D&D 5th Edition: Thread #7

[Clawhound]

I sort attack spells into these categories:

• Damage
• Death
• Environment

The important point to note that a "death" spell is any spell that takes out a character or opponent for this battle. For example, "hold person" is a death spell. A character can find themselves out of the entire battle. For them, there's no difference between being held and being dead. By observing the unhappiness of players, I eventually decided that I did not like how "death" spells worked.

In most cases, especially for 3E, a spell only needs to sideline a character for 1-2 rounds to be powerful.

If I am going to use powerful "death" magic, like petrification, then projecting that as information to the characters works best. I want them anxious and paranoid going into a deadly situation, but I also want them to choose that encounter.

[Water_Bear]

The thing is, Save or Die effects are important because they're part of the fantasy stories D&D is based on. Whether Perseus and the Medusa or Harry and Voldemort, the idea of magic so powerful it can kill you in one shot is just too cool not to use. Plus, it's at least as *Iconic* as the Alignment System and a ton of other controversial things they're bringing over.

I'd say the ideal SoD/SoL is something where you get a precaution and a Save. So, instead of Finger of Death, you get a Gaze attack like the Medusa's. Or for Phantasmal Killer, changing it to follow standard rules for illusion disbelief. Or just letting people hold their breath when a Cloudkill passes over. That way careful players will have more recourse to avoid getting one-shotted, while spell-casters have to make actual tactical decisions rather than trying to track everyone's HP totals.

[TuggyNE]

The thing is, Save or Die effects are important because they're part of the fantasy stories D&D is based on. Whether Perseus and the Medusa or Harry and Voldemort, the idea of magic so powerful it can kill you in one shot is just too cool not to use. Plus, it's at least as *Iconic* as the Alignment System and a ton of other controversial things they're bringing over.

Strongly agreed. Essentially, the fluff is too compelling to ignore, and the mechanics just need to follow in some reasonably-usable form.

Or just letting people hold their breath when a Cloudkill passes over.

I've never been able to understand why this was specifically disallowed in 3.x (except perhaps that breath-holding was so easy anyone would do it?).

[PairO'Dice Lost]

I've never been able to understand why this was specifically disallowed in 3.x (except perhaps that breath-holding was so easy anyone would do it?).

Mechanically, probably because it only has a radius of 20 feet and moves at 10 feet per round, which already makes it incredibly easy to escape, so if breath-holding were allowed you couldn't even guarantee one round's worth of effects and if you're going to use a 5th-level 20-foot radius spell to kill things instead of a 3rd-level 20-foot radius spell to deal 9d6 damage (which will probably kill a lot of 6 HD critters anyway) there's got to be a benefit.

Flavor-wise, probably because it's basically mustard gas, chlorine gas, or another chemical weapon (inspired by WW2 wargames the same way fireballs and lightning bolts are based on Chainmail catapults and ballistae), which can get in through your eyes, mouth, and even skin pores, so holding your breath wouldn't negate it--in fact, the Fort save for 4-6 HD critters can easily represent the difference between taking a lungful of it (and thus choking to death) and holding your breath (and merely suffering extreme blistering and irritation).

[Ashdate]

I can only imagine players creating a "SOP" for when they enter new rooms. They slowly add to it as they discover more and more counter measures to spells, forcing character sheets to have a space for declaring when a character stops taking extreme measures on the off-chance that they walk into the room with an illusionary Medusa who can cast Cloudkill.

[TuggyNE]

Flavor-wise, probably because it's basically mustard gas, chlorine gas, or another chemical weapon (inspired by WW2 wargames the same way fireballs and lightning bolts are based on Chainmail catapults and ballistae), which can get in through your eyes, mouth, and even skin pores, so holding your breath wouldn't negate it--in fact, the Fort save for 4-6 HD critters can easily represent the difference between taking a lungful of it (and thus choking to death) and holding your breath (and merely suffering extreme blistering and irritation).

Hmm, fair enough, I can buy that.

[invinible]

I think the way the Pokémon series handles their one hit ko moves is the way that Dungeons & Dragons should handle their Save or Lose moves.

[ThiagoMartell]

I think the way the Pokémon series handles their one hit ko moves is the way that Dungeons & Dragons should handle their Save or Lose moves.

So you mean save or lose abilities should be banned anywhere it matters?

[Stubbazubba]

I don't think Pokemon is a good model. In Pokemon, when you lose a Pokemon to a one-hit KO, you keep playing at a disadvantage. In D&D, when you lose a character to a one-hit KO, you stop playing. Until one player controls multiple characters in D&D, Pokemon is not a very useful comparison.

[PairO'Dice Lost]

I don't think Pokemon is a good model. In Pokemon, when you lose a Pokemon to a one-hit KO, you keep playing at a disadvantage. In D&D, when you lose a character to a one-hit KO, you stop playing. Until one player controls multiple characters in D&D, Pokemon is not a very useful comparison.

I think the point he was going for was that in Pokemon it's hard to raise the accuracy of the one-hit KOs much higher than 30%, and they just don't work on higher-level targets. I don't like balancing SoL to only work on lower-level enemies, since that essentially reserves it for PC use against mooks (when basic blasting spells would work better anyway) or BBEG use against PCs (because they tend to be party level +2 or higher) which basically defeats the purpose.

[Stubbazubba]

I think the point he was going for was that in Pokemon it's hard to raise the accuracy of the one-hit KOs much higher than 30%, and they just don't work on higher-level targets. I don't like balancing SoL to only work on lower-level enemies, since that essentially reserves it for PC use against mooks (when basic blasting spells would work better anyway) or BBEG use against PCs (because they tend to be party level +2 or higher) which basically defeats the purpose.

Agreed, but that doesn't solve the root problem that SoLs are uninteresting for the target, regardless of the odds and regardless of how fun it is for the user. See my post before that one.

[Saph]

Agreed, but that doesn't solve the root problem that SoLs are uninteresting for the target, regardless of the odds and regardless of how fun it is for the user.

The problem with this line of reasoning is that if you keep on following it long enough then you end up with the conclusion that you shouldn't do anything harmful to a PC.

After all, getting killed by hit point damage isn't "interesting" for the target, either – it stops them acting just as permanently as a SoD does, right? If you have 30 HP, a 50-damage fireball is a Save Or Die by definition. So this would mean that we'd have to ban any kind of attack that has the potential to do more damage in one turn than the PC's hitpoints.

But you can achieve the same effect by having multiple creatures focus-fire a target. If the same 30 HP character is targeted by five orcs that all do an average of 10 HP per attack, then you're effectively playing the Save-Or-Die game all over again, just with a different probability curve. (In fact, the PC has even less input in this situation, since they aren't even the one rolling the dice – they just get told how much damage they take.)

This takes a PC out of the battle without them being able to do anything about it. This, as we've previously agreed, isn't fun, so we're going to have to ban this too. Maybe put in a rule saying no more than two enemies are allowed to attack the PC in one turn.

But the monsters can still produce the same effect by attacking a PC over multiple turns. In ideal conditions a PC can avoid this by getting healed, or running away – but there are lots of ways to prevent a PC from being supported or from moving. This leads to a situation where the PC gets locked down and slowly killed over multiple turns. Since this isn't fun, either, we're going to have to get rid of effects that split up PCs or stop them moving, which means battlefield control is out of our game, too . . .

. . . and you can see where this is going.

The thing is, as a DM, I can see how the "you shouldn't take players out of the game" point of view makes complete sense. You DO want players as engaged and involved as possible, and it IS a bad idea to take away a player's agency when there's a way around it. However, you've got a case of conflicting priorities here – if you want to play a combat game, then there's a natural tendency to have sudden and violent death be a potential outcome, because well, that's what happens in combat. You can deal with that by introducing "Death is Cheap" rules, but that brings its own problems.

The way I see it, if you're going for the following three things:

• You're playing a game that revolves around lethal combat where both sides are trying to kill the other.
• Death/serious injury is scary and meaningful.
• Players can't be taken out of the game to any significant degree.

. . . then it's "pick two out of three". You can't have all of them.

If you delete the first, you get a non-combat game. Combat is deadly and has major consequences, and so people avoid it. Rare in RPGs, because gamers like killing stuff.

If you delete the second, you get "Death Is Cheap". Characters go down fast and they get up fast. This is the default setting in comic books, Descent, and high-level 3.5 games (where it's quite common to have multiple deaths and resurrections in a single combat).

If you delete the third, you get the "old-school" approach. This is the default setting in survival horror, Warhammer, and in older versions of D&D where even a high-level character could get permanently killed from one mistake or piece of bad luck. High risk, but exciting.

I think the arguments over SoDs mostly come down to whether you prefer Type Two games or Type Threes.

[1337 b4k4]

I can only imagine players creating a "SOP" for when they enter new rooms. They slowly add to it as they discover more and more counter measures to spells, forcing character sheets to have a space for declaring when a character stops taking extreme measures on the off-chance that they walk into the room with an illusionary Medusa who can cast Cloudkill.

Or alternatively, your DM could not abuse the tools he's given. I mean I can one shot a player without SoD effects, I'll just put an invisible monster mage 20 levels above the party in the room with damage and magic reduction and a sword of +20 vs Humans, Elves and Dwarves. SoDs are tools, just like any other tool in your DMs kit and they can be used and abused for both good and evil. That they are a complicated tool that requires careful application is not a good reason to not have them at all.

In D&D, when you lose a character to a one-hit KO, you stop playing. Until one player controls multiple characters in D&D, Pokemon is not a very useful comparison.

Until? Heck, up to about 2e, that was the default. You had a retinue of characters, not to mention the men at arms that were in your party. Your PC goes down and someone is getting a field promotion ("Minion 1, I choose you!").

[Friv]

The way I see it, if you're going for the following three things:

• You're playing a game that revolves around lethal combat where both sides are trying to kill the other.
• Death/serious injury is scary and meaningful.
• Players can't be taken out of the game to any significant degree.

. . . then it's "pick two out of three". You can't have all of them.

I disagree with your premise, and with the slippery slope that you presented.

The complaint about SoD effects is not that players can be taken out of the game to a significant degree. It is that players can be taken out of the game in a single action, and that can last for a long period of real-world time.

If large groups of enemies have to gang up on someone to take them out of action, the question becomes how they got themselves into a tactical position for every enemy in play to be ganging up on a single guy to that degree. If the DM is deliberately arranging things for that to be a reasonable tactic, yeah, it's probably not a good plan.

If enemies take several actions to take a character entirely out of play, that's fine. That is, in fact, working as intended.

If players can die, that's also fine, although if death is going to be common I would suggest that players have some kind of reasonably available back-up for the post-fight so that they can get back into play quickly. Similarly, SoD is less of a problem

The goal is to avoid having bored players sitting at the table for long stretches of time. Reducing the impact of single-hit kills is one way to do this.

(Another way to do this involve giving players narrative control over situations that is not dependent on their character's consciousness, but that's never been somewhere D&D intended to go. A more common D&D approach is to just give access to various sorts of minions to every player, which helps the SoD problem but introduces complexity and confusion issues.)

[Clawhound]

SoD also have some connection to Con one-shot games, where getting taken out is presumed. In that context, I'm just fine with one shotting a character out of the game. Hopefully it was a truly gruesome death that you get to brag about.

[Bacon Elemental]

Don't forget the Squishy Wizard rule. Here's a scenario.

Bob the wizard is engaged, along with his party in a large fight, when he separated from the group. After fending off a few goblins, he's at about two thirds HP. To his horror, he sees a gigantic troll charging at him, club larger than his whole body raised to strike. He has only one turn to take action before it reaches him, so he uses Epic Finger Blast - a Save or Die spell. Now, there are two ways this could turn out.
1) The troll fails its save and is instantly killed with a single blast from his finger, while he turns his head and smokes a cigar. It's still-moving form crashes to the ground just inches from his feet. He is the god of badass.

2) The troll makes the save, and continues it's enraged rush. Bob barely had time to scream before the club swing smashes the life out of his body, draining all his hit points. The other heroes cry out, and the monster is soon slain in revenge for bob's life. Bill, his player, is pissed off.

[PairO'Dice Lost]

Don't forget the Squishy Wizard rule. Here's a scenario.

There are two other scenarios if Bob instead casts epic burnination, a non-SoL spell:

3) The troll fails its save, takes 120% of its HP in damage, and is instantly killed with a single blast of fire, while he turns his head and smokes a cigar. It's still-moving form starts to disintegrate as it slides forward, and it collapses to ash just inches from his feet. He is the god of badass.

4) The troll makes the save for half damage, and though it is covered in copious third-degree burns, it continues its enraged rush. Bob barely has time to scream before the club swing smashes the life out of his body, draining all his hit points. The other heroes cry out, and the monster is soon slain in revenge for Bob's life. Bill, his player, is pissed off.

And yet another scenario if Bob casts anywhere but here, a teleportation spell:

5) The troll doesn't roll a save at all and continues its enraged rush. Bob barely has time to scream before he disappears in a pop of displaced air and the club swings through the space his body just occupied. The other heroes cry out at the near miss, and will rib him later for screaming like a girl when he was in danger. Bill is relieved at the narrow escape.


If Bob is separated from his party, weakened, and within charge range of a troll, whether an offensive spell is a SoL is irrelevant: the Squishy Wizard will be happy if his spell is powerful enough to take out a threat on a failed save, and will be pissed off if it isn't powerful enough on a successful saving throw to save him from his tactical errors. It's good to provide the wizard with ways to save himself after he screws up, but SoL spells are not the best candidate for that.

[Icewraith]

Yet another situation where Grease or Wall of Force has a more reliable effect.

The real problem with powerful death effects, is it's awesome when it works for you and sucks when it doesn't, and it when it's used against you you're one bad roll away from eating the rest of the chips while your buddies finish up the combat, and depending on the amount of effort you put into making your character (and the power level you're playing at) you've either

(I started rambling)

Spoiler

Show

1: just seen a ton of research go down the drain,

2: the suspension of disbelief is about to be shattered as with the red dragon lying dead at the remainder of the party's feet, Not-Bob the rogue (identical twin brother of bob the rogue) rides up with a message for his brother, who is now a smoldering pile of ash. He is accompanied by Grunty the greatsword wielding dwarf fighter/barbarian, who Not-Bob hired as muscle for the dangerous road. Grunty happens to be the second cousin of Snorri the greataxe wielding dwarf fighter barbarian, whose corpse is slowly cooling on various parts of the battlefield. His beard is red instead of black... and shut up guys, at least I came up with a better name than Not-Bob! Do we have any more chips? Can I loot what's left of my old character?

3: you frantically roll up a new character, try to come up with a backstory, and buy ALL the +saving throw gear you can afford before too much time has elapsed so you can play again. In your mind, you weigh the probability of your DM finally letting you play a half-dragon werepanther with the improved level adjustment rules you worked out last week, vs him making you reroll a more reasonable character and having to avoid advancing the next part of the plot so the party's not undermanned when they run into the next encounter, which he's had planned for two weeks....

4: Your dm hands you a sheet which, while it might have a perfectly good character, you had no input on and now have to figure out how the hell this thing is supposed to work, and now have to either work with a backstory someone else made OR figure out a backstory and personality on the fly that matches the sheet you've just been handed. Also playing your own character is a big part of the fun in these games, that's why character death sucks so much

5: You had a stack of character sheets of appropriate level and semi-filled out setting appropriate backstory for just such an emergency, which requires an excessive amount of planning before play actually begins

6: You happen to have your character from another game with you, but he's four levels higher than the rest of the party and just visited ye olde magic shoppe. Prepare for Wolverine syndrome as you now have the same character in different games at the same time and you try to keep the level, gear and history of the character relatively coherent.

If this sort of thing keeps up you start making redundant copies of character concepts that you can swap into games in the hope that this time, one of them will save vs. finger of death. You have your dice exorcised and if the pattern still continues, you destroy them so the remains can never be used to randomly generate rolls again and, with a heart full of hope, buy a new set of dice and hope for the best.

[Stubbazubba]

-snip-

Excellent strawman.

Did no one read my post at the bottom of the last page? Here it is again, spoilered for length;

Spoiler

Show

Extra Credits did an episode on a relevant topic a few weeks ago, "Counter Play."

The main thrust of the idea is this:

When designing an ability or a mechanic, you can't only be thinking about how to make that ability or mechanic interesting for the player who gets to use it, you also have to think about how its interesting for the players its used on. And on a more rigorous level, it's the idea that a mechanic or ability in a multiplayer game should increase the number of meaningful choices available both to the player using it and the player its being used on.

TTRPGs are not considered multiplayer games, but the psychology and importance of this principle is true because at the combat round level, they function exactly like one; the DM is one player controlling a single monster on any given turn (mostly), and the player is controlling their one character, and they are slinging these abilities back and forth in a way that is essentially indistinguishable from a competitive multiplayer game.

EC goes on to make the point that abilities that are an interesting tactical option for the user but not for the target is a good way to create frustrated targets. However, when you consider both sides of that equation, you create a richer play experience for both. So the question of whether or not SoDs are cool for the SoD-user is not the only consideration we have to take into account when designing SoDs. We also have to account for the SoD receiver's experience and what options SoDs provide to them. Obviously, the only tactical implication of a traditional SoD for a target is "jack up that save modifier in your build!" That is one-dimensional (it's not really a choice if it's the only way) and irrelevant in combat (the decision is made outside of combat and nothing in combat will change it). This is not an enriching option as-is.

So SoDs need to be counter-able by the party, whether that's by beginning an SoD at the end of one turn and then casting on the next where taking any damage in-between either negates or greatly diminishes its effect if cast, SoDs only working on targets below a certain HP threshold, or something else that gives the opposing party/character an actual tactical option it can take in the midst of combat to attempt to prevent or counter it.

A third consideration for these mechanics in a TTRPG, I would say, is how it interacts with the user's allies. You want abilities that interlock with the roles/actions of others, and gives them interesting options on their turns, too. The mundane half of the party's contribution to the battle can't be meaningless with one successful SoD. The mundanes have to contribute to SoDs somehow, whether that's as simple as protecting the caster from having their concentration broken during casting, or contributing to meeting the necessary HP threshold for the spell to work, or some other combination of tactics.

Also, giving mundanes SoD abilities certainly couldn't hurt, either. At some point a rogue should be able to just sneak up and stab a guy through the heart, and the fighter should be able to cut off the monster's head with one mighty blow, so long as those have tactically interesting mechanics backing them up.

So when I say 'interesting for the target,' I don't mean harmless, I mean tactically engaging at all. SoDs don't create any tactical options for the target whatsoever, they are simply uncounterable.

When you die to hit point damage it's because of a series of tactical choices made by both you and the DM, and in the end, the DM's choices served him better, but who wins or loses is based on tactical choices. Hit point damage is a tactically-rich option because any hit point-depleting maneuver has ways of being mitigated or countered, even if its just after-the-fact healing (even if its as simple as "do I fight defensively for a bit? Do I try to get my allies' help? Do I run away?"). (And yes, if a DM is throwing a 50 damage Fireball at a 30 HP PC, that doesn't sound very level-appropriate to me. If anything can take you from full to 0, it had better take more than one turn to do so.) SoDs, on the other hand, are tactic-less; it is the result of one choice and nothing the other side has done or can do affects the choice or the result, and therefore, it's uninteresting for them to even be in a fight where SoDs are on the table, not just after they get hit by one.

They are a completely one-sided tactic, thus they are boring because their existence doesn't influence the tactics of the other side whatsoever. SoDs don't engage anyone, they're one-press win buttons at best, which is barely acceptable in a single-player game, and given the multiplayer nature of TTRPGs they are flat-out inappropriate, regardless of how cool they may be for that one PC in the cases where they actually hit.

EDIT:

@Bacon: If Bob the Wizard is in that state, why on earth is he trying to kill the troll? He should be teleporting either himself or the troll away, or otherwise focusing on avoiding that hit, not putting all of his eggs into a SoD basket. That's a very poor tactical call, if Bob just wants to live to see the next round.

Nevertheless, it's a great example of a tactically interesting scenario; turn it around and realize that usually Bob doesn't get a turn between when the troll starts charging and when he gets hit, so this 'SoD' troll attack giving him a chance to do anything is already a step forward. So, at this point, Bob the Wizard has a tactical decision to make; try to kill the Troll before he completes the charge, or teleport away, or deal enough damage that he either dies or is mostly dead, or employ some battlefield control spell (Wall of Force/Grease), or buff his AC so the Troll has no hope of hitting, etc., etc. That, IMO, is a great example of how it should work.

[Yora]

The new Playtest package is supposed to be up sometime today.

When it comes to prestige class, we have some pretty simple design goals. We want them to represent interesting elements of the world, rather than just new mechanical options or a source of power. Prestige classes should be something that you earn membership in through your actions in the campaign world, rather than just a new set of feats or special abilities that come with prerequisites. They're another reward for a character to strive toward. Obviously, since a prestige class offers new mechanics, you can always look at it just like a list of new abilities, but that doesn't mean we have to start our design work with that approach.

Let's wait and see about that. This is exactly how Prestige Classes were supposed to be in the 3.0 DMG.

[Water_Bear]

Ugh. Prestige Classes with built in fluff requirements. Why does anyone think that's a good idea!?!

This is supposed to be a modular game right? Release that **** as modules, don't bake an idea that dumb into the core system.

[Kurald Galain]

Based on today's D&D survey, does anyone else see the problem with multiple prestige classes with synonymous names? For instance, if the handbook contains a Beastmaster and an Animal Lord, what if anything is supposed to be the difference between that? Or how is an Eldritch Knight not the same thing as a Spellsword?

[Yora]

Where is the survey anyway? That article doesn't provide a link.

[Cavelcade]

It shows up at the bottom of the page, for me anyway.

Take our survey and let us know what your favorite prestige classes are.

[Craft (Cheese)]

I see a pretty big problem with prestige classes as Mearls has written their goals behind them: They're utterly backwards.

It sounds like what they really want to do is give the player an incentive to give their character some concrete motivations, so that the DM can play off of these motivations to make more interesting situations.

A noble goal, but problem is PrCs are pretty much the worst possible design I can imagine for this. Let me count the ways:

1. Prestige classes were so problematic in 3.5 because they're quite difficult to balance. For every PrC that's reasonably balanced against just taking straight levels in your base class there's a dozen that are strictly better or strictly worse. Furthermore PrCs are often the only halfway viable way to make a certain character concept, making you more or less railroaded into them if you want to play that type of character.

2. PrCs are long term investments into your character. Too long-term. The requirements for most PrCs make them impossible to enter any earlier than level 6, some of them you have to wait as long as level 9. Yes you can use early-entry tricks but those are loopholes rather than intentional design features. Depending on where your game starts level-wise PrC entry will either have already happened or be too far away to meaningfully work into the story. Sometimes you'll hit a sweet spot but this is rather rare. Long-term goals are fine, but what's the DM to work with session-to-session? Unless the entire campaign is based around one of the players tracking down Elfslayer's descendants, it doesn't really solve the problem it's purported to.

3. So you've entered your PrC. Now what? After the quest to get entry is done, it's back to square one: The character has no concerete motivations and goals aside from "react to whatever the DM throws at you" and has no mechanical incentive to create more. The story-building function of the PrC is one-shot, which is doubly bad because of the long-term problem above. Or at least, it should be one-shot. Unless you want weird situations like "I'm bored of the Knights of Solamnia now. Time to join another faction!"

4. It's quite.... limiting. Could you imagine what types of PrCs would fit with goals like "I want to find the rare cure for my father's disease" or "I want to find a suitable husband for my sister" or "I want to spread the influence of my god's religion?" Me neither. Sure, you could make PrCs whose fluff requirement involves doing one or more of those things, but really PrCs only make sense when they represent the character coming into new types of personal power in some way. Like the Rainbow Servant PrC: You find a Coatl who teaches you the secrets of using divine magic as an arcane caster in exchange for helping them with some task. PrCs just break down when you try to apply a reason to entering them other than personal power and prestige.


I don't think PrCs in D&D Next will be anything more than a disastrous failure.

[Clawhound]

I really want to argue with you Craft (Cheese). I really really wanna wanna.

And I think that you are far, far too correct, and that makes me sad.

Some things would help prestige classes, or at least minimize their hazard. We already know how they get used and abused by character designers and that won't change.

First, limit a character to one prestige class. Promiscuous multiclassing is hard as heck to balance in design without adding prestige classing.

Second, have no requirements for the prestige class other than level. That eliminates the crazy planning which solely existed to access the prestige class. Most combination will simply not work, as they don't synergize.

With those two things, character designers can still pick well, but so could anyone else. You've made it that much easier for non-designers to create an engaging character, and that is far more important.

[Yora]

I expect the very same thing as in 3rd Edition. So just play without them.

[Craft (Cheese)]

Some things would help prestige classes, or at least minimize their hazard. We already know how they get used and abused by character designers and that won't change.

I think the best way to "fix" PrCs would be to make them work like racial substitution levels: You can trade, say, your 4th level class feature in exchange for this alternate (but balanced) feature if you meet the requirements. Except the requirement is based on meeting an in-story requirement rather than being a particular race. If you meet the requirement later, you can switch out your substitution levels retroactively, and if your DM permits you can start the game having already met the requirement as part of your backstory. So, if you're a member of the Knights of the Black Axe, you can trade some of your Fighter maneuvers to get exclusive, special maneuvers in their place, and when you level up, you can learn more of the special Black Axe maneuvers instead of taking new ones from the standard Fighter list.

It's not perfect as problems 3 and 4 still apply, but I do think it would nicely handle problems 1 and 2.

[noparlpf]

What if base classes only went up to level 5-6, maybe 8, and then you have to branch into a more specific subset of, say, "Rogue", for example, "Assassin" or "Combat Trapsmith" or whatever.

[Yora]

It would make things more complicated for people who want to play single class characters.