when those opportunities arise. But inevitably there will not always be ropes to cut, and sometimes the building just won't catch on fire.

Agreed on the skill system being kinda unreliable. Luckily, there's ways around that and I do think optimised parties should include some of those ways.

Isn't that exactly what I said?

"If you cover these bases well, you'll have a fairly optimized party."

[Skylivedk]

You inevitably will face encounters where you have to fight. The only way to win those encounters is to deal damage. So yes, damage is the cornerstone of an optimized party because it's inevitable that you will have to result to it at some point.

No, it isn't. Not even a fight will always require damage, and sometimes dealing damage is your worst option.

But that doesn't mean only damage, which is how you seem to be taking it.

Reading your post you say it's the most important. I don't agree.

I specifically call out a number of non-combat abilities and your accusation is that I'm ignoring exploration and social. I don't understand how you can even entertain for a moment that your extreme caricature of my position in any way realistically reflects it.

I think I made that clear: those instances referred back to combat encounters rather than world building.

Isn't that exactly what I said?

"If you cover these bases well, you'll have a fairly optimized party."

I don't disagree with everything you write ;-) I disagree with how you prioritise the capabilities of a party and how you link the capabilities. To be more exact, I disagree with my perception of what you wrote which was linking all pillars back to the Combat pillar. If that's not your intention, my bad. It was how it came across to me. Reading #damageisnr1 probably influenced my perspective on the rest of your post.

I still disagree with damage in a vacuum being the best thing to optimise for. I'd much rather have a big toolbox with a lot of excellent tools/solutions in it than optimise for damage first and tack on the capabilities after.

[MaxWilson]

I don't disagree with everything you write ;-) I disagree with how you prioritise the capabilities of a party and how you link the capabilities. To be more exact, I disagree with my perception of what you wrote which was linking all pillars back to the Combat pillar. If that's not your intention, my bad. It was how it came across to me. Reading #damageisnr1 probably influenced my perspective on the rest of your post.

I still disagree with damage in a vacuum being the best thing to optimise for. I'd much rather have a big toolbox with a lot of excellent tools/solutions in it than optimise for damage first and tack on the capabilities after.

I can imagine an interesting playstyle in which players try very, very hard to kill zero or at most one creature per adventure, and where they are able to succeed despite having a collective party DPR of only about 20 HP per round at level 11. E.g. Cloud of Daggers might be the only damage-dealing method in the party, and the goal might be to force the enemy to surrender before it's reduced to zero HP. That campaign might look a lot more like a police procedural than a dungeon crawl.

I'm not sure that 5E (as opposed to e.g. DramaSystem or FATE or GUMSHOE) is the best system for such a campaign but it's interesting to think about. If your group is either very comfortable with freeform roleplaying, or very good at inventing new game structures/stealing them from other systems, it could work. What 5E brings to the table here is additional "technologies" for solving procedural problems in play, e.g. if you are in a scene with e.g. an uncooperative witness, 5E gives you Detect Thoughts or Suggestion or Zone of Truth or Disguise Self to help you make them more cooperative. What 5E lacks is the basic, non-technological underlying part: 5E has no real guidance for you on what players can or should do if they want to get information out of the witness the way Sherlock Holmes would do it, without falling back to technologies like Detect Thoughts. 5E magic/special abilities are kind of a blunt instrument.

It could work, or it could fail disastrously (by which I mean, turn out to be confusing and boring for the players). I can imagine it going either way for a given group, but it's interesting to think about.

[Eldariel]

Just a quick response WRT damage: Damage is the thing any PC can natively do. It doesn't take much to get enough damage to kill limping enemies, especially in a party with 3+ characters (maybe one cantrip pick or carrying a weapon depending on the class).

So the math turns into whether the amount of reduction in damage/effects dealt by a hostile you try to kill is worth enough for the resources you needed to put into achieving the increase in damage. Which is far less forgiving a metric than just the "maximum damage"-metric and generally seems to point towards "benefits of damage increase are insufficient compared to the resources invested". I posit that the best damage increasing investments are those that also provide you with an increase in versatility or that come as a byproduct of investments that also provide you with lateral angles of solving problems (generally caster levels).

[Frogreaver]

Just a quick response WRT damage: Damage is the thing any PC can natively do. It doesn't take much to get enough damage to kill limping enemies, especially in a party with 3+ characters (maybe one cantrip pick or carrying a weapon depending on the class).

Not sure the relevance of looking at killing limping enemies...

So the math turns into whether the amount of reduction in damage/effects dealt by a hostile you try to kill is worth enough for the resources you needed to put into achieving the increase in damage.

I don't think the game can necessarily be boiled down solely to a probabilistic expected value problem. There's just to much volatility and too many unknowns to be able to reasonably pre-assess the expected value of an ability/tactic/etc over a whole campaign.

Instead optimization is primarily about ensuring the party can bypass bypass/setup many encounters, have enough party defense/healing to be able to extend the fight when things go bad, and have enough damage that the fight is ended before the enemies can cause to much trouble. Optimization becomes about layering on hefty amounts of each of those so that your party performs well and has a contingency when your current tactic isn't working in your favor.

Which is far less forgiving a metric than just the "maximum damage"-metric and generally seems to point towards "benefits of damage increase are insufficient compared to the resources invested".

No one is suggesting you optimize for maximum damage. Only that you must also account for damage.

I posit that the best damage increasing investments are those that also provide you with an increase in versatility or that come as a byproduct of investments that also provide you with lateral angles of solving problems (generally caster levels).

I posit that when dealing with a party situation that isn't necessarily the case. Having one character of the party focused primarily on damage is going to be more optimal in many situations than ensuring a party full of full casters with none focused on damage. If the damage character is doing adaquate damage then the casters will use less resources on offense and defense - freeing up more spells for other things.

[Eldariel]

Not sure the relevance of looking at killing limping enemies...

Chances are you are going to levy some sort of sanctions on enemies of any relevance anyways especially if the party is not especially high damage (because what else is a non-damage specced character going to be specced for?) so enemies will most likely be either disabled or annihilated and to that end, whether you maim and then kill the maimed enemies slightly slower or just nuke the enemies, there is probably rather low variance between the average expected result

I don't think the game can necessarily be boiled down solely to a probabilistic expected value problem. There's just to much volatility and too many unknowns to be able to reasonably pre-assess the expected value of an ability/tactic/etc over a whole campaign.

Instead optimization is primarily about ensuring the party can bypass bypass/setup many encounters, have enough party defense/healing to be able to extend the fight when things go bad, and have enough damage that the fight is ended before the enemies can cause to much trouble. Optimization becomes about layering on hefty amounts of each of those so that your party performs well and has a contingency when your current tactic isn't working in your favor.

Probabilities are all the game is. Everything in the game boils down to probabilities; the probability of succeeding in any given task, the probability of having any given resource available, the probability of having to use your contingencies, the probability of said contingencies working, etc. It's just a matter of describing the problem.

No one is suggesting you optimize for maximum damage. Only that you must also account for damage.

I posit that minimum investment damage is so sufficient that it's not necessarily worth it to specialise any one character for damage beyond what individual spells known or similar abilities might provide.

I posit that when dealing with a party situation that isn't necessarily the case. Having one character of the party focused primarily on damage is going to be more optimal in many situations than ensuring a party full of full casters with none focused on damage. If the damage character is doing adaquate damage then the casters will use less resources on offense and defense - freeing up more spells for other things.

OTOH if the party has no dedicated damage characters, the party has more and more varied CC effects by necessity since the same resources are put elsewhere. Actually, this is an important point we should solve: at which point is it worthwhile to dedicate a character to damage over a character that brings a reasonable mix of CC and damage. Things like many Sorcerer and Wizard builds, warrior Bards/Wizards/Clerics/Druids, many Warlocks, etc. can provide decent utility and CC while also having extremely good nova and fairly good at-will damage. It needs to be immense damage indeed from the non-utility character or an immense amount of redundancy for it to do enough more damage to make up for bringing nothing else to the table.

Simple Agonizing Blast Warlock for instance is quite competitive with most non-casters for damage while also having (admittedly a low amount) of spell slots, invocations and then the higher level spells eventually. Similarly, Wizard/Bard/Warlock have access to the Extra Attack chassis plus magic for advantage and even on-hits/extra attacks for pretty solid Sharpshooter builds (which admittedly sink a feat or even two or three into damage if you also go XBE/Elven Accuracy), which doesn't do poorly for damage and yet also brings a reasonable (if not quite full in the case of the Bards and Warlocks) set of utility and control magic to the table catering to both needs. Mostly these options give up subclass options, which is generally a secondary part of a caster chassis compared to the actual spell list and spell slots. Thus the opportunity cost for bringing a dedicated warrior is pretty big: it has to be enough better than these builds for the party as a whole to make up for not bringing other stuff to the table.

[Frogreaver]

Chances are you are going to levy some sort of sanctions on enemies of any relevance anyways especially if the party is not especially high damage (because what else is a non-damage specced character going to be specced for?)

Maybe Out of combat - ie stealth, persuasion, perception, teleportation, diviniation, etc

so enemies will most likely be either disabled or annihilated and to that end, whether you maim and then kill the maimed enemies slightly slower or just nuke the enemies, there is probably rather low variance between the average expected result

Actually, control abilities are very high variance and fairly resource expensive. So while they may on average be very effective, there will be plenty of times when they outright miss or hit very few enemies.

Probabilities are all the game is. Everything in the game boils down to probabilities; the probability of succeeding in any given task, the probability of having any given resource available, the probability of having to use your contingencies, the probability of said contingencies working, etc. It's just a matter of describing the problem.

Expected value is a certain type of very long term probability problem. Expected value absent variance/volatility for a more short term situation is an extremely bad measure.

I posit that minimum investment damage is so sufficient that it's not necessarily worth it to specialise any one character for damage beyond what individual spells known or similar abilities might provide.

Maybe that's a difference in definitions. A caster can be about damage just as much as a Fighter. A sorcerer taking all the blasty spells for example.

OTOH if the party has no dedicated damage characters, the party has more and more varied CC effects by necessity since the same resources are put elsewhere. Actually, this is an important point we should solve: at which point is it worthwhile to dedicate a character to damage over a character that brings a reasonable mix of CC and damage.

Exactly. But remember, CC has very diminishing returns. Against the same group of enemies, the first hypnotic pattern will Control about half the enemies. The 2nd about 25%, etc. The major benefit of more and varied CC is that you can target weaker saves. But then there's always the inherent risk of losing concentration (or a requirement for major inventment into concentration saves).

Things like many Sorcerer and Wizard builds, warrior Bards/Wizards/Clerics/Druids, many Warlocks, etc. can provide decent utility and CC while also having extremely good nova and fairly good at-will damage. It needs to be immense damage indeed from the non-utility character or an immense amount of redundancy for it to do enough more damage to make up for bringing nothing else to the table.

I think it would help if you defined immense. That's a rather big sounding word that might just mean 50% more party damage or double party damage.

Simple Agonizing Blast Warlock for instance is quite competitive with most non-casters for damage while also having (admittedly a low amount) of spell slots, invocations and then the higher level spells eventually.

Those Warlocks are pretty terrible at damage and that's even with hex.

Similarly, Wizard/Bard/Warlock have access to the Extra Attack chassis plus magic for advantage and even on-hits/extra attacks for pretty solid Sharpshooter builds (which admittedly sink a feat or even two or three into damage if you also go XBE/Elven Accuracy), which doesn't do poorly for damage and yet also brings a reasonable (if not quite full in the case of the Bards and Warlocks) set of utility and control magic to the table catering to both needs.

Two things:
1. You seem to be under the impression that a full caster cannot be a damage focused build? It can. But then there
s a different discussion - how beneficial is high at will damage vs high damage that relies on expendable resources and concentration.

2. Many of those builds you cite aren't particularly good at damage.

Mostly these options give up subclass options, which is generally a secondary part of a caster chassis compared to the actual spell list and spell slots. Thus the opportunity cost for bringing a dedicated warrior is pretty big: it has to be enough better than these builds for the party as a whole to make up for not bringing other stuff to the table.

I've been talking about a primary damage dealer. You seem to be taking that to mean it can't be a caster. Strange assumption on your part. That said, I think for most of the game we can make a great case for a non-caster damage dealer. Level 1 and 2 spell slots just don't add enough damage. And you don't get enough level 3+ slots to really push hard into damage till almost tier 3.

[KorvinStarmast]

If I could try to steer this discussion back somewhat to the framework we are going after, which is to first establish what category of {challenge?} or {play style} or {team concept} a party is being optimized for and once those are mostly agreed, anyone can then take "4 PC YOLO" or "3 PC disaster prevention" or whatever models and propose a party optimized for that.

I'd like to get away from tangents like DPR au outrnace, since we had a previous post that I think most of us agreed on that is "optimization for all three pillars of the game" is a basic assumption or point of departure.

Do we agree on that?

The reason I say this is that a significant volume of discourse over the past six or seven years on D&D 5e DPR optimization has been written/posted, albeit mostly focused on single PCs.

What we are offering is something different from that.

[MaxWilson]

If I could try to steer this discussion back somewhat to the framework we are going after, which is to first establish what category of {challenge?} or {play style} or {team concept} a party is being optimized for and once those are mostly agreed, anyone can then take "4 PC YOLO" or "3 PC disaster prevention" or whatever models and propose a party optimized for that.

I'd like to get away from tangents like DPR au outrnace, since we had a previous post that I think most of us agreed on that is "optimization for all three pillars of the game" is a basic assumption or point of departure.

Do we agree on that?

The reason I say this is that a significant volume of discourse over the past six or seven years on D&D 5e DPR optimization has been written/posted, albeit mostly focused on single PCs.

What we are offering is something different from that.

I'm okay with that in principle but am not ready to propose any parties yet. There's a piece that I am still working on first.

[KorvinStarmast]

I'm okay with that in principle but am not ready to propose any parties yet. There's a piece that I am still working on first.

Thank you, Max. When I get some time {there may be some tomorrow} I am going to review all of the posts and try to create a table that captures the categories as best as I understand them.

[Frogreaver]

If I could try to steer this discussion back somewhat to the framework we are going after, which is to first establish what category of {challenge?} or {play style} or {team concept} a party is being optimized for and once those are mostly agreed, anyone can then take "4 PC YOLO" or "3 PC disaster prevention" or whatever models and propose a party optimized for that.

I'd like to get away from tangents like DPR au outrnace, since we had a previous post that I think most of us agreed on that is "optimization for all three pillars of the game" is a basic assumption or point of departure.

Do we agree on that?

The reason I say this is that a significant volume of discourse over the past six or seven years on D&D 5e DPR optimization has been written/posted, albeit mostly focused on single PCs.

What we are offering is something different from that.

I think that’s doing a bit of a disservice to the reason damage was brought up in the first place. Having a high damage character can save party resources, hp, spell slots, etc - and thus be potentially better in many circumstances than a low damage alternative character.

[MaxWilson]

I think that’s doing a bit of a disservice to the reason damage was brought up in the first place. Having a high damage character can save party resources, hp, spell slots, etc - and thus be potentially better in many circumstances than a low damage alternative character.

You may take that into consideration when you propose a party for a specific set of constraints/playstyle.

[Frogreaver]

You may take that into consideration when you propose a party for a specific set of constraints/playstyle.

For a general adventuring party playing level 1-20 I would suggest something like this is about as optimized as you can get. Maybe splash in a bit of multiclassing.

SS+CE BattleMaster Fighter (Damage)
Subclass Undecided Cleric (Tanking/Defense/Healing/Out of Combat Utility)
Divination Wizard (Control/Out of Combat Utility)
Lore Bard (Social Pillar/Backs up other characters in their respective roles as needed)


Fighter helps ensures casters don't need to use many resources except in hard battles. Especially important in tier 1 and tier 2 when spell resources are less powerful/run out quicker, but still useful in tier 3 and 4.

Cleric tanks as needed. Early game the high AC and healing should be plenty enough for that. Tier 2+ you get Spirit Guardians to round that out. Clerics can also make ranged attacks with cantrips when not trying to tank.

Wizards, it's a wizard which I think is enough said.

Lore Bard, the social king and back up healer/controller/damager/additional out of combat utility.

[Benny89]

I always play min. 5 players team so for me one of the optimal setups would be:

1. On demand Nova + tank + support + hard to kill + social proficiences - Paladin (or Pala-builds like Sorcadin etc.)
2. Something tanky with healing + good damge + ressurection if things go south + good in front line - I prefer here strong cleric like Arcana for example. But Shep Druid is also great if he can choose his summons as he should. Some nice Divine Soul Sorc build for Twin Haste/Heal/Healing Word/Regeneration etc. can also be great if we take dip in Life Cleric/Hexblade/Paladin for example for some armor.
3. Pure DPS - this can be anything from PAM GWM Hexblade/Battlemaster/Zealot barbarian to Xbow Master/SS BM or EA Samurai. A lot of options. This is THE optimal target for spell buffs like Haste, Holy Weapon etc.
4. Wizard- but I prefer wizard that is hard to kill and one that will take care of "anti-magic" defenses of whole party - hence I prefer Abjuration Wizard, but as we know all wizards are great in general. Divination is ok, but from my experience his "thing" rarely has its moment so I prefer some 21 AC Wizard which is almost impossible to kill while still does all what "God Wizard" can do + boosted resistances andd saves. But some powerbuilds like 1 Hexblade/X Evocation Wizard would also be super strong here for pure punch.
5. Skill monkey/utility/flex - Bard or Arcane Trickster would be best for skill monkey + utility for party and expertise is some roles that party lacks - it would be stealth for example.

Of course above is just one of many setups. There is no "best setup" as it all depends on campaign and DM and style table is playing.