Party optimisation philosophy

[5eNeedsDarksun]

Maybe I'm misinterpreting some of these posts, and if I am then please say so, as my intent isn't to be smashing a strawman.
Having said that, it looks like having a decent Stealth stat for all party members is being labeled as a bit of a gimmick by some posts. I really don't think that is the case. Reality is almost every character I've played with, with the exception of Strength based melees and a minority of Clerics have at least a 14 Dex and can easily achieve decent Stealth by wearing a Breastplate and avoiding the 1/2 plate when necessary. Further, it is viable in 5e to make a Dex based warrior who is comparable to a Strength based one in Melee and better at range. Further, thematically I don't think it is a stretch to imagine an adventuring group as 'sneaky' depending on the campaign.
I've played in more than one campaign where the lone heavily armored warrior was asked routinely to 'wait outside' for a round while everyone else used stealth to move in and get a free wack at the baddies, meaning on round 1 they were not able to benefit the party. Despite the benefits of Str based characters in Heavy Armor, this drawback is big enough to not include one in my optimized group.

[Eldariel]

Maybe I'm misinterpreting some of these posts, and if I am then please say so, as my intent isn't to be smashing a strawman.
Having said that, it looks like having a decent Stealth stat for all party members is being labeled as a bit of a gimmick by some posts. I really don't think that is the case. Reality is almost every character I've played with, with the exception of Strength based melees and a minority of Clerics have at least a 14 Dex and can easily achieve decent Stealth by wearing a Breastplate and avoiding the 1/2 plate when necessary. Further, it is viable in 5e to make a Dex based warrior who is comparable to a Strength based one in Melee and better at range. Further, thematically I don't think it is a stretch to imagine an adventuring group as 'sneaky' depending on the campaign.
I've played in more than one campaign where the lone heavily armored warrior was asked routinely to 'wait outside' for a round while everyone else used stealth to move in and get a free wack at the baddies, meaning on round 1 they were not able to benefit the party. Despite the benefits of Str based characters in Heavy Armor, this drawback is big enough to not include one in my optimized group.

I believe the term "gimmick" is being used only in the sense of hyperfocusing on it (to the point of everyone getting Expertise, maxing Dex, etc.) - I'm certainly for having everyone have decent stealth abilities since the opportunity cost is miniscule and the potential benefit is massive but I don't think it's worth delaying spell access for the Expertise bonuses for instance since while it's true that this'll somewhat include the amount of enemies you can negate with it, the more relevant obstacle is having environments that don't cater to stealth or missions where you need to engage an enemy, and thus I believe it's more useful to use build resources towards other options while having stealth as a good general option.

[MaxWilson]

Further, it is viable in 5e to make a Dex based warrior who is comparable to a Strength based one in Melee and better at range.

I don't think this is actually true: Str is much, much better in melee. If you were for example fighting in a campaign where ranged weapons were considered both dishonorable and illegal, and most of your fights were going to take place in public (e.g. gladiatorial combat or duels), a high-level Str-based fighter would beat a Dex fighter almost every time, partly because of better AC (although lower initiative partly offsets that) but mostly through the sheer strength of Athletics.

Grapple the enemy, disarm them and/knock them prone, and beat on them with advantage until they die. Dex fighters have that option too if they go Prodigy (Athletics) but Str fighters are significantly better.

But that's only if you remove ranged combat from the equation somehow.

[Nagog]

So for clarity: Are you optimizing specifically for combat, or general use?

[Eldariel]

So for clarity: Are you optimizing specifically for combat, or general use?

Since we're trying to do useful optimisation here, I think "general use" has to be the answer. Though I think a combat-based listing would have a place as well. But hmm, the framework for the framework is getting pretty cumbruous already so this seems like it'll be quite the project.

[MaxWilson]

This is probably an issue with how I used the terminology: a campaign arc is equivalent to a single adventure in my parlance. An arc is a part of a campaign that has a given goal. The campaign is a set of arcs intervowen in some way or another. As an example, the campaign arcs of Lost Mine of Phandelver would be getting to Phandelver, rescuing Gundred, finding and holding Wave Echo Caves. Essentially three separate arcs though the players can engage in optional arcs such as dealing with the Old Owl Well Orcs, getting knowledge of the spellbook location for Sister Garaele, helping out Reidoth, or dealing with Hamun Kost. Each of those would be a separate arc (except Orcs and Kost, since they're basically the same arc but just from two different angles).

That doesn't really change anything; whether the arc is player- or DM-imposed is trivial. In this case an arc would be precisely something like "take over a faction" or "achieve a personal goal" or "defend gains against an usurper" or any of the sort.

I do think there are plenty of rules for this. Nondetection, Mind Blank, Scrying, Contact Other Plane, Divination, etc. are a part of the rules set as always and in a way, every class actually gets to participate a bit since Ritual Caster feat exists. Whether it's the most fun way to play the system? Ehh, depends on whom you ask. I feel like this question is irrelevant. I definitely think a PC enemy with 9th level spells could and should try just something like what's being suggested here. The PCs have weeks to enact their own plan in the meanwhile. They have divinations to figure out that somebody is moving against them and they have teleportation and the like to take the fight to them or minionmancy to raise their own army for the encounter or any such options. I do think plans like these are precisely what makes e.g. high level Liches fearsome opponents: not their combat potential but their out-of-combat potential.

Actually, in my own LMoP game right now the party is at a stand-off with Glasstaff, unable to leave Phandalin since they know he'll take it over if they leave but as he was driven out he's not willing to give them a fight unless he has overtly favourable terms of engagement, and they have few means to force an engagement with an enemy on the run avoiding them. In short, they're fighting an information war and it's quite different bringing the party ranger in particular to bear in rather surprising ways (what's actually exposed Glasstaff is that he's animating the dead from the engagement at the hideout and the Rev. Ranger Gloomstalker has Favored Enemy: Undead so Primal Awareness actually gives him a compass straight to Glasstaff without him suspecting that it is a known quality; the party hasn't fully realised this yet which makes for a cool dynamic of both parties playing in k

I don't think you can ever solve D&D. RAW D&D every ability players have is also available to NPCs so there's nothing CaW PCs can do that can't be done better by their opposition against them. Thus by definition it is impossible to ever reach a point where the party is more powerful than the challenge the world provides. Sandbox is indeed the best way to autogenerate appropriate challenge but it's of course quite trivial to activate opposition in existing modules to match this as well. There are, after all, powerful NPC demi-deity level enemies and general big movers and powerful spellcasters in pretty much all of them so they're fully capable of engaging in 5d Chess with the party.

However, this requires the ability from the party to play the same game as well, because otherwise they will simply not have the ability to pick their fights and they'll soon be overpowered by completely unfair encounters as the enemy hits them in their weak spot when they're already spent. In short, I'd rather say that a powerful party is just a permission for the DM to actually play the world and the opposition naturally without having to dumb movers down to party level.

*snip*

I'm pretty sure we should be able to include this in some broader umbrella category of game style considerations. I think this is too niche to be a consideration as a separate consideration, but "Pacing" in general is probably a worthwhile topic to cover. This would also cover things like whether resting is at player prerogative or DM purview, how do encounters occur, etc. Should probably be included under "Campaign style" - I feel like this might be a subset of "sandbox vs. plot-driven" (sandbox tends towards player-driven resting, while plot-driven tends towards doomsday clocks and DM-driven resting, but it's also influenced by DM style).


I'll have to think about these a bit: I'm pretty sure these should all be coverable with just two-three major DM/campaign style questions instead of a set of questions on minutiae.

Okay, thanks for clarifying "campaign" arc. That lets us focus in on the CAW thing, and the rules (or lack thereof) in RAW for structuring it.

My suspicion is that you can't have all three of these at the same time:

(1) Require RAW as a common denominator for discussion,
(2) Meaningful analysis of optimal strategies/builds with maximum narrative power.
(3) Combat As War playstyle. [Together with #2 this implies: CAW on both sides, vs. powerful enemies who are playing to win.]

I feel that you can do meaningful RAW-driven analysis if you restrict yourself to essentially Combat As Sport scenarios, like "fighting a bunch of MM monsters in a dungeon crawl," or limited Combat As War scenarios like "fighting waaaaay too many monsters before they can massacre your hometown," but there is no meaningful way to analyze conflict against "powerful NPC demi-deity level enemies and general big movers... engaging in 5d Chess" in 5E because 5E's ruleset is essentially focused on squad-level tactical combat, and everything else is... just kind of handwaved, so the determining factor is not "what is the party build" but rather "how is the DM running things?" which in no way is a RAW-based discussion.

What's more important, the ability to cast Nondetection or the ability to recognize when you're being tailed on the street? RAW can't tell you. RAW can't even tell you which of them is more difficult, let alone which of them is more common. Does DPR even matter AT ALL in a CAW game, or are things just dealt with by hiring mercs with poisoned weapons, blackmailing people, and poisoning key NPCs to move more-pliable NPCs into position? Is there a meaningful difference between casting Scrying and blackmailing a wizard into casting Scrying? Does blackmail involve die rolling and Intimidation checks or is it simply a matter of gaining leverage over something precious to the wizard, like keeping his granddaughter out of prison?

You could analyze CAW within a specific context, with specific rules and DMing style, but each set of rules and style guidelines is essentially its own game based on the 5E engine. Is it possible to meaningfully discuss a game whose rules haven't even been written yet at the point where you're discussing it?

TL;DR in any CAW game, the optimal approach will be based on something not in RAW because RAW has very little to say about the game aspects CAW cares about. (It has a few things to say about what is definitely possible with spells, but nothing to say about whether those things like spying are also possible without spells.)

[Eldariel]

Okay, thanks for clarifying "campaign" arc. That lets us focus in on the CAW thing, and the rules (or lack thereof) in RAW for structuring it.

My suspicion is that you can't have all three of these at the same time:

(1) Require RAW as a common denominator for discussion,
(2) Meaningful analysis of optimal strategies/builds with maximum narrative power.
(3) Combat As War playstyle. [Together with #2 this implies: CAW on both sides, vs. powerful enemies who are playing to win.]

I feel that you can do meaningful RAW-driven analysis if you restrict yourself to essentially Combat As Sport scenarios, like "fighting a bunch of MM monsters in a dungeon crawl," or limited Combat As War scenarios like "fighting waaaaay too many monsters before they can massacre your hometown," but there is no meaningful way to analyze conflict against "powerful NPC demi-deity level enemies and general big movers... engaging in 5d Chess" in 5E because 5E's ruleset is essentially focused on squad-level tactical combat, and everything else is... just kind of handwaved, so the determining factor is not "what is the party build" but rather "how is the DM running things?" which in no way is a RAW-based discussion.

What's more important, the ability to cast Nondetection or the ability to recognize when you're being tailed on the street? RAW can't tell you. RAW can't even tell you which of them is more difficult, let alone which of them is more common. Does DPR even matter AT ALL in a CAW game, or are things just dealt with by hiring mercs with poisoned weapons, blackmailing people, and poisoning key NPCs to move more-pliable NPCs into position? Is there a meaningful difference between casting Scrying and blackmailing a wizard into casting Scrying? Does blackmail involve die rolling and Intimidation checks or is it simply a matter of gaining leverage over something precious to the wizard, like keeping his granddaughter out of prison?

You could analyze CAW within a specific context, with specific rules and DMing style, but each set of rules and style guidelines is essentially its own game based on the 5E engine. Is it possible to meaningfully discuss a game whose rules haven't even been written yet at the point where you're discussing it?

TL;DR in any CAW game, the optimal approach will be based on something not in RAW because RAW has very little to say about the game aspects CAW cares about. (It has a few things to say about what is definitely possible with spells, but nothing to say about whether those things like spying are also possible without spells.)

Hmm, I'm not entirely in agreement. Yes, of course much is beyond what you have available. But we can lay a groundwork:
1) You need resources to e.g. blackmail people. You can't just tell someone to go do something for you, you need to get something that they hold in value.
2) Hiring NPC mercs...requires that there are mercs of sufficient power available. Similarly, it requires that the enemy hasn't hired them for you. We can also surmise it heavily depends on your social skills, related magic, etc. as well as what you are able to offer to the mercs.
3) We can probably assume that if engaging a big league event (say, OotA; demon lords on the material plane, hooray!) there probably aren't mercs able/willing to help you. You need help from other movers in the world, who probably aren't that easy to gain under your power (ergo power is something that enables you to gain people under your power).
4) There's a very meaningful difference between Scrying and blackmailing somebody to Scrying. Scrying is something you can do yourself and at will. Blackmailing is a complex, multistep process that produces enmity of the blackmailee, and might be fleeting as whatever you're blackmailing them with might come out of your power making them free to move against you. Procuring the leverage in the first place is far from trivial. First and foremost though, if you blackmail somebody you need to ensure they cannot deadify you before you can pull on your leverage.
5) DPR is...always useful. If you can kill somebody before they can act and they know it, they're more likely to do what you want. However, if that's how you procure power, you will create numerous enemies making your position always precarious.

So what this comes back to: All of CaW options still depend on the party's resources, i.e. how good the party is in e.g. making the merc gang rather take their offer than the opponent's and what's the failcase (since mercs are never all that trustworthy if we go by their modus operandi of "work for the highest bidder"). Things like social power and monetary power (the ability to produce money too) are of course highlighted.

Further, no matter how carefully you play in the shadows, the world has a lot of powers in the shadows and you probably cannot avoid crossing enemies if you want to accomplish anything. In that sense, you'll always be under threat and thus your personal power, defense and survivability (especially the tools therein; how safe a place can you rest in, how quickly and efficiently you can move, how unpredictable you can remain, etc.) is an advantage pretty much regardless of how the game is run since if we make the assumption that some powers in the world probably want you dead, you need to always stay one step ahead.

All of this does come back to the information war, which highlights a specific subset of magic as well as certain skills, but those certainly do have mechanical frameworks in the system (I can't call skill system "rules" since they're so out there but they are what we have). In general, whether DM sets the DC at 15 or 20 to e.g. convince the Black Rose to reveal their previous employer with a generous offer of gold is not that relevant; higher bonus is better anyways. Similarly, if you can just get to the boss (finding the boss of a shadowy merc organisation is, again, of course easier said than done but there are plenty of ways characters can engage in such a task) and Suggest that they tell it to you in exchange for a fair amount of money, well, that's probably a fine plan too. Or mayhap you create a facade of a conversation while somebody is covertly trying to read the important information out of your subject's mind.

Either way, I don't believe CaW really meaningfully changes how you approach the characters/rolls themselves. It merely just highlights the game outside the Encounter itself; which, true, has less rigid guidelines but guidelines nonetheless on what you can and can't do or what kinds of options are open to given character classes.

[5eNeedsDarksun]

I don't think this is actually true: Str is much, much better in melee. If you were for example fighting in a campaign where ranged weapons were considered both dishonorable and illegal, and most of your fights were going to take place in public (e.g. gladiatorial combat or duels), a high-level Str-based fighter would beat a Dex fighter almost every time, partly because of better AC (although lower initiative partly offsets that) but mostly through the sheer strength of Athletics.

Grapple the enemy, disarm them and/knock them prone, and beat on them with advantage until they die. Dex fighters have that option too if they go Prodigy (Athletics) but Str fighters are significantly better.

But that's only if you remove ranged combat from the equation somehow.

So I'll agree to the extent that 1v1 in close quarters against a Dex based character or monster the Strength based character will likely win. That's not really D&D though. You are 4v? in varied conditions and ranges. Further the monster(s) can be resistant or totally immune to the conditions you describe (grappled, disarmed, prone). You might be fighting a giant and a dozen cultists or any number of creatures where those tactics are not useful.

[MaxWilson]

Eldariel I owe you a better response than I can really write on my phone. Can't right now.

Short response to the other topic:

So I'll agree to the extent that 1v1 in close quarters against a Dex based character or monster the Strength based character will likely win. That's not really D&D though. You are 4v? in varied conditions and ranges. Further the monster(s) can be resistant or totally immune to the conditions you describe (grappled, disarmed, prone). You might be fighting a giant and a dozen cultists or any number of creatures where those tactics are not useful.

It _is_ D&D, it's just not this version of D&D. Some versions of D&D are very melee-oriented, but 5E has a large number of rules that tilt it towards D&D: Gunfight Edition. Adding Dexterity to longbow damage, having arrows shot from magical bows count as magical for purposes by weapon resistance, limiting opportunity attacks (both number of attacks made and making it cost a reaction and letting the target use its full Dex and shield bonuses to AC), allowing spellcasters to move while spellcasting, the mounted combat rules... that's five ways off the top of my head that ranged combat in 5E has been strengthened since AD&D.

Against a giant you should Disarm, and have somebody run off with the weapon. If the giant pursues, opportunity attack him each time he Dashes away until he stops chasing or dies from opportunity attacks. (If cultists pursue you both they're putting themselves in perfect Fireball Formation, so you actually want them to chase you too.)

I agree that against some enemies, Str is no better in melee than Dex, but against a majority of monsters it is much better (in melee), perhaps as much as 50-70% better. They aren't comparable. Dex is clearly better at range, and Str is clearly better in melee. (But ranged combat is clearly better than melee unless the DM's setting adds extra constraints.)

[MinotaurWarrior]

"neutral DM who neither panders for the party nor builds against them" and default optimisation goal is "campaign arc efficiency";

Sure on DM, but I don't think you can really talk about "arc efficiency". What does that even mean? A party of level one dragonborn clerics of Tiamat can stop playing HotDQ immediately and say, "this is fine". Really though, I'd argue that below you're talking about encounter efficiency anyway and it doesn't matter.

ergo, a party that can just divine the big bad behind the story, teleport to them, imprison them forever in a ring and then proceed on to the next campaign is better than the party that has to slog through the encounters to get to the oracle, then fight through the BBEG's lieutenants and underlings, then get the McGuffin and then disable them via. McGuffin. There are other options: if you want to optimise towards "encounters" instead of "campaign", then things like stealth and flight parties become more relevant; they aren't very good at dealing with campaign level stuff but their gimmick is good for encounters. Similarly, they're good with neutral or favourable DM but absolutely worthless with adversial DM who will obviously just make sure no encounter ever caters to your gimmick and thus it's useless.

Your wizard plan, though you didn't name all the spells, requires being a higher level (13th for Teleport), is less repeatable (at 13th level, teleport is your only spell that day), and has more risks of failure (saves, 25% chance of missing when teleporting based on a scry).

Also a hostile DM dunks on a party that relies on a wizard at least as hard as they do on one relying on a gimmick. Most modules have a silly lack of abjuration that a hostile DM can throw in to prevent a wizard from his shennanigans.

If we assume "campaign" and "neutral" as our optimisation paradigm, we can discuss your idea further. The problem with stealth optimisation as suggested is that Reliable Talent means 11 levels in Rogue. At that point a casting class would be pretty close to trivialising every single encounter that doesn't deal with the object itself via e.g. Scrying, Contact Other Plane and Teleport.

Teleport is a 7th level spell requiring a 13th level wizard. Until level 15, he can only use it once per day. Scry is a 5th level spell he can use twice a day, and has a substantial risk of failure. Contact Other Plane is a ritual he can cast, but the risk of insanity prevents spamming it. Even if those three spells solve every challenge at level 13, that's not trivial. It's basically using all of your resources, with a risk it doesn't work.

Stealth is first and foremost a low level power; we want to ensure sufficient minimum stealth to never be detected by standard enemies while stealthing without overspecialising. The thing is, overspecialising detracts from your ability to deal with the encounters that do beat your gimmick: a Rogue is just far worse at dealing with [thing] than any casting class, almost regardless of what [thing] is

That's a bold claim without support.

(call it 95% of [thing] on higher levels, which we're talking about).

I don't think we ever stated that, and I'd argue this is a very important point in establishing what we're trying to get at.

Most games are lower level.

The number of enemies that have 19 but not 22 PP is not very significant. I posit you gain more encounter points by focusing on your ability to deal with the 19+ PP enemies than by trying to further go for 22 PP enemies, especially since most cases where this matters are where stealth fails (that is, the circumstances don't lend themselves to stealth).

Did you think about the math I provided? The fact that there aren't many enemies with 23 but not 33 PP is precisely why that delta is so substantial. Also, the point about circumstances is why I mentioned for example invisibility - going all in on the gimmick includes obtaining the ability to employ the gimmick in adverse circumstances (e.g. without any natural obscuration)

Also IDK where you're getting the numbers 19 and 22. A sort of "dude with 16 DEX, Stealth Prof, and PWT" at level 11 can't roll lower than 18. A CR0 Eagle can find them. And without bonus action hiding, they don't have the action economy advantage .

More to the point, you gain way more campaign solving points by having the right spell to the task instead of having to engage with whatever mundane nonsense the campaign expects you to (take SKT for example; instead of going through the whole nonsense you can just figure out Iymrith and Slarkather are behind the whole deal, unveil their facade and proceed to kick their asses so hard they kiss the moons and skip ~5+ levels worth of encounters entirely).

SKT is an adventure for levels 1-11 where none of the tricks you've mentioned thus far are possible. I don't own it, so I can't assess it in depth. But I'm also going to throw out the likelihood that you even guess to try to scry them is less than 100%, and that you are able to convince anyone else is also less than 100%

Overall, I think all this just speaks for maximising your casting, specifically because campaign level magic does skipping encounters way better than skills. Better than Teleport is Rope Trick + Plane Shift, which we can include many classes in and thus get in and get out at will. The more 7th level spells the party has, the easier it is to, instead of bypassing encounters in some circumstances, to entirely just circumvent the whole arc of "get to X". I posit that Stealth is a Tier 1-2 strategy; Tier 3 changes the landscape of the whole game with the ability to just skip travel and movement entirely. Tier 1 travel is walking/flying, Tier 2 travel is Phantom Steed/equivalent, Tier 3 travel is teleportation. Tier 3 travel negates the whole concept of "encounter".

7th level spells exist for substantially less than half of the game. It also, again, is much more expensive (at most you get 6 7th+ spells per day) and I just have to disagree that it is more effective. Every strategy you've talked about has a risk of failure.

If we add flight, that's again a really worthwhile Tier 1-2 strategy. It comes at the cost of Tier 3 power (where everyone can fly of their own volition and more to the point, encounter skipping forms other than flight become more important).

They can start flying at level 5, for twenty minutes, by spending all of their resources.

At that point, bird gang (Warlock 4 / Sorc 1, CHA 15, Spell Sniper, Eldritch Spear, Lance of Lethargy) can cast expeditious retreat for 600 rounds of kiting an ancient blue dragon to death.

I don't think overspecialising for the few percentages is really worth it compared to just using minimum resources for a gimmick and then using your actual build resources towards making sure you can overcome scenarios that completely negate your gimmick.

Here's the reason I disagree: the few percentages translate into a much longer stretch of time between non-trivial encounters. That then allows you to overcome the very rare (less than 1/3 "fair" days) by nova'ing if you have to. Those scenarios you can't overcome with your exploit were designed to be beaten by a fair party that was wittled down by previous encounters. You will beat them too.

Again, increasing your combat range from 600' to 1200' is just not that major; the bigger issue in 600' vs. 1200' is spotting and terrain rather than your attack range per ce. You can get 600' combat range and flight fairly cheap which can easily be worth it but I posit that the difference between 600' and 1200' is marginal; the number of encounters that can deal with one but not the other is just minor. Neither is good for dealing with any non-combat encounter, trap, social encoutner, etc. for instance

But that marginal difference has a huge effect I've described many times.

Now I'll admit I'm coming into this with a focus on, not necessarily combat, but let's call it "encounters you'd pull out a map for". However:

1. Again, the nova rule applies. If you spent very few resources to bypass everything else, you can very easily and safely spend them on whatever challenge the gimmick doesn't apply to
2. Both gimmicks I've described, as well as several others, also bypass many social / environmental challenges (fly over the city gate instead of arguing to get in. sneak in the front door instead of using the treacherous pass to the side entrance)
3. The AT Party especially also incidentally owns at non-combat encounters, with party members having split up extra expertise and proficiency on other skills

In particular, the AT is actually better at using a lot of important wizard tricks because magical ambush is probably the most reliable way to increase the likelihood of a spell going off. Sneak up to someone. Suggestion. Bonus action hide. Disappear.

I actually think the opposite. In normal D&D as well as janky exploits parties, I think the party with a Wizard simply has options normal parties do not, which makes it more powerful. I don't think any individual gimmick is all-encompassing enough to make up for lacking e.g. Contact Other Plane, Teleport, Scrying, Clairvoyance, familiars or company.

If we're playing campaign level D&D, nothing bypasses campaign arcs like a Wizard. Similarly, if we think combat, nothing solves [enemy] like a Wizard simply because Wizard spell list is the most encompassing. But Wizard power is more in altering encounter nature to be more favourable and fighting CaW style information wars (Find Familiar, Magic Mouth, Phantom Steed/Tiny Hut, etc. are all scaling effects of the same style) - though it certainly also brings irreplaceable combat effects (Wall of Force is the best way to negate 99% of the Monster Manual; even if the enemy survives, it doesn't matter since you're miles away by the time it can move again and if the party has multiple casters, you can ensure the enemy will not survive with a DoT of course).

Here, again, you're mostly just saying that you disagree.

Wall of force is arguably the best rocket tag spell. But it exposes you to the danger of being visible within range and hoping to win initiative or survive a round. It's also something you can only do a limited number of times.

Re: Janky exploits specifically, I think they're just too one-sided. If they ever do run into a campaign arc where their janky exploit doesn't work, they lack the ability to bypass said arc like a party with all the 9th level spell lists available does. I think it's more important to be able to skip encounters than it is to defeat them, and more important to outmaneuver the enemies than it is to defeat them. Though I think you also want a decent degree of ability to defeat the enemy in cases where that is the campaign arc. If you must kill Zariel in hell, you must kill Zariel in hell; no amount of encounter bypassing gets you around that fact (though it can give you advantage in the act itself). I don't think a single module for instance is solvable via. Stealth without extremely loose stealth DMing (so not with a neutral DM). Many, however, are quite solvable via. teleportation and planar shifting combined with the ability to use a couple of Glyphs of Warding to nova an encounter dead with couple of hundred gold pieces (that you can trivially generate).

No level 1-12 module is solvable via teleportation.

I know I've gone through LMoP before and shown that either Kiting or Stealth can 100% RAW own it.

As for Zariel, that's actually a great example of a a nontrivial encounter.

AT Team level 11: The party comes in stealthed. The ATs cast Fog Cloud as a second level spell (40ft radius / 80ft diameter), then bonus action hide. Zariel is surprised.

Next turn +:

Zariel has bad options. Because of the action economy she can at most do one targeted attack, by doing a search action on her turn (+16 At best a 25% chance against the pure ATs, At best even odds against the multiclass), hope the one she finds doesn't go right after her, and then use her legendary action Immolating Gaze). So she has to rely on AOE (Blade Barrier, Fireball, or WoF). But she also doesn't know where they are, so where she targets has a random chance of targeting a space that hits nobody. Also those are both dex save spells and everyone has evasion and uncanny dodge (which is precisely as effective as her alternating between spending her turn going invisible and fireballing, so let's ignore that tactic). There's something to be said about using blade barrier to divide the fog up first (Three of the rings), and then narrowing the search space to the quadrant the PCs last attacked from. But even then, it's going to be less than 100% target rate, and will deal on average 7 damage on a failed save. And if push comes to shove (e.g. lots of good WoF placements) the PCs can just take the BB damage (again about 7) and even cast a / move to a new fog clouds.

PCs: On the second turn, ATs caste haste(action)-attack(haste)-hide(bonus). On subsequent turns ATs Attack(Haste)-Hide(Bonus)-Move-Ready(Action) (Attack right before my turn starts). On the third turn, ATs Hide(Haste)-Attack(Action)-Hide(Cunning). Rinse, wash, repeat, for 1.5 Sneak attacks per AT per round. Assuming just silvered arrows, 7d6+5 = 29.5 damage, hit 69.75% of the time (Unseen attacker advantage) 9.75% crit, average attack gets 22.965 damage, 103.3425 DPR, -20 for regeneration, ~83, she's down in about 6 hasted rounds (not counting the likely damage from the first round of attacks)

The multiclass, honestly, doesn't even need to do anything after round one except move and use a cantrip to keep her down.

And then they've exhausted some resources, but can continue to auto-win lots of encounters the rest of the day. They can even, very likely, take on multiple equivalent challenges.

EDIT: I suppose I should also defend my position on Wizard vs. a Generic DPS Class With Healing Word. I think the most important way of preventing "cascading character failure" as you put it is to minimize the chances of it happening. I think the best way to do it is magic that disables; this is because enemy HP scales ridiculously but enemy saves don't. Thus the most reliable way of making sure the enemy isn't outputting more damage than you're capable of taking without the whole party being downed before the healers get to it. As an example, Princes of the Apocalypse features an encounter for level 5 characters featuring up to half a dozen Fireball casting enemies with 50+ HP each. The only real way to deal with something like that is to use magic like Hypnotic Pattern and to beat their initiative.

Back in "fair" mode:

Sure, you're absolutely want to be able to target saves. But while the wizard is good at that, non-Blade Singer wizards have a lot of that effect attributable to the PCs who keep them alive in the encounters they're not using their top 2 spell slots. And even then, the strategy here (and many others) are essentially hammer-and-anvil strategies where a lot of the power of Hypnotic Pattern comes from the other classes with the AC HP Saves and DPR to take on the monsters that make their saves and then pick off the rest one by one.

Even your note "and to beat their initiative" concedes the vulnerability Wizards have when playing rocket tag. The Lore Bard has 5 more HP +3 better dex saves and higher initiative with which to win rocket tag.

This isn't to say that Wizards suck - I might agree that they are generally the most powerful class. But it's to explain why their value as party members is a bit lower than their power level.

Again, I think Wizard is just the best party to such an end though Bard also works. I don't think a Healing Word class really adds much to it; it's a failsafe after the "TPK potential" is disabled but the part that makes the party survive the encounter is a couple of concentration disables that prevent the Fireball Doom. And that's something only CC classes can provide; they do so proactively so it's not as obvious since the party doesn't get to know what would've been but that's all the more valuable since it not only saves the party but also significantly conserves party resources.

This is the "or averting TPK". But I disagree that only CC classes, or especially only Wizards, provide it in a fair party. In boring medium encounters where the wizard is conserving resources and just using cantrips, the Paladin tanking (exploiting the exponential returns of higher AC and Saves in terms of attacks-survived) may be averting a TPK (or lost adventuring day, by having to spend extra resources and rest early). When there are threats worth Nova'ing, the Paladin Smitefest can avert a TPK. Now I agree that this is a smaller effect. In a fair party playing fair DnD, the guy who can cast Hypnotic Pattern twice a session does get more "disaster avoided" points. But it's important to not go all-or-nothing.

And I think that within the realm of CC classes the Wizard doesn't have that huge of an edge. The Lore Bard can do most of what the wizard can do, but they also have better saves, better AC, better HP, expertise, and bardic inspiration. Sorcerers can do a fair share of what wizards can do, but they have slightly better saves, can do what they choose to do better (WM w/ Heightened Spell is the best at making enemies fail saves) and Divine Souls gain access to some critical tools the Wizard lacks. Again, to reiterate, this isn't to deny the Wizard's power. They are very powerful, but I think that as a party member their value isn't as high as their abstract "power".

To try to formulate what I believe is the ultimate measurement of party power, I think narrative overrides are the de-facto thing we should be looking at since that allows the party to pick exactly what to engage in and thus picks what's the most favourable to them. Of course, this links purely to CaW, but I don't think CaS really makes sense in the context of ~a third of the spells chapter of the PHB, so provided we consider the whole system, I posit we should assume CaW framework.


But it's true that the framework needs at least four components before we can get to the party optimisation framework itself:
1) Combat-as-War or Combat-as-Sports?
2) Pandering, neutral or adversary DMing?
3) Encounter-level or campaign-level optimisation?
3b) Maximal difficulty or maximal endurance?

Are there any aspects I'm missing here?

I think we should assume RAW WRT e.g. yoyo healing since that is the only common framework we all have and probably the most commonly used, though we could also try and address common houserules (Exhaustion upon healing or negative HP, I think). I think that would make the list most broadly applicable. It would probably be too cumbrous to consider more than few outliers (but perhaps yoyo healing specifically is common enough a house rule consideration that it warrants mentioning in rankings).

I think you have to assume RAW and a Neutral DM (or maybe let's call it a sort of "Robot" DM, since sometimes strict RAW feels either pandering or hostile towards the edges).

The other points though I think are coming at it from a slightly odd angle. CaW and CaS are, as I understand them, more about the strategy you choose, and could theoretically vary from party to party (e.g. the Birdgang is very CaS just pewpew lazering everyone, the AT Team are more CaW getting the job done with minimal fuss). Campaign optimization, when detatched from encounter optimazation, gives us "The only winning move is not to play." Just be OK with whatever happens without PC intervention and you win immediately. What I think you're really talking about, even with Teleport strategies, is the ability to resolve key encounters.

I think, fundamentally, in normal fair DnD, there's basically a single success criteria with a single level (you can't "win more") everyone can achieve (fair DnD doesn't include a monster with a ridiculous speed and a ridiculous starting distance running away as fast as it can that you need to catch, when only the Bird Gang can) and so the chief issue is just not losing.

I think there's some secondary points (Adventuring day efficiency, ability to take on challenges you're underleveled for, ability to take on "impossible" challenges, etc) but they're not that important in fair parties with normal DnD. I'd say a party that spends twelve days to accomplish what another party can do in one is a worse party, but anyone who is accomplishing a standard adventuring day in a day, and a standard session in about a session, is fine imo. Maybe call this "failure due to time out".

The other issue I think you're passing over is this:

Are we talking about a party made all at once (as in a 1-on-1 DnD game with a single player playing all PCs) or are we talking about assessing classes by their party contribution in a more organic party.

In a party made all at once, maybe (I'm not convinced, but it's plausible) you can argue that there are fair parties with wizards that outclass all fair parties without wizards because you've compensated for all of the downsides and the "spells no spells-known caster would ever pick" gives you a slight edge.

But realistically, with everyone coming to the table together making their own choices on what to play, I think it's reasonable to guess that the lore bard adds more value to the party than most wizard sub classes. The lore bard can pick up those few elite powerhouse spells (including ones the wizard can't), is a little tougher, does better on skill checks, and goes first more often (all meaning less of their success is attributable to other classes compensating for their weak points).

Someone mentioned googling shapley values and getting lost, which is 100% fair because the wikipedia pages on math stuff are often totally unreadable unless you already understand them. The key point is this: to asses one element's contribution to the whole, you've got to be agnostic to the order in which elements are added (to be fair in considering synergy and redundancy), and to achieve that, you've got to imagine every order of adding elements.

So the (non-BS) wizard alone sucking so bad (I think everyone would agree with that. Wizards don't make good solo characters) should count against it, and the advantage it brings to a party that compensates for those weak points is a synergy that needs to be attributed not just to the wizard, but also to the classes that compensate for it's weak points.

[MaxWilson]

Hmm, I'm not entirely in agreement. Yes, of course much is beyond what you have available. But we can lay a groundwork:
1) You need resources to e.g. blackmail people. You can't just tell someone to go do something for you, you need to get something that they hold in value.
2) Hiring NPC mercs...requires that there are mercs of sufficient power available. Similarly, it requires that the enemy hasn't hired them for you. We can also surmise it heavily depends on your social skills, related magic, etc. as well as what you are able to offer to the mercs.
3) We can probably assume that if engaging a big league event (say, OotA; demon lords on the material plane, hooray!) there probably aren't mercs able/willing to help you. You need help from other movers in the world, who probably aren't that easy to gain under your power (ergo power is something that enables you to gain people under your power).
4) There's a very meaningful difference between Scrying and blackmailing somebody to Scrying. Scrying is something you can do yourself and at will. Blackmailing is a complex, multistep process that produces enmity of the blackmailee, and might be fleeting as whatever you're blackmailing them with might come out of your power making them free to move against you. Procuring the leverage in the first place is far from trivial. First and foremost though, if you blackmail somebody you need to ensure they cannot deadify you before you can pull on your leverage.
5) DPR is...always useful. If you can kill somebody before they can act and they know it, they're more likely to do what you want. However, if that's how you procure power, you will create numerous enemies making your position always precarious.

So what this comes back to: All of CaW options still depend on the party's resources, i.e. how good the party is in e.g. making the merc gang rather take their offer than the opponent's and what's the failcase (since mercs are never all that trustworthy if we go by their modus operandi of "work for the highest bidder"). Things like social power and monetary power (the ability to produce money too) are of course highlighted.

Further, no matter how carefully you play in the shadows, the world has a lot of powers in the shadows and you probably cannot avoid crossing enemies if you want to accomplish anything. In that sense, you'll always be under threat and thus your personal power, defense and survivability (especially the tools therein; how safe a place can you rest in, how quickly and efficiently you can move, how unpredictable you can remain, etc.) is an advantage pretty much regardless of how the game is run since if we make the assumption that some powers in the world probably want you dead, you need to always stay one step ahead.

All of this does come back to the information war, which highlights a specific subset of magic as well as certain skills, but those certainly do have mechanical frameworks in the system (I can't call skill system "rules" since they're so out there but they are what we have). In general, whether DM sets the DC at 15 or 20 to e.g. convince the Black Rose to reveal their previous employer with a generous offer of gold is not that relevant; higher bonus is better anyways. Similarly, if you can just get to the boss (finding the boss of a shadowy merc organisation is, again, of course easier said than done but there are plenty of ways characters can engage in such a task) and Suggest that they tell it to you in exchange for a fair amount of money, well, that's probably a fine plan too. Or mayhap you create a facade of a conversation while somebody is covertly trying to read the important information out of your subject's mind.

Either way, I don't believe CaW really meaningfully changes how you approach the characters/rolls themselves. It merely just highlights the game outside the Encounter itself; which, true, has less rigid guidelines but guidelines nonetheless on what you can and can't do or what kinds of options are open to given character classes.

While the five observations you make are valid, I don't believe the observations constitute meaningful analysis about which builds are best. You can't do meaningful analysis without knowing more specifics than are available in RAW. Are wizards better than beastmaster rangers at tracking down relatives of wizards whom you want to blackmail? Who knows!

You're not wrong that personal power still matters, but again, no meaningful analysis is possible in the absence of data. Is a Bladesinger with Mage Armor and Foresight for 8 out of 24 hours every day better-protected against threats than an Assassin with a false identity? Who knows!

[Merudo]

You can't do meaningful analysis without knowing more specifics than are available in RAW. Are wizards better than beastmaster rangers at tracking down relatives of wizards whom you want to blackmail? Who knows!

Absolutely.

I view RAW as mostly describing the physical laws of the game world: how many spells can a Wizard who just learned his craft cast in a day, how much damage does falling 100 feet cause, how long does a character have to rest to restore their strength, etc..

What's missing from RAW are the sociological / ecological "laws" of the game world: how are societies organized, what's the distribution of the creatures and monsters in the world, what types of ecosystem exist, etc.

D&D was designed to cover a wide range of worlds, such as the Forgotten Realms, Eberron, Magic the Gathering planes, and homebrew worlds. Attempting to create comprehensive sociological/ecological RAW that would simultaneously cover all these disparate universes is a fool's errand.

It would, however, be perfectly doable for WotC or third party publishers to produce additional sociological / ecological rules to cover specific worlds such as the Forgotten Realms. Such rulesets would be invaluable to support a CAW playstyle.

Unfortunately, there has never been such a comprehensive ruleset for any world in any edition of D&D. Even when covering a very narrow location (say, Baldur's Gate or Waterdeep), D&D material typically only covers notable NPCs and places, and omit any kind of sociological / ecological description of the setting.

A good supplement should, to support CAW in a specific locale, cover at the minimum:

• Population employment by major industry sector, with special emphasis on the prevalence of spellcasters, rogues, and for-hire mercenaries
• Style of governance, including functioning of institutions and prevalence of corruption
• Code of law of the locale, including juridical procedures and criminal punishment
• Trade and diplomatic relations with the neighboring entities
• Price elasticity of demand, in particular for poison, magical items and spell components

All of this matters a great deal for CoW. For example, the success of PCs partaking in crime, spying or blackmailing is highly dependent on the prevalence of magic users able to cast spells such as Zone of Truth, Augury, and Divination; if the Enthralling Performance ability of Glamour Bards is well-known, singing and dancing might be viewed with suspicion, if not banned outright; inheritance law is likely to be much different in a setting where the Resurrection spell is accessible; the Locate Object spell will be useless if folks commonly cover their house and backpacks in thin sheets of lead as a protection.

The absence of CoW-enabling rules mean that DMs either have to painstakingly flesh out the world they are using, or come up with these sort of things in an ad hoc manner at the table. Neither is desirable, both for the DM, who has to put a lot of work for his world to support CoW gameplay, and the players, who cannot meaningfully predict what effects their CoW actions will have on the DM's world.

Against a giant you should Disarm, and have somebody run off with the weapon.

It's the second time in this thread that I've seen you mention Disarm, and I'd like to point out that unless the character is a Battle Master, disarming requires the DM to adopt the optional Disarming rule.

[KorvinStarmast]

optimizing to minimize the chances of catastrophe is a perfectly reasonable approach.

Then why don't we make that "Optimization Case 1" and work on it's detailed features?

A related way to think about party optimization could be "how robust is this party against player error?"

Yes.

Are there certain PC classes which rely less on player skill, perhaps to the point where an incompetent or malicious player still cannot lose?

Sheer incompetence is a powerful force in the multiverse. (This is a riff off of Heinlein's "there are only two universal constants, Hydrogen and Stupidity"). For my money, and in the West Marches spirit, incompetence and foolishness ought to be rewarded with character death and a chance to roll up another one.

The Holy Grail of another approach (Avoiding Catastrophe) might be "a party build which is so reliable that you can run 10,000 Adventurer's League tables with this build through the same WotC adventure without a single TPK."

I'll further suggest that "Is AL compatible" must be a constraint on the optimization philosophy, since the point you make about DM's varying from a core conceit - yo you healing - has a major impact on PC mortality.

The Holy Grail of a third approach (Minimizing Incompetence) might be "a party build which is so reliable that one semi-competent newbie and three covertly-malicious saboteurs

I don't think this is a fruitful pursuit. The point of party optimization is that the players care enough to do this as a team. I'll suggest we table this category for the purposes of this exercise.

Furthermore, builds are only optimal w/rt a given set of rules and a style of DMing. E.g.

Hence, let's frame this whole exercise as "AL compatible" and go from there.

First of all, excellent posts everyone, thank you. This is quite the fruitful conversation, I can see.

Yeah, thanks for starting the thread.

What are we optimising towards?

While I suggest our first category ought to be Max's "minimizing catastrophe" there is another category for "high risk high reward" party style. Not sure how to describe that yet.

Stealth is first and foremost a low level power; we want to ensure sufficient minimum stealth to never be detected by standard enemies while stealthing without overspecialising.

Sneaking around in the dungeon was an OD&D standard approach to surviving, and fits IMO into the "avoiding catastrophe" class of party optimization. It is also how you fold the "scouting" and "fighting on your own terms" element of CaW into a party's MO. I'll go a little Sun Tzu here and suggest that an optimized team/party doesn't just kick in doors and find out "so, what's in here to fight?" as so many Video Game trained players seem to do. Listening at the door first was a standard OD&D habit ingrained into us by lots of PC deaths for those who did not ... or poking something with a 10' pole, or pouring water on the floor and seeing where it flows ... this is where the exploration pillar stands out.
Do we need to fight?
Do we know our terrain?
Do we know our opponent?
Are we fighting on terms of our own choosing?

All of these contribute to Max's "avoiding catastrophe" party theme and optimization framework.

If we're playing campaign level D&D, nothing bypasses campaign arcs like a Wizard. Similarly, if we think combat, nothing solves [enemy] like a Wizard simply because Wizard spell list is the most encompassing. But Wizard power is more in altering encounter nature to be more favourable and fighting CaW style information wars (Find Familiar, Magic Mouth, Phantom Steed/Tiny Hut, etc. are all scaling effects of the same style) - though it certainly also brings irreplaceable combat effects (Wall of Force is the best way to negate 99% of the Monster Manual; even if the enemy survives, it doesn't matter since you're miles away by the time it can move again and if the party has multiple casters, you can ensure the enemy will not survive with a DoT of course).

Does this mean that any optimized party has a wizard?
If we go to the Basic Rules, the four classes for a party are Fighter (Champion), Cleric (Life) Rogue (Thief) and Wizard (Evoker). Is that pre-optimized?

But it's true that the framework needs at least four components before we can get to the party optimisation framework itself:

1) Combat-as-War or Combat-as-Sports?
2) Pandering, neutral or adversary DMing?
3) Encounter-level or campaign-level optimisation?
3b) Maximal difficulty or maximal endurance?

I suggest that item 2 leads us to an unsolvable problem.
Rocks fall, everybody dies isn't something you can design for.

1) Combat-as-War or Combat-as-Sports?
2) Pandering, neutral or adversary DMing?
2) Encounter-level or campaign-level optimisation?
3) Maximal difficulty or maximal endurance?

For Max: I'd have rather seen (short and long) bows do "to hit with dex" bonus" and "damage using strength bonus" and crossbows only get one shot per round without the feat unless using object interactino to load the second shot. But that's fiddly, and water under the bridge, and moves away from the "keep it simple for new players" approach that is one of 5e's strong points.

What's missing from RAW are the sociological / ecological "laws" of the game

While a productive line of thinking, particularly as a DM, I don't think this fits into the thread in terms of establishing party optimization criteria and goals.

[Merudo]

While a productive line of thinking, particularly as a DM, I don't think this fits into the thread in terms of establishing party optimization criteria and goals.

My goal was to (1) echo MaxWilson on how CAW-based optimization is not really possible if we limit ourselves to RAW, and (2) create a tentative outline of what would need to meaningfully discuss CAW optimization.

I believe my contribution fits perfectly into the thread in term of establishing optimization criteria for CAW play.

[MaxWilson]

My goal was to (1) echo MaxWilson on how CAW-based optimization is not really possible if we limit ourselves to RAW, and (2) create a tentative outline of what would need to meaningfully discuss CAW optimization.

I believe my contribution fits perfectly into the thread in term of establishing optimization criteria for CAW play.

I think your critique is a valid one ("Unfortunately, there has never been such a comprehensive ruleset for any world in any edition of D&D") but the conclusion I draw from this is, "There is no unified CAW system in use. Your only choices are to optimize for a specific DM's style, or to write off CAW as unsolvable and either stick to discussing useful CAW tactics/strategies without trying to impose a ranking on them, or stick to discussing CAS."

For my part I'm interested in hearing about cool CAW-applicable tricks, moderately interested in CAS tricks, and not actually all that interested in figuring out which CAW parties are globally "optimal." For example, I'm interested in any cool ideas people have for ways to abuse Deception/Changeling shapeshifting/wildshape/Disguise Self/Seeming/Minor Illusion/Major Image/Programmed Illusion, tricks that I might not have thought of. I'm not super-interested in comparing these tricks to the value of simply killing whatever creatures you might otherwise trick--that judgment is best made in play, at the actual table. It could turn out that one of these approaches is more "optimal" at a given table than another (some DMs just don't like it when you bypass fights), but there will still be a measure of subjectivity involved, and besides, the discussion is the interesting part, not the ultimate ranking.

[KorvinStarmast]

My goal was to (1) echo MaxWilson on how CAW-based optimization is not really possible if we limit ourselves to RAW, and (2) create a tentative outline of what would need to meaningfully discuss CAW optimization.

I believe my contribution fits perfectly into the thread in term of establishing optimization criteria for CAW play.

Except that I think it's DM facing, not player facing. There is an entire game system of rules missing, per your insightful post (and I roughly agree with you there) - heck the economics system alone is basically unaddressed. I was under the impression that we were attempting to arrive at a party optimization framework for D&D 5e as it is.

Did I misunderstand that?

If we are not approaching this exercise from the point of view of "we, the players, are building a party, and we want to optimize it for (something)" then I wonder at what we are doing with this sharing of thoughts.

This is where I think that Max's category of "Avoid Catastrophe" is one approach, a "high risk high reward" concept for a party is another one, and IMO to be useful it needs to attach an AL compliant rider if this philosophy or tool is to apply to as many tables as is practical. Otherwise, how does this tool aid the player?

I won't comment on how CAS or CAW is or isn't achievable, or rather to what extent it is achievable with the RAW, other than to suggest that it is very dependent on DM style.

And do you really know that style before you begin play?

Hence my previous suggestion that optimizing for a DM style seems to unlikely to be achievable.

[Merudo]

Except that I think it's DM facing, not player facing.

I would view it as community facing. If DMs, players, and content creators are on board on the idea of a CAW-based supplement for D&D, then maybe such a supplement will actually get released .

Not only would it be a boom for CAW lovers everywhere, it would actually become possible to discuss party optimization for the supplement.

I was under the impression that we were attempting to arrive at a party optimization framework for D&D 5e as it is.

The title of the thread is "Party optimization philosophy", so you'll excuse me if I'm writing about what theoretically could-be instead of what-is .

[Klorox]

Fascinating thread idea.

General question: if you discuss party makeup with the other players before a campaign begins, do you actually bring up things like this, or do you concentrate on just your own characters?


I'm also wondering what tier(s) these rankings apply to.

For instance, I love the bard, and I love the paladin. But I think the paladin is a much stronger class in the early game than the bard. And while the support aura abilities of the paladin can never be undervalued, the bard really starts standing out after they reach tier 2.

[KorvinStarmast]

I would view it as community facing. If DMs, players, and content creators are on board on the idea of a CAW-based supplement for D&D, then maybe such a supplement will actually get released

I think that without bounding the question we cannot arrive at an optimization philosophy, or scheme, or even categories. And calling on WoTC to put out such a supplement strikes me a really, really unrealistic. I proposed further up an "AL Valid" constraint otherwise ... just how broadly useful is this approach? (I wish I could answer that, but I really don't know).

The title of the thread is "Party optimization philosophy", so you'll excuse me if I'm writing about what theoretically could-be instead of what-is .

While an interesting line of inquiry on the theoretical side, I was under the impression that we were trying to arrive at a framework, or a set of principles, that sufficiently bounded the problem so that something applicable can be arrived at.

This thread would be a necessary first step: how does one clearly explain how we look at party optimization? Step two would be something along the lines of
"Given the three/five/seven optimization categories we agreed on in Step One, lets examine a few packages to see if they fit our created model, or if the model/models already have some glaring shortcomings."

I'd expect it to be an iterative process.

General question: if you discuss party makeup with the other players before a campaign begins, do you actually bring up things like this, or do you concentrate on just your own characters?

Me? As a player? Always. I always discuss how the party is made up. I do not like to have blatant capability gaps.

As a DM? Not my job, man. Bring whatever party you have. We'll figure out a way to have furn.

I have discovered this: if it's a published adventure, look out, the party (one poorly built) may hit a roadblock ...

[5eNeedsDarksun]

Fascinating thread idea.

General question: if you discuss party makeup with the other players before a campaign begins, do you actually bring up things like this, or do you concentrate on just your own characters?


I'm also wondering what tier(s) these rankings apply to.

For instance, I love the bard, and I love the paladin. But I think the paladin is a much stronger class in the early game than the bard. And while the support aura abilities of the paladin can never be undervalued, the bard really starts standing out after they reach tier 2.

The party I described earlier in this thread as our most lethal was definitely somewhat pre-planned. We knew we were starting an 'intown' campaign in Waterdeep and made a decision to have at least some level of stealth in every character. We didn't necessarily set out to make an overpowered group, and on paper it didn't look as good as our previous one. However, as the campaign went on and we encountered various terrain/ obstacles/ monsters we started to realize just how nasty we were and how impactful it was to have an entire party that could get in optimal spots before most fights started (or avoid them when we wanted). So I'd continue to advocate that optimal parties have nobody in heavy armor.

[Corran]

I have to respond to this specifically, since I see this sentiment a lot when discussing the necessity of Paladin in an optimal party. While Paladin is a strong class, it is my own experience that Aura of Protection is a bit of a paper tiger. Sure, it looks absolutely bonkers on paper to get +5 to all saves for the whole party (and even just personal +5 to all saves is incredibly strong), but the party benefits are often less impressive than that IME, mostly due to the 10' range (up until level 18, which is after full casters already became demigods).

Often, the things you want a saving throw bonus for are AOEs (Fireballs, dragon breaths, Banshee Wail, etc.) and in those encounters, huddling close together for the save bonus guarantees that basically the whole party gets hit. OTOH if you have enough room to split up and be ~30' apart (in the "Fireball formation"), the enemy will have to hit at most two characters. So to gain benefits of Aura of Protection, the party will have to accept more hits from the AOE (obviously this doesn't apply to non-AOE but I find those aren't as frightening all that often; most seem to be like max HP reduction or poisoning or curses or things of that nature that you can tend to out of combat, though save-or-dies and mind control effects do of course exist), which kinda defeats the purpose. Even if everyone saves thanks to the Aura, the total damage split across the party is the same as if two guys got hit and failed the save (let alone if two guys got hit and succeeded the save).

You dont need to be in fireball formation to make good use of aura of protection, that's silly (as you rightly demonstrate). Aura of protection is important when it helps cover for a weakness you dont want the enemy to exploit (and yes, it's a very good point that bardic inspiration is a great way of doing that). For example, if I know/suspect that my barbarian ally is only one save away from being taken out of the fight, I might want to put my paladin close to them, even if it means that I will also get hit by the same effect (perhaps even if it means that I am not putting myself in a good position to tank/go nova). That would probably be because the barbarian is the most important asset during that fight for some reason (for example because he can throw enemies off a cliff faster/more efficiently than any other pc can; in which case succeeding at the save could mean 4 enemies thrown off the cliff, and failing the save could mean 0 enemies and 1 barbarian thrown off the cliff). Of course, in an optimized party there will probably not be a barbarian and everyone will take care of their saves to some degree, so I am thinking that the value of the paladin drops the more optimized the party and its tactics are (two caveats; first, with sufficient control, the paladin makes an excellent tank assuming a dip in caster while also holding onto some support and nova capabilities, and second, crusader's mantle could be useful in a party that uses lots of minions -though I have a feeling that I am overestimating this last one and I am waiting to be corrected). Still, there might be a small number of tactics where you want the party to actually fight in fireball formation (eg fighting inside the sickening radiance of an evoker, or trying -most likely unsuccessfully- to bait a dragon's breath -which you plan to counter with a wall of force/stone- and its frightful presence -which you'll counter with aura of courage; I am not sold on either of these and I haven't tried or put that much thought into them), in which case the various paladin auras might be handy (courage is a core piece of the strategy and protection is a tiny risk mitigating factor), but I suspect that would be very rare (and there are probably better alternatives most of the time, given the constraints we would need in order to fall back to tactics like those). In short, it (aura of protection specifically) is a reactive ability. A great one (but a poor contingency at the same time), but even the best reactive abilities are at their best when you didn't plan something right (either when building characters or ahead of a fight).

[MaxWilson]

The party I described earlier in this thread as our most lethal was definitely somewhat pre-planned. We knew we were starting an 'intown' campaign in Waterdeep and made a decision to have at least some level of stealth in every character. We didn't necessarily set out to make an overpowered group, and on paper it didn't look as good as our previous one. However, as the campaign went on and we encountered various terrain/ obstacles/ monsters we started to realize just how nasty we were and how impactful it was to have an entire party that could get in optimal spots before most fights started (or avoid them when we wanted). So I'd continue to advocate that optimal parties have nobody in heavy armor.

I agree that everybody in the party should be stealth-proficient, but not that nobody should have heavy armor. Whether stealth is applicable to a given scenario depends, how much stealth you need to go undetected also depends, how many PCs are needed on the stealthy part of the mission depends***, and there are ways to mitigate heavy armor disadvantage anyway. (Enhance Ability (Dexterity) is one way, Bardic Inspiration is another, Guidance can help some, and Pass Without Trace trivializes stealth despite heavy armor disadvantage, as long as you are proficient in Stealth.)

*** For instance, if the goal is to defeat the enemy in detail by repeatedly infiltrating their territory and repeatedly provoking individual groups into attacks on a prepared defensive position, you probably don't even WANT to Enhance Ability (Dexterity) the heavily-armored Defense style paladin so he can go with the Shadow Monk. Instead you want him guarding the caltrop-littered chokepoint on level 1 with the Necromancer and his dozens of skeletons. Let the Bard with Stealth Expertise go with the Shadow Monk instead, prepared to pull them both out via Dimension Door if necessary.

You dont need to be in fireball formation to make good use of aura of protection, that's silly (as you rightly demonstrate). Aura of protection is important when it helps cover for a weakness you dont want the enemy to exploit (and yes, it's a very good point that bardic inspiration is a great way of doing that). For example, if I know/suspect that my barbarian ally is only one save away from being taken out of the fight, I might want to put my paladin close to them, even if it means that I will also get hit by the same effect (perhaps even if it means that I am not putting myself in a good position to tank/go nova). That would probably be because the barbarian is the most important asset during that fight for some reason (for example because he can throw enemies off a cliff faster/more efficiently than any other pc can; in which case succeeding at the save could mean 4 enemies thrown off the cliff, and failing the save could mean 0 enemies and 1 barbarian thrown off the cliff). Of course, in an optimized party there will probably not be a barbarian and everyone will take care of their saves to some degree, so I am thinking that the value of the paladin drops the more optimized the party and its tactics are (two caveats; first, with sufficient control, the paladin makes an excellent tank assuming a dip in caster while also holding onto some support and nova capabilities, and second, crusader's mantle could be useful in a party that uses lots of minions -though I have a feeling that I am overestimating this and I am waiting to be corrected). Still, there might be a small number of tactics where you want the party to actually fight in fireball formation (eg fighting inside the sickening radiance of an evoker, or trying -most likely unsuccessfully- to bait a dragon's breath -which you plan to counter with a wall of force/stone- and its frightful presence -which you'll counter with aura of courage; I am not sold on either of these and I haven't tried them), in which case the various paladin auras might be handy (courage is a core piece of the strategy and protection is a tiny risk mitigating factor), but I suspect that would be very rare (and there are probably better alternatives most of the time, given the constraints we would need in order to fall back to tactics like those). In short, it (aura of protection specifically) is a reactive ability. A great one (but a poor contingency at the same time), but even the best reactive abilities are at their best when you didn't plan something right (either when building characters or ahead of a fight).

A fair point. Paladins auras are definitely nice in certain situations, e.g. wizard casts an important Wall of Force spell on the biggest bad guy in the fight, paladin moves to boost his concentration saves (and Wisdom/Constitution saves vs. incapacitation). Bardic Inspiration tends to be useful in the same situations, although Paladin aura gives a more predictable bonus.

Likewise they're good for keeping self-Polymorphed PCs in Polymorphed form longer.

I dunno if I would call that "reactive" though, just "situationally excellent."

[Corran]

I dunno if I would call that "reactive" though, just "situationally excellent."

Yeah, you are probably right. Being not always able to account for the terrain is enough reason to call it so.

Self polymorphing sounds like a very desperate move, even with a paladin's aura backing the caster. But perhaps I am missing something, so I'll try to think a little harder on it (after a long rest).

[MaxWilson]

Self polymorphing sounds like a very desperate move, even with a paladin's aura backing the caster. But perhaps I am missing something, so I'll try to think a little harder on it (after a long rest).

It depends on the form. E.g. a Brontosaurus has +6 to Con saves already, so if you can get a little bit of a boost from a paladin aura there's a good chance you'll be able to stay dino all the way down to 0 dino HP. T-Rex and Giant Ape both have +4, so if your paladin grants +5 you can only fail the save if the damage >= 22 HP.

Mage Armor + self-Polymorph (Giant Ape, AC 15, HP 157) is really quite a good use of resources when you happen to already have a +5 Paladin aura in the party for other reasons. Make sure you get a DM ruling first on what counts as "in the Paladin's aura", since you obviously can't squeeze your whole body into range unless the paladin is riding you as a mount. (Speaking of which, Mounted Combatant partymates are also good for the self-Polymorph strategy.)

N.b. The prime virtue of Polymorphing yourself instead of someone else is that if you have low DPR you can bring the party average DPR up by switching, while also avoiding taking another spellcaster out of circulation who might otherwise find something useful to do with their concentration.

N.b. #2 Some people might worry about the saving throw penalty from lower stats, because since you can release concentration literally at any time, it's safer than it seems. If a Mind Flayer comes out of nowhere and Mind Blasts you, you can instantly turn back into a human Druid or Wizard even while you are in the middle of fighting off the effects (a.k.a. making an Int save).

[Eldariel]

Since there's still interest in discussing this topic, I'll just update that I'll try and work on a synthesis for the framework over the next week (though anyone else is of course free to try as well). The criticism from MaxWilson and Merudo is certainly something that needs to be tackled if something like this is to have any practical value.

[KorvinStarmast]

Since we're trying to do useful optimisation here, I think "general use" has to be the answer. Though I think a combat-based listing would have a place as well. But hmm, the framework for the framework is getting pretty cumbruous already so this seems like it'll be quite the project.

Assumptions and frameworkd ideas listed by me for vetting by the kindly participants of this thread.

1. No DMG optional rules in use. (Facing, flanking, lingering wounds, massive damage death, and so on)

2. vHuman and feats are in play, and multiclassing is also acceptable.
a. With multicalssing
b. Without multiclassing
(Not sure how much this has an impact, but it might on small party sizes like 3)

3. The optimization baseline is that a party is confronted with Exploration, Social, and Combat challenges.

4. Assumptions on campaign length. I suggest that we need two separate optimization categories.
a. A campaign that will go from level 1-20.
b. A campaign that will go from level 1-12. (Which is where a lot of published adventures seem to tail off or end)

5. Party size. Three cases
a. Three member party
b. Four member party
c. Five member party

6. Optimization purpose.
a. Max's "Avoid Disaster" concept
b. High Risk High Reward / YOLO concept
c. Something else?

7. AL compliant: a separate optimization category since it limits sources. This case would be presented so that folks can take the final products of the effort and apply it to a public play experience.

Each of the above seems to me to be important cases that will influence optimization; in particular, party size.

[MaxWilson]

7. AL compliant such that folks can take the final products of the effort and apply it to a public play experience.

I think AL compliance should be its own separate section, like party size, not a baseline assumption. AL has requirements like mandatory point buy and PHB+1 which aren't valid assumptions outside of AL.

Perhaps a good way to think of it is together with other chargen constraints like multiclassing: Basic only, PHB no optional rules (including point buy), PHB mandatory point buy and feats but no multiclassing, PHB all optional rules allowed, PHB/Xanathar/SCAG/Volo/MTOF with a be without multiclassing, all WotC sources including UA with all optional rules allowed, AL-style mandatory point buy and PHB+1. I believe these are all common configurations.

[KorvinStarmast]

I think AL compliance should be its own separate section, like party size, not a baseline assumption. AL has requirements like mandatory point buy and PHB+1 which aren't valid assumptions outside of AL.

Perhaps a good way to think of it is together with other chargen constraints like multiclassing: Basic only, PHB no optional rules (including point buy), PHB mandatory point buy and feats but no multiclassing, PHB all optional rules allowed, PHB/Xanathar/SCAG/Volo/MTOF with a be without multiclassing, all WotC sources including UA with all optional rules allowed, AL-style mandatory point buy and PHB+1. I believe these are all common configurations.

OK, fair point, I'll edit that. Bounding the problem by identifying constraints seems to me an important piece of "Step 0" before we can have proposals for party optimization.

[MinotaurWarrior]

Restating my premises more clearly:

The goal of all optimization is to maximize the chance of victory. Unlike in other cooperative ventures (e.g. business) there is no such thing as "winning more". Victory is generally defined as either bypassing an obstacle without dying, or killing a bunch of things without dying, plus some other less mechanically interesting stuff (e.g. picking up the McGuffin after winning the encounter).

In an exploitable system, you've got to ignore the corner cases. There are hugely overpowered builds and parties that rely on exploiting vulnerabilities in the system to defy the expectations it was built upon. For a non-controversial example, consider the coffee-lock who spent the past hundred years acquiring spell slots. We're just going to constrain the space we're exploring to not include those extremes.

Variance in the E side of PvE make comprehensive computation impossible. Even just in the MM, there are ways to tailor encounters that make otherwise weak parties strong and otherwise strong parties cry (or even have no chance). And nothing stops a DM from creating a noncombat encounter solvable by one party and unsolvable by another (e.g. You need Thieves Cant). 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.

Non-adventuring can't be called better adventuring. At the extreme, simply deciding that you are fine with the status quo / the bad guys winning immediately solves every adventure. The more moderate version, while still less prevalent than in 3.P or 2e, is just screwing off, doing something other than adventuring, and using those resources to win. For example, throwing hirelings at everything. I would also include any strategy that relies on getting more than one long rest to deal with one "adventuring day" worth of encounters as the mild version of this issue. Permanently bypassing a challenge is fine though.

Obviously, adventuring is a terrible, sub-optimal idea. Nobody should do it, which is why nobody does it IRL, which is why we are so entertained by fictional characters doing it. The point of the game is to get that entertainment by having our PCs do this objectively dumb thing.

So you've got to apply a cost function of "time spent not adventuring" to the party's victory chance.

You've got to get there first a party's high power level at level 20 is meaningless if they all died at level 1. You've got to discount later success by earlier failure. The meaning here can be more obvious if you think of it in terms of days. If on adventuring day 1 you have a 50% chance of success, and adventuring day 2 you have a 95% chance of success, your overall chance of succeeding on day 2 is actually 47.5%.

If you can only control the build of a single PC, you have to evaluate your contribution to the group acknowledging synergy and redundancies, which is best through the lens of Shapely Values. You can't actually compute any of the numbers we're talking about, but you can approximate.

"When my PC is alone, how well would they do?"
"When my PC is added to each other PC alone, how much value does he add?"
"When my PC is added to each combination of two, three.... [Party Size -1] PCs, how much value does he add?"

What are the answers for every other PC?

Divide the total score by your individual score to get the best sense of your contribution to the group's overall success rate.

---

My personal take on it is that when you do this, you get something closer to that Chinese forum guide than the typical western tier guide. A lore bard + an ancients paladin is really, really solid.

Spitballing, but I think the best single-class parties for every number of PCs, 1-6, is probably:

1) Lore Bard: Skills, Spells, Initiative, and at least you have light armor.
2) Lore Bard + Ancients Paladin: You have someone who can tank, hit, and heal
3) Ancients Paladin + Arcane Trickster + Divine Soul: AT covers skills, allowing a switch to a more highly supportive full caster
4) Sentinel Forge Cleric + Sharpshooter BM Fighter + Arcane Trickster + Wizard: The Forge cleric takes up tanking duties from the paladin, and divine casting from the sorcerer, freeing up a slot for a higher sustained DPR Fighter, and a more versatile wizard (for when you really need that one obscure spell)
5) Sentinel Forge Cleric + Sharpshooter BM Fighter + Arcane Trickster + Wizard + Celestial Chainlock: Most critically, the warlock doesn't detract from anyone else doing their job. They add acceptable DPR, backstop healing, and the ability to get more spellcasting done in a day by upping the number of short rests, covering days where, for whatever reason, you need to do more casting than the Cleric, Wizard, and AT can handle.
6) Sentinel Forge Cleric + Sharpshooter BM Fighter + Arcane Trickster + Wizard + Celestial Chainlock + Divine Soul: The DS gets reintroduced here because of their economy of scale with a larger party.

[MaxWilson]

OK, fair point, I'll edit that. Bounding the problem by identifying constraints seems to me an important piece of "Step 0" before we can have proposals for party optimization.

I think we're not that far away from actually starting. People can start submitting example parties, while noting which constraints they satisfy and which gameplay assumptions they are making, and show how to use those parties in a game.

In some ways this dovetails nicely with @da newt's various tactical challenges.

Restating my premises more clearly:

The goal of all optimization is to maximize the chance of victory. Unlike in other cooperative ventures (e.g. business) there is no such thing as "winning more". Victory is generally defined as either bypassing an obstacle without dying, or killing a bunch of things without dying, plus some other less mechanically interesting stuff (e.g. picking up the McGuffin after winning the encounter).

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.