Racial Level Adjustment - Bad Idea or Poorly Implemented?

[Remuko]

While I think this has probably been addressed else where, I would argue the simplest fix is to just hard-set ECL to CR, as PC classes are explicitly defined as having CR=HD. Then you have to separate CR/Power-level from Wealth level, by saying that PC's WBL is that of a PC of their ECL minus the CR of their monstrous race. E.g. An Erinyes (CR8) with 2 levels of Warblade would be ECL/CR 10, and have gained 30Kgp wealth (13K across the ECL/CR 8>9 transition, and 17K across the ECL/CR 9>10 transition).

I’d go with very good idea, very badly executed.

I will absolutely say that if the game will not allow me to play monster PCs, it will be a huge factor in me not wanting to play that system. For example one of the reasons PF2 is such a pile of garbage is that their race system functionally prohibits nonstandard races without building an entire set of feats.

I also think that some of the complaints about it in 3.5 are overstated. Like it is difficult to balance 1-20. That’s just not an issue in any game I play. I very rarely see play over 15, and that rarely in games that started at level 1. And if it doesn’t balance 3-15, well, neither do fighter and wizard and we make those work. Balancing a centaur fighter just isn’t all that much harder than balancing an orc fighter or monk. I mean if we are shooting for T3 it could be as easy as tweaking the combat numbers and then dropping some fancy magic horseshoes when you notice them lagging.

My biggest complaint was the artificial LA inflation to discourage non standard races. There absolutely are good, playable monster races. Marrulurk for example is perfectly decent in many Skillmonkey builds. But they are so rare as to be more an accident than design. But the LA reassignment project is fantastic and proves the concept to my satisfaction.

i agree pretty much entirely with these two comments. (especially that first one as anyone in the LA reassignment threads would know)

[RandomPeasant]

Balancing a centaur fighter just isn’t all that much harder than balancing an orc fighter or monk.

But that's not really the hard part. Centaur HD are basically Fighter levels already. The Centaur fights with a weapon in physical combat, Fighters fight with weapons in physical combat. The hard part is balancing things where the monster does not flow directly into the class. Like a Stone Giant Wizard or a Minotaur Truenamer. You could probably massage LA into something vaguely workable for those kinds of simple transitions (though I would argue that if your proposal is anything along the lines of "CR=EL", you've already given up on the notion of LA). But fixing more complicated things requires you to solve the multicaster problem, and that is a problem that is as yet unsolved.

[ThanatosZero]

I recommend to make 4 pools of XP

One for classes, one for Racial HDs, one for Level Adjustments and one to count all the XP together to determine the Effective Character Level.

The Class level and RHD pools uses the normal amount of XP needed for a increase in ECL.

Spoiler: Normal Experience Chart

Show

1st 0 XP
2nd 1,000 XP
3rd 3,000 XP
4th 6,000 XP
5th 10,000 XP
6th 15,000 XP
7th 21,000 XP
8th 28,000 XP
9th 36,000 XP
10th 45,000 XP
11th 55,000 XP
12th 66,000 XP
13th 78,000 XP
14th 91,000 XP
15th 105,000 XP
16th 120,000 XP
17th 136,000 XP
18th 153,000 XP
19th 171,000 XP
20th 190,000 XP

The LA Pool uses this XP chart.

Spoiler: Experience Chart for LA

Show

+1 1,000 XP
+2 3,000 XP
+3 6,000 XP
+4 10,000 XP
+5 15,000 XP
+6 21,000 XP
+7 28,000 XP
+8 36,000 XP
+9 45,000 XP
+10 55,000 XP
+11 66,000 XP
+12 78,000 XP
+13 91,000 XP
+14 105,000 XP
+15 120,000 XP
+16 136,000 XP
+17 153,000 XP
+18 171,000 XP
+19 190,000 XP
+20 210,000 XP

Let us take a Astral Deva for example.

They have 12 Outsider HDs and a Level Adjustment of +8.
Normally this makes them a ECL 20 Character.

But using the alternative method, the ECL will become lower.

Astral Deva
RHD 12 (66,000 XP)/LA +8 (36,000 XP) = ECL 14 (102,000 XP)
BAB 12, HP 12d8 + CON Modifier

TEOUltimus suggested, that for a creature to gain access to class levels, they need to earn the effective amount of XP needed for a normal level up +1,000 XP, by only taking the current ECL into account.
In this case for ECL 13 they need 14,000 XP, for to enable the class level pool.
[(ECL+1) x 1000 XP = Total XP needed]

Note: The XP collected for to activate the pool, doesn't count towards the ECL pool.


Rule of thumb using this system
The ECL determines how many HDs for Hitpoints may be active at any given time (always take the best). Also it limits maximum BAB, Skillranks in a skill, Spells per day and known, Caster and Initiator Levels, and so on with the exception of saves and new skillpoints with each level up in a class.


After earning 14,000 XP
Cleric 1 (0 XP)/RHD 12 (66,000 XP)/LA +8 (36,000 XP) = ECL 14 (102,000 XP)
BAB 12, HP 1d8+12d8 + CON Modifier

Reaching Cleric 8
Cleric 8 (28,000 XP)/RHD 12 (66,000 XP)/LA +8 (36,000 XP) = ECL 16 (130,000 XP)
BAB 16, 8/20 Cleric Spellcasting, HP 4d8+12d8 + CON Modifier

Reaching Cleric 14
Cleric 14 (91,000 XP)/RHD 12 (66,000 XP)/LA +8 (36,000 XP) = ECL 20 (193,000 XP)
BAB 20, 14/20 Cleric Spellcasting, HP 8d8+12d8 + CON Modifier


As for inherent spellcasting, these are normally covered by LA and RHD, but also limited by ECL.
If either inherent spellcasting or abilities are still too strong later in comparrison with normal PCs in the sessions, I recommend the DM to increase LA as they see fit.

A alternative is to treat paragon classes as special classes, which go towards the RHD instead and thus the RHD pool. Therefore humans and any other normal PC class can become as powerful as monster PCs, to even out the playing field.

If you have no RHD and have class levels, you need to activate the RHD pool, as done with the class level pool for characters without class levels.
[(ECL+1) x 1000 XP = Total XP needed]

(I will copy later these revisions into my homebrew thread. You can find the link to it in my signature)

[Gnaeus]

But that's not really the hard part. Centaur HD are basically Fighter levels already. The Centaur fights with a weapon in physical combat, Fighters fight with weapons in physical combat. The hard part is balancing things where the monster does not flow directly into the class. Like a Stone Giant Wizard or a Minotaur Truenamer. You could probably massage LA into something vaguely workable for those kinds of simple transitions (though I would argue that if your proposal is anything along the lines of "CR=EL", you've already given up on the notion of LA). But fixing more complicated things requires you to solve the multicaster problem, and that is a problem that is as yet unsolved.

I don’t think that is truly necessary for concept. Why is stone giant/wizard an issue when fighter 17/wizard isn’t. Or if it is, you have a larger problem with 3.5 and multiclassing, not with nonstandard races.

It seems pretty obvious that in 3.5, if your goal is fairly optimized characters (and for many groups it isn’t) that all the elements in your build have to advance the focus of your build. “I can’t make a troll mage that plays at my tables op point without home brew” isn’t actually a worse problem or even a different problem than “I can’t make a Barbarian 8/sorcerer that plays at my tables op point without homebrew”. You could play an aranea sorcerer (if you adjusted the LA) or a marrulurk rogue and be on expected power level.

3.5 and to a greater extent Dreamscarred Press also had some multiclass friendly casting options. You can make a stone giant initiator or Akashic class without undue hardship.

[RandomPeasant]

I don’t think that is truly necessary for concept. Why is stone giant/wizard an issue when fighter 17/wizard isn’t. Or if it is, you have a larger problem with 3.5 and multiclassing, not with nonstandard races.

Yes. 3e multiclassing does not work for that sort of thing (that's what I mean by "multi-caster problem"). The issue is not that LA is a separate problem, it's that LA locks you into having that problem. You're right that Stone Giant Wizard isn't going to be the optimal way to build a Wizard. But if your paradigm is LA, it's not even going to be a playable way to build a Wizard. It's the difference between Truenamer v Wizard and Sorcerer v Wizard. In the abstract, it's true that they're both class imbalances, but the latter is far easier to make work in practice.

[Gnaeus]

Yes. 3e multiclassing does not work for that sort of thing (that's what I mean by "multi-caster problem"). The issue is not that LA is a separate problem, it's that LA locks you into having that problem.

I think the problems of 3.5 are pretty set in stone. It has been out of print for quite a while. So since your issue with LA is actually that it is in 3.5, please feel free to go write a different game.

You're right that Stone Giant Wizard isn't going to be the optimal way to build a Wizard. But if your paradigm is LA, it's not even going to be a playable way to build a Wizard. It's the difference between Truenamer v Wizard and Sorcerer v Wizard. In the abstract, it's true that they're both class imbalances, but the latter is far easier to make work in practice.

1. It isn’t a playable way to build a wizard at your table. Assuming that you tweaked Stone Giant to be playable at your table AT ALL (by reducing LA or improving what you get to balance with other stuff at table) wizard 1 still gives more utility than most choices.

2. Your position seems to be that all classes should be playable by all races and at the same balance point. That doesn’t seem at all viable. There are other games where I could play a giant, or a wizard, but I can’t think of any game where I could play an anything wizard and have that be as good AS A WIZARD as a race that makes good wizards. I could build a GURPS giant wizard. But he would spend so many points being a giant that he would be handicapped as a wizard compared to a human wizard. Of course, like in 3.5, he would have the advantages of being a giant. Being a giant is better in combat than being a human. If you could be as good a wizard as a giant as you can as a human, all players would play giant wizards.

3. Again, we have the tools to make giant caster types, just not giant vancian caster types. I can make a stone giant mystic (a PoW initiator class) and fly and throw lightning bolts and dispel magic and teleport and do all kinds of magic stuff.

If all races were equal at all classes, that would be a bug not a feature. If your design goal were implemented, you would get a worse RPG.

[Melcar]

I've seen a lot of people on this forum complaining about level adjustment, and treating any creature with more than LA +1 or so as pretty much unplayable unless you can do something to get rid of your LA at some point.

Do you think the idea of level adjustment is just plain bad? Or was it implemented poorly (ie assigning the wrong LA to races)?

How would you recommend balancing things when one PC's racial features are substantially stronger than another's?

I think that LA is just plain when applied to different creatures, but that it might have a role when applied to templates!

To balance things out, you have racial HD... So a racial HD 7 creature with 1 level is equal to a level 8 PC... that balances things out nicely!

[RandomPeasant]

I think the problems of 3.5 are pretty set in stone. It has been out of print for quite a while. So since your issue with LA is actually that it is in 3.5, please feel free to go write a different game.

The question posed by the thread is "is LA a bad idea or just poorly implemented". I question what you think the point of the thread is if not to talk about whether 3e's design decisions are problematic or not.

1. It isn’t a playable way to build a wizard at your table. Assuming that you tweaked Stone Giant to be playable at your table AT ALL (by reducing LA or improving what you get to balance with other stuff at table) wizard 1 still gives more utility than most choices.

What an utterly facile argument. Commoner is a playable class, it just isn't playable "at your table". Some things are bad. Beyond that, Stone Giant Wizard clearly fails at being a Wizard, even if you want to argue that some people are playing at a low enough power level that it's viable.

2. Your position seems to be that all classes should be playable by all races and at the same balance point.

Depends what you mean by "balance point". I very clearly didn't say that all race/class combinations should be at the same exact power level, but generally when we talk about balance points we mean a range. The range of things that you could mean by "Wizard" is quite large. A core-only Evoker is a Wizard. A BFC/Utility-focused Master Specialist Conjurer is a Wizard. An Incantatrix with a laundry list of personal buffs is a Wizard. And the range of things that can reasonably hang with Wizards is also large. A Sorcerer or Dread Necromancer or Favored Soul can contribute to a party with a Wizard, and might even be MVP if they happened to be played by someone more skilled or if the challenges broke in favor of their character's skillset. But we can still acknowledge that those classes aren't as abstractly powerful as the Wizard.

It's entirely reasonable to design a system where the downside of being a Stone Giant Wizard vanishes into the variance that is always going to exist as a result of build choices, encounter design, player skill, DM pity, and the thousand other things that vary between games. That's a totally obtainable design goal. But it's not a design goal you can achieve if you marry yourself to LA, because LA is a bad idea, not merely something that was implemented poorly.

[ThanatosZero]

Hm... A stone giant wizard.
Let me see what can be done with the homebrew variant rule.
(For context: https://forums.giantitp.com/showsing...6&postcount=33)

Stone giant RHD 14 (Giant) and LA+4
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 14 (101,000 XP)
10 BAB, HP 14d8 +Con Modifier

After earning 15,000 XP
Class 1 (Wizard 1) (0 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 14 (101,000 XP)
10 BAB, 1/20 Wizard Spellcasting, HP 14d8 +Con Modifier

Reaching 7th level
Class 7 (Wizard 2/Abjurant Champion 5) (21,000 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 16 (122,000 XP)
16 BAB, CL 16, 7/20 Wizard Spellcasting, HP 5d10+11d8 +Con Modifier

Reaching 14th level
Class 14 (Wizard 2/Abjurant Champion 5/Dragonslayer 1/Spellsword 1/Sacred Exorcist 5) (91,000 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 20 (192,000 XP)
20 BAB, CL 20, 14/20 Wizard Spellcasting, HP 6d10+14d8 +Con Modifier

No 9th level spells, but access to 7th level arcane spells.

What do you think? Too strong or acceptable compared to a Militia Wizard 5/Knight Phantom 10/Abjurant Champion 5?

[ThanatosZero]

Arent you forgetting the wizard 1d4 HD?

Not at all. That is why I added the link on my post for context.

Rule of thumb using this system
The ECL determines how many HDs for Hitpoints may be active at any given time (always take the best). Also it limits maximum BAB, Skillranks in a skill, Spells per day and known, Caster and Initiator Levels, and so on with the exception of saves and new skillpoints with each level up in a class.

[Psyren]

That said if you're going to keep stats as they exist in D&D, getting rid of racial stat mods entirely like they do now in 5e is dumb.

Isn't that rule optional? (And yeah I know, technically every rule is optional, but I'm pretty sure that one got labelled especially as such.)

[BioCharge]

Here's an idea that struck me: what if we use some form of Pathfinder's Simple Class Templates to make monster's more playable? We'd need to make a template for each base class, sure, but that seems infinitely easier than adjusting every monster printed.

We apply the template (with CR adjustment), adjust the hit dice up or down to equal the CR and then call it a day. This may be more viable with Pathfinder base classes having more features than their 3.5 counterparts, and thus the abilities lost would, theoretically, be replaced by whatever abilities the monster has.

Given, it doesn't solve the problem of high CR monsters for low-level play, but I honestly think that's laregly impossible to solve without designing a whole monster class, as has already been discussed. It also doesn't mesh well with multiclassing before the monster's hit dice (i.e. a 10 hit die creature can't start with, say, 5 wizard/5 rogue) or prestige classes before then. But it could be a decent ish start?

[Remuko]

Hm... A stone giant wizard.
Let me see what can be done with the homebrew variant rule.
(For context: https://forums.giantitp.com/showsing...6&postcount=33)

Stone giant RHD 14 (Giant) and LA+4
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 14 (101,000 XP)
10 BAB, HP 14d8 +Con Modifier

After earning 15,000 XP
Class 1 (Wizard 1) (0 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 14 (101,000 XP)
10 BAB, 1/20 Wizard Spellcasting, HP 14d8 +Con Modifier

Reaching 7th level
Class 7 (Wizard 2/Abjurant Champion 5) (21,000 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 16 (122,000 XP)
16 BAB, CL 16, 7/20 Wizard Spellcasting, HP 5d10+11d8 +Con Modifier

Reaching 14th level
Class 14 (Wizard 2/Abjurant Champion 5/Dragonslayer 1/Spellsword 1/Sacred Exorcist 5) (91,000 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 20 (192,000 XP)
20 BAB, CL 20, 14/20 Wizard Spellcasting, HP 6d10+14d8 +Con Modifier

No 9th level spells, but access to 7th level arcane spells.

What do you think? Too strong or acceptable compared to a Militia Wizard 5/Knight Phantom 10/Abjurant Champion 5?

that doesnt actually seem that bad. its clearly not as good as a wizard 20 in base rules but its not a fighter 20 either or the WotC Giant + LA levels of bad. I'd say this shows promise.

[RandomPeasant]

Stone giant RHD 14 (Giant) and LA+4
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 14 (101,000 XP)
10 BAB, HP 14d8 +Con Modifier

Right off the bat that seems unworkable. If your system outputs that a CR 8 monster is appropriate as a level 14 character, your system is broken. Especially if that monster is a dumb melee bruiser.

After earning 15,000 XP
Class 1 (Wizard 1) (0 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 14 (101,000 XP)
10 BAB, 1/20 Wizard Spellcasting, HP 14d8 +Con Modifier

So you gained a level but are the same ECL? That seems almost definitionally broken. Beyond that, this is a pretty crappy Wizard. You're almost exclusively reliant on your giant-based combat abilities, with your spellcasting being a curiosity at best, even if those combat abilities are adequate to your level. If we grant that this is a playable character that is both a Stone Giant and a Wizard, it would still seem like a failure to me if my goal was to play something like one of the Gigantes from A Practical Guide to Evil (a reasonable reference point for "magic-user giant", I think).

Reaching 7th level
Class 7 (Wizard 2/Abjurant Champion 5) (21,000 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 16 (122,000 XP)
16 BAB, CL 16, 7/20 Wizard Spellcasting, HP 5d10+11d8 +Con Modifier

This seems... really bad. As an ECL 16 character, a 7th level Wizard would be a trash mob for you, relying on that kind of casting is unlikely to be adequate.

Reaching 14th level
Class 14 (Wizard 2/Abjurant Champion 5/Dragonslayer 1/Spellsword 1/Sacred Exorcist 5) (91,000 XP)/
RHD 14 (Giant) (91,000 XP)/
LA +4 (10,000 XP)
= ECL 20 (192,000 XP)
20 BAB, CL 20, 14/20 Wizard Spellcasting, HP 6d10+14d8 +Con Modifier

That's probably the best you're doing relatively speaking, but I don't really care what results any system outputs at 20th level. People have been able to cast wish, shapechange, and gate with their spell slots for four levels at this point, if the game is still functional what's being played is so far removed from the written rules that it's not really meaningful.

What do you think? Too strong or acceptable compared to a Militia Wizard 5/Knight Phantom 10/Abjurant Champion 5?

It's definitely not too strong. My alternative proposal would be something like this:

Stone Giant
Minimum Level: 6
* Your 1st level feat is "Large Size".
* Your 3rd level feat is "Giant Racial Stat Boosts".
* Your 6th level feat is "Stone Giant Racial Stat Boosts".
* Rock Throwing
* Rock Catching
* Maybe a WBL penalty if you don't think the feats balance all the stat boosts

That seems way more playable as a Wizard, Rogue, or anything other than a front-line melee class than any kind of LA-based muckery. It's not, despite what Gnaeus seems to think I want, an optimal choice for a Wizard, but it's going to produce a Wizard that casts level-appropriate spells, and that's going to be acceptable in the overwhelming majority of games. A Stone Giant Wizard built in this way is also going to feel like a Wizard in a way that one that tries to layer 1st level Wizard casting on top of an 8th (let alone 14th) level character will not.

[Morphic tide]

Right off the bat that seems unworkable. If your system outputs that a CR 8 monster is appropriate as a level 14 character, your system is broken. Especially if that monster is a dumb melee bruiser.

Monster math stacks with PC math. Monsters build in the numbers of magic items to compete against PCs who actually have them on relatively even footing. This is why LA needs to be a thing, because any "Big Dumb Melee Brusier" at its CR has approximately double the bonuses over a base humanoid of a normal PC Big Dumb Melee Bruiser. If not considerably more from equalizing damage with raw Strength.

So you gained a level but are the same ECL? That seems almost definitionally broken. Beyond that, this is a pretty crappy Wizard. You're almost exclusively reliant on your giant-based combat abilities, with your spellcasting being a curiosity at best, even if those combat abilities are adequate to your level. If we grant that this is a playable character that is both a Stone Giant and a Wizard, it would still seem like a failure to me if my goal was to play something like one of the Gigantes from A Practical Guide to Evil (a reasonable reference point for "magic-user giant", I think).

Yeah, that's a big mess covered in abuse, particularly given the thing above about over-stacking bonuses. The Aranea is bragging about its +6 Cha with CL=RHD, and Huge monsters of all stripes show vicious contempt for the normal races with their massive Natural Armor and wide-ranging damage amplification.

This seems... really bad. As an ECL 16 character, a 7th level Wizard would be a trash mob for you, relying on that kind of casting is unlikely to be adequate.

You have at worst the melee abilities of an 8th-level Martial, and more likely something like a 12th-level Martial if you can get appropriate gear because you're then just flat adding all the items of a Marital on top of a Stone Giant. And 7th-level Gish spellcasting is quite the sizable boon to an already passable Martial when it's at the cost of but two levels. Plenty enough low-level buffs on Wizard to make a viable Gish, and the framework being suggested is horrifyingly efficient at such shenanigans.

That's probably the best you're doing relatively speaking, but I don't really care what results any system outputs at 20th level. People have been able to cast wish, shapechange, and gate with their spell slots for four levels at this point, if the game is still functional what's being played is so far removed from the written rules that it's not really meaningful.

Three levels, not four, and viciously demolishing the insane game-shattering peaks of casters in practice is a very old time-honored tradition of GMs. Any situation where that comes up is a situation not worth designing for because, as you've mentioned, the game doesn't function with those particular things in play.

It's definitely not too strong. My alternative proposal would be something like this:

Stone Giant
Minimum Level: 6
* Your 1st level feat is "Large Size".
* Your 3rd level feat is "Giant Racial Stat Boosts".
* Your 6th level feat is "Stone Giant Racial Stat Boosts".
* Rock Throwing
* Rock Catching
* Maybe a WBL penalty if you don't think the feats balance all the stat boosts

That seems way more playable as a Wizard, Rogue, or anything other than a front-line melee class than any kind of LA-based muckery. It's not, despite what Gnaeus seems to think I want, an optimal choice for a Wizard, but it's going to produce a Wizard that casts level-appropriate spells, and that's going to be acceptable in the overwhelming majority of games. A Stone Giant Wizard built in this way is also going to feel like a Wizard in a way that one that tries to layer 1st level Wizard casting on top of an 8th (let alone 14th) level character will not.

But then you're not actually playing a Stone Giant in any meaningful capacity, you have a thin emulation of a Stone Giant. Your statistics have literally nothing to do with the monster. This is, to my understanding, a major complaint with the way PF handles Polymorph effects, because you don't get to actually be the creature in question, you get to borrow a bare handful of properties that vaguely suggest the creature. And worse, this becomes something that must be defined for every monster separately rather than being a genericized mass like the PF Polymorph.

Some form of level adjustment system is required if PCs and monsters get their numbers in ways that combine, simply because the monsters are going to end up at an advantage when you then add PC characteristics to them.

[Dante & Vergil]

Wasn't there people on these forums that had created monster classes from the ground up? I remember them being well received, as they did away with LAs.

[Silly Name]

LA has always struck me as a very weird situation where the designers didn't want to let certain creatures be PCs, yet gave the players rules to technically make them PCs. Except they suck, so you either don't bother with those rules and make your own, or you never play those unusual creatures.

If you don't want players to have access to a thing, just don't give it to them.

Something with 14 RHD is already pretty bad because unless you play exactly in the creature's strengths, there's just too little you can do so with few levels open, so you're very much stuck in playing those creatures to type: melee bruisers will remain melee bruisers, casters will remain casters, roguish assassins will stay roguish assassins. Sticking LA on top of this is just silly.

And most games don't start at that high a level in the first place anyway, which gates potential cool concepts from the get-go.

[sreservoir]

Yes. 3e multiclassing does not work for that sort of thing (that's what I mean by "multi-caster problem"). The issue is not that LA is a separate problem, it's that LA locks you into having that problem. You're right that Stone Giant Wizard isn't going to be the optimal way to build a Wizard. But if your paradigm is LA, it's not even going to be a playable way to build a Wizard. It's the difference between Truenamer v Wizard and Sorcerer v Wizard. In the abstract, it's true that they're both class imbalances, but the latter is far easier to make work in practice.

If you start your build with Fighter 18, are you going to build a "playable" Wizard?

There are two problems here. One is that an awful lot of LAs are assigned "conservatively": they're assigned something needlessly high, not because they're consciously balanced against that many levels, but basically to throw a number on there that's intentionally high enough to deter players from taking it. That's mostly an issue of, well, not really wanting players to play certain races, but someone along the way feeling obliged to fit them into the rules that allow that anyway? It's a self-contradictory design-by-committee probably, probably.

The other problem is that the value of a single class level isn't uniform, and in particular a lot of classes have abilities placed at class level X that are expected to be effective around character level X. Worse, there are then other classes whose core features basically scale linearly, without no allowances for what other characters of the same level are doing. And the former are the more powerful options.

[Crake]

I recommend to make 4 pools of XP

One for classes, one for Racial HDs, one for Level Adjustments and one to count all the XP together to determine the Effective Character Level.

The Class level and RHD pools uses the normal amount of XP needed for a increase in ECL.

Spoiler: Normal Experience Chart

Show

1st 0 XP
2nd 1,000 XP
3rd 3,000 XP
4th 6,000 XP
5th 10,000 XP
6th 15,000 XP
7th 21,000 XP
8th 28,000 XP
9th 36,000 XP
10th 45,000 XP
11th 55,000 XP
12th 66,000 XP
13th 78,000 XP
14th 91,000 XP
15th 105,000 XP
16th 120,000 XP
17th 136,000 XP
18th 153,000 XP
19th 171,000 XP
20th 190,000 XP

The LA Pool uses this XP chart.

Spoiler: Experience Chart for LA

Show

+1 1,000 XP
+2 3,000 XP
+3 6,000 XP
+4 10,000 XP
+5 15,000 XP
+6 21,000 XP
+7 28,000 XP
+8 36,000 XP
+9 45,000 XP
+10 55,000 XP
+11 66,000 XP
+12 78,000 XP
+13 91,000 XP
+14 105,000 XP
+15 120,000 XP
+16 136,000 XP
+17 153,000 XP
+18 171,000 XP
+19 190,000 XP
+20 210,000 XP

Let us take a Astral Deva for example.

They have 12 Outsider HDs and a Level Adjustment of +8.
Normally this makes them a ECL 20 Character.

But using the alternative method, the ECL will become lower.

Astral Deva
RHD 12 (66,000 XP)/LA +8 (36,000 XP) = ECL 14 (102,000 XP)
BAB 12, HP 12d8 + CON Modifier

TEOUltimus suggested, that for a creature to gain access to class levels, they need to earn the effective amount of XP needed for a normal level up +1,000 XP, by only taking the current ECL into account.
In this case for ECL 13 they need 14,000 XP, for to enable the class level pool.
[(ECL+1) x 1000 XP = Total XP needed]

Note: The XP collected for to activate the pool, doesn't count towards the ECL pool.


Rule of thumb using this system
The ECL determines how many HDs for Hitpoints may be active at any given time (always take the best). Also it limits maximum BAB, Skillranks in a skill, Spells per day and known, Caster and Initiator Levels, and so on with the exception of saves and new skillpoints with each level up in a class.


After earning 14,000 XP
Cleric 1 (0 XP)/RHD 12 (66,000 XP)/LA +8 (36,000 XP) = ECL 14 (102,000 XP)
BAB 12, HP 1d8+12d8 + CON Modifier

Reaching Cleric 8
Cleric 8 (28,000 XP)/RHD 12 (66,000 XP)/LA +8 (36,000 XP) = ECL 16 (130,000 XP)
BAB 16, 8/20 Cleric Spellcasting, HP 4d8+12d8 + CON Modifier

Reaching Cleric 14
Cleric 14 (91,000 XP)/RHD 12 (66,000 XP)/LA +8 (36,000 XP) = ECL 20 (193,000 XP)
BAB 20, 14/20 Cleric Spellcasting, HP 8d8+12d8 + CON Modifier


As for inherent spellcasting, these are normally covered by LA and RHD, but also limited by ECL.
If either inherent spellcasting or abilities are still too strong later in comparrison with normal PCs in the sessions, I recommend the DM to increase LA as they see fit.

A alternative is to treat paragon classes as special classes, which go towards the RHD instead and thus the RHD pool. Therefore humans and any other normal PC class can become as powerful as monster PCs, to even out the playing field.

If you have no RHD and have class levels, you need to activate the RHD pool, as done with the class level pool for characters without class levels.
[(ECL+1) x 1000 XP = Total XP needed]

(I will copy later these revisions into my homebrew thread)

This is actually pretty similar to how I run LA in my games. I basically run an optional gestalt system, and LA/RHD always go on the gestalt side. The xp costs are flat and don't affect the progression of your main gestalt side, but your xp values are totaled to determine your ECL. That way you don't nerf the xp costs of future levels, and isntead only pay a flat xp cost for the LA you do have.

I also use this as an alternative to multiclassing, so you can have a Wizard 17//Fighter 16 fighting alongside a Wizard 20 (as the most basic example, im sure people can come up with better multiclasses)

[RandomPeasant]

Monster math stacks with PC math. Monsters build in the numbers of magic items to compete against PCs who actually have them on relatively even footing. This is why LA needs to be a thing, because any "Big Dumb Melee Brusier" at its CR has approximately double the bonuses over a base humanoid of a normal PC Big Dumb Melee Bruiser. If not considerably more from equalizing damage with raw Strength.

I agree that you can't simply give someone the racial bonuses of a Stone Giant or Couatl for free. That would pretty obviously be broken. But I don't buy the conclusion that you need LA to balance things. Because (as I said in the very first post in this thread) the value of a level isn't constant. The difference between 5th level and 6th level is not the same as the difference between 6th level and 7th level, let alone 15th level and 16th level.

LA can't work. It just mathematically can't. If you're buying fixed benefits, you have to buy them with fixed costs. That means either you need monsters to smoothly transition into PC classes (I am skeptical that this is possible) or you need monster abilities to be bought with something other than levels (feats or WBL are good choices, and the latter addresses your concerns quite directly). Or, I suppose, you need to go to a model where the value of a level is supposed to be linear. But if your system maintains 3e's exponential level scaling. LA is mathematically unworkable even before you consider how well it works for any particular build.

Three levels, not four

I was counting the current level in my math.

But then you're not actually playing a Stone Giant in any meaningful capacity, you have a thin emulation of a Stone Giant.

That depends what you think the Stone Giant is. The MM entries for Orcs or Goblins or Kobolds come with a level of Warrior by default. If you had to keep that to play an Orc Barbarian or a Goblin Rogue or a Kobold Sorcerer, those characters would be much worse than they are using the normal rules. But no one thinks that you aren't really playing a Kobold when you give up that level of Warrior. I view the Stone Giant's racial hit dice in the same way. Ideally, the system would be written in such a way that the Stone Giant wouldn't have any racial hit dice, and would simply be an 8th level Warrior with the Stone Giant race, in the same way that generic Orc warriors are represented by the Warrior class rather than racial hit dice.

And worse, this becomes something that must be defined for every monster separately rather than being a genericized mass like the PF Polymorph.

That's going to be true for whatever monsters-as-PCs system you choose. There's simply no general rule that can cover all the ways for a monster to be. Consider the Medusa or the Bodak. They get most of their power from a passive ability, making PC classes a much greater upgrade for them than for Hill Giants or Couatls whose active abilities trade off with a Warblade's Maneuvers or a Warlock's Invocations. And a system like this wouldn't be all downside. It would make it much easier to generate advanced monsters, as they could simply use the PC/NPC advancement rules without any modification.

If you start your build with Fighter 18, are you going to build a "playable" Wizard?

If the system is going to present Fighter 18/Wizard 1 and Fighter 19 as equal choices, they would be. It's not clear to me why imbalance here is any more acceptable than class imbalance, which people are rightly upset about.

The other problem is that the value of a single class level isn't uniform

This is the thing that sinks LA as a concept more than anything else. The difference between the 7th level of Wizard and the 8th level of Wizard is a 3rd level spell slot and a 4th level spell slot, plus some stuff that Wizards get at every level. The difference between the 14th level of Wizard and the 15th level of Wizard is a 5th level spell slot, access to 8th level spells at all, an 8th level spell slot, and a bonus feat, plus the same stuff that Wizards get at every level. There's no +1 LA race or template in the world that's balanced against both of those things.

[Morphic tide]

I agree that you can't simply give someone the racial bonuses of a Stone Giant or Couatl for free. That would pretty obviously be broken. But I don't buy the conclusion that you need LA to balance things. Because (as I said in the very first post in this thread) the value of a level isn't constant. The difference between 5th level and 6th level is not the same as the difference between 6th level and 7th level, let alone 15th level and 16th level.

LA can't work. It just mathematically can't. If you're buying fixed benefits, you have to buy them with fixed costs. That means either you need monsters to smoothly transition into PC classes (I am skeptical that this is possible) or you need monster abilities to be bought with something other than levels (feats or WBL are good choices, and the latter addresses your concerns quite directly). Or, I suppose, you need to go to a model where the value of a level is supposed to be linear. But if your system maintains 3e's exponential level scaling. LA is mathematically unworkable even before you consider how well it works for any particular build.

Not in the Martial space. From a powergaming "Cannot Lose 9ths" perspective maybe, but the vast majority of Martial routes have enormously more limited value from further levels. At 11 RHD you still have room for the premier Grappler PRC's big selling point (Black Blood Cultist 8 for every Natural Attack's damage on every successful Grapple check). Size bonuses are obscene value propositions to such builds, making them legitimately viable where they aren't previously.

I have run the numbers quite a few times on Psychic Warrior 4, for quite a number of monsters, and it is rather consistent that the only point monster math doesn't make a solidly competitive Natural Attack PsyWar is when you start considering Form of Doom and Combat Transformation at level 16, and even then it remains a genuine question because you can run through a lot more fights as the monster than the Humanoid. And are virtually always and forever a vastly better Grappler, because the reason Grapplers are bad is because monster-math is better at it than PC-math.

My own point is specifically that monsters already aren't paying the costs. Because monsters are not meant to be paying such costs. You can't balance it out with WBL because item costs are exponential. Do you judge the +6 Strength by its low-level 4k GP, or the fact it saves you 20k GP if you stick to the +4 Belt of Giant's Strength? Do you judge +2 Natural Armor as 8k or 32k? What about the lunatic +20s to ability scores or double-digit natural armor values? The overall system is not constructed in a way conductive to directly using monster statblocks for PCs, because monster statblocks operate differently in very basic fashions, because monsters are built from very blunt raw numbers and not bundles of magic items that mesh together to generate large numbers.

LA costs you BAB, skill points, HP, and delays feats. BAB is directly compensated by the enormous Strength bonuses and size increases, skill points can be exceeded with point-buy by taking the top 2 off an 18 or "dumping" Constitution to jump Intelligence for more skill ranks for every class level you do get, the HP is made up for by Natural Armor turning into AC advantage and the Con bonuses, and many, many monsters have a Special Attack or Special Quality far more valuable than a single feat unless you're desperately digging for the spectacular outliers. LA is a fixed cost that can have its impact measured to derive the point where something is better or worse than an example Human.

That depends what you think the Stone Giant is. The MM entries for Orcs or Goblins or Kobolds come with a level of Warrior by default. If you had to keep that to play an Orc Barbarian or a Goblin Rogue or a Kobold Sorcerer, those characters would be much worse than they are using the normal rules. But no one thinks that you aren't really playing a Kobold when you give up that level of Warrior. I view the Stone Giant's racial hit dice in the same way. Ideally, the system would be written in such a way that the Stone Giant wouldn't have any racial hit dice, and would simply be an 8th level Warrior with the Stone Giant race, in the same way that generic Orc warriors are represented by the Warrior class rather than racial hit dice.

But that isn't how monsters are constructed. Stone Giants are not Orcs or Goblins or Kobolds. They have a "for player characters" entry as a formality, not something intended to be used. Monsters are built as statblocks that challenge PCs directly without requiring itemized equipment. I have run numbers for a by the book Awakened Bear in this very thread to show that it quite readily dumpsters a Water Orc Barbarian for versatility. The pressures are fundamentally different, you can't properly construct the vast majority of theoretically-playable monsters as a racial statblock with so many class levels, because they do huge numbers of things there aren't classes for and have to compete with kitted-out PCs on the average aray of 10s and 11s.

In literally every single space but caster-likes, including Incarnum and certain applications of Initiators and Psychic Warrior, the vast majority of monsters can do very ridiculously more than a standard PC race, because they can trade a small part of their enormous bonuses for a huge chunk of versatility. Taking the top + off half a dozen of your magic items is a huge amount saved for much more niche functionalities to make up for lacking class features, where you aren't getting very expensive stuff like Huge size or Good-maneuverability Flight. If monster statblocks are for average members of a kind of creature and simultaneously meaningfully challenge PCs, they will always end up able to wildly diverge from PC build constraints, because they're competing with PCs without using PC bonuses.

For a by-the-book Stone Giant to be usable as a CR 8 monster for the whole party to go to town on, it has to be competing with a very unusually durable level 8 Barbarian with the standard array of 10s and 11s, no magic items, and the bluntest feats imaginable. If it were 8 RHD, it would become wildly strictly superior to a regular PC, because it has numbers to compete with magic items, that stack with magic items. Orcs, Goblins, and Kobolds, meanwhile, are all PC races. Their default statblock carries class levels, that are specifically mentioned as replaceable. They have no level adjustment. They have distinct advantages and disadvantages to other normal races, even if such is slanted against them. They even have specific splatbook support. None of this is true of any Giant in particular, and incidental PC-friendly type support can be mostly traced to the Goliath or Half-Giant, or is shenanigans like Alter Self.

To not need Level Adjustment, you must have monster-math actually be directly equivalent to PC-math. With exponential cost to improve magic items stacking with racial bonuses, this cannot be the case. Basic functions of how characters differentiate early and scale later on result in the problem I mention. You'd need to carefully weigh extra RHD to match the value of WBL and the Elite array, and doing that for Dungeons and Dragons revised third edition is simply logistically impossible. The way the product design was handled could not have allowed such a delicate measure of functionality. The way the game is constructed is not conductive to any consistent translation. You have to radically rewrite every technically-playable monster in the game to have them operate on PC math. They're just too different to have a clean framework for playability without lopping off levels to give a chance to catch up on raw numbers.

[Psyren]

"Transition monsters into PC classes" is pretty much what LA buyoff is meant to simulate. The monster LA starts out higher when their abilities are more valuable than class levels, and over time erodes as this balance shifts in favor of the greater value higher level class features provide.

The PF1 Monsters as PCs system is that as well, but with the buyoff being automatic/built into the progression.

[RandomPeasant]

Not in the Martial space. From a powergaming "Cannot Lose 9ths" perspective maybe, but the vast majority of Martial routes have enormously more limited value from further levels. At 11 RHD you still have room for the premier Grappler PRC's big selling point (Black Blood Cultist 8 for every Natural Attack's damage on every successful Grapple check). Size bonuses are obscene value propositions to such builds, making them legitimately viable where they aren't previously.

I agree that martial builds are underpowered, and martial classes offer insufficient returns at high levels. Changing the game so that those characters end up more viable is a good thing. I agree that aesthetically it may be weird if all martial characters are Ettins or Ogres or Storm Giants, but if that produces melee combatants who are reasonably effective, it's not mechanically problematic.

But that isn't how monsters are constructed. Stone Giants are not Orcs or Goblins or Kobolds. They have a "for player characters" entry as a formality, not something intended to be used.

Yes, I agree that the monsters-as-PCs rules are broken by design. I am proposing that they should instead be written in a way that is not broken by design so that monsters can be playable as PCs instead of that not being true. Again, the question posed by the thread is "is LA a bad design choice", and my view is that the answer to that question is "yes" because it locks you into a bunch of decisions that make monsters unworkable as PCs.

you can't properly construct the vast majority of theoretically-playable monsters as a racial statblock with so many class levels, because they do huge numbers of things there aren't classes for and have to compete with kitted-out PCs on the average aray of 10s and 11s.

It seems like you're making a couple of unnecessary assumptions there. First, who says that monsters and PCs have to use different basic stat arrays? Second, your argument would benefit from an example of a monster that A) is reasonably playable as a PC and B) doesn't work as a PC class. Third, while it's true that the math behind monsters differs from the math behind PCs, much of that is a result of contingent design decisions -- there's no reason a Giant needs to have average BAB and an outsized pile of HP and STR.

For a by-the-book Stone Giant to be usable as a CR 8 monster for the whole party to go to town on, it has to be competing with a very unusually durable level 8 Barbarian with the standard array of 10s and 11s, no magic items, and the bluntest feats imaginable.

I think you're overestimating the amount of challenge a CR 8 monster is supposed to be for an 8th level party.

If it were 8 RHD, it would become wildly strictly superior to a regular PC, because it has numbers to compete with magic items, that stack with magic items.

Who says those numbers have to stack with magic items? It's called a Belt of Giant Strength because it used to literally give you the strength of a giant.

and doing that for Dungeons and Dragons revised third edition is simply logistically impossible.

Well, sure. But so is every other solution. There isn't any easy fix in the context of the rules we have now. The only way to make monsters-as-PCs work in the system we currently have is to look at the monsters your players want to play and create bespoke homebrew to make it work. That scales horribly, but it's the only solution that works at all in the 3e rules framework. And maybe the way you choose to create your bespoke homebrew is based on LA. If you hammer at it enough in the context of the one to four specific characters it is going to need to work for in whatever limited level range your campaign covers, I'm sure you can get something that is no more problematic than Greg wanting to play a Monk while Sally wants to play a Druid. But there's no general solution.

"Transition monsters into PC classes" is pretty much what LA buyoff is meant to simulate. The monster LA starts out higher when their abilities are more valuable than class levels, and over time erodes as this balance shifts in favor of the greater value higher level class features provide.

The PF1 Monsters as PCs system is that as well, but with the buyoff being automatic/built into the progression.

LA buyoff has problems of its own, as it makes LA power-for-nothing in the long run. This is less bad than monster PCs being flatly unplayable, but such a system can clearly be improved upon. Last I checked (which was admittedly a while ago, perhaps this has been fixed), PF monsters-as-PCs had the same problem, with it being a suckers bet not to pick a synergistic monster as the base for your character if you expected the game to run long enough. LA is too punishing of a cost, but there does need to be some cost.

[HouseRules]

The maximum amount of automatic "LA payoff" in PF1 is half CR rounded down.

PF1 Monsters as NPC buys at a rate of 3 class levels per 2 ECL.
PF1 Monsters as PC buys at a rate of 4 class levels per 3 ECL.

[Necroticplague]

It was a bad idea that was then poorly implemented.

I think that something along the line's of PF's 'create a race' system should be used in place for all PCs. And as you level up, you get more points to spend in it. Monster abilities merely have point costs and prerequisite, you can pick them up using these points. This creates an effective minimum level to be a full-fledged member of a race (when you'd have enough points for all their abilities), while allowing play to be technically available from any level, or to easily make variants (who have merely spent their points differently).

[Morphic tide]

I agree that martial builds are underpowered, and martial classes offer insufficient returns at high levels. Changing the game so that those characters end up more viable is a good thing. I agree that aesthetically it may be weird if all martial characters are Ettins or Ogres or Storm Giants, but if that produces melee combatants who are reasonably effective, it's not mechanically problematic.

Can we stop moaning about spellcaster supremacy? Again, it is a time-honored tradition for DMs to navigate this problem, it's been there from the start, there are loads of remarkably low impact methods that cut off the staggering abuses, including such things as just flat nuking the bypasses to risk and distended utility functions by doing stuff like banning the secondary Charisma check modifiers that make Planar Binding trivial and mandating the Specific Summon Monster variant. It is vastly easier to clear out the tools that render casters absurd than it is to rebuild literally the entire rest of the system to make those outliers few players can actually execute in practice "balanced".

Because those caster tools are well-analyzed and catalogued to tone down piece by piece, whereas the phenomenon of monster-math is constantly swept under the rug unless one speaks of why Transmutation is overpowered or why Grappling is terrible. I have participated in the LA reassignment thread, I am literally the only person who habitually looks at direct build comparisons to actually try to measure monster-math. Nobody else goes at the numbers of how much PP a PsyWar is burning to keep up. Nobody else is citing progression breakpoints. Not on a regular basis. I am still intensely angered by the Glaistig getting a -0. Go to the LA reassignment threads and run Advanced Search for my username from the Search Thread drop-down menu. I do it nearly every time I decided to comment on a monster. I never get hard numbers disproving me. Literally never.

I will literally cite combinations of metamagic shenanigans involving three different sourcebooks to point out a 4rhd Undead is a perfectly capable Necromancer-backing Paladin of Slaughter, irrespective of how viable such a specific character build "really" is, going so far as to specifically measure Deadly Touch pools and total spell slots. Inbuilt metamagic reduction and a significant boost to casting scores alleviates enormous swaths of the downsides to going full spellcaster, provided you actually build for what you are. You're not the big game-shaking do-everything Wizard, you're the Corpsecrafter with a bucket of metamagic to make the top-quality Undead and keep them from breaking.

Yes, I agree that the monsters-as-PCs rules are broken by design. I am proposing that they should instead be written in a way that is not broken by design so that monsters can be playable as PCs instead of that not being true. Again, the question posed by the thread is "is LA a bad design choice", and my view is that the answer to that question is "yes" because it locks you into a bunch of decisions that make monsters unworkable as PCs.

Most monsters are perfectly fine PCs with all of one or two less RHD or LA +1 or +2, and usually have an acceptable combat niche right at their statblock RHD. Level Adjustment is the answer to the design choices that make monsters poor PCs, because those design choices came first and level adjustment was the decision to resolve the issue in question. Monsters do not make good spellcasters, because monster-math virtually always needs to answer the question of how you handle fighting a party of four. Which has generally been large durability skews until 5e introduced Legendary Actions to directly address the action economy issue.

It seems like you're making a couple of unnecessary assumptions there. First, who says that monsters and PCs have to use different basic stat arrays? Second, your argument would benefit from an example of a monster that A) is reasonably playable as a PC and B) doesn't work as a PC class. Third, while it's true that the math behind monsters differs from the math behind PCs, much of that is a result of contingent design decisions -- there's no reason a Giant needs to have average BAB and an outsized pile of HP and STR.

The fact that PCs are meant to be exceptional? It's why the Elite array even exists at all? For a monster that doesn't properly work as a PC, look at the Aranea. Its racial hit dice are literally Warrior gestalted on Sorcerer with a side-order of Monstrous Spider and +4 to three relevant ability scores. An inherent poisonous bite is not how the extremely vast majority of PC classes work, because PC classes are nearly always learned abilities. A Giant does need outsized Strength if it is to compete with PCs, unless you give the Giant the same Standard Magic Items to get the same resultant to-hit bonus, or painstakingly design a system to have monster-math take up the same positions as PC-math, which is an extremely large workload of overhauling the actual system itself.

The game was not designed from the start with monster PCs in mind. Monsters are not designed with being PCs in mind. Therefore, playability of monsters must be an add-on, and level adjustment works with this question very bluntly by directly asking what level of PC a monster is roughly equivalent to as a question of comparing raw outputs. It is the natural answer, and the most straightforward, because it's a direct mathematical comparison of the value of playing a monster versus a typical Humanoid.

I think you're overestimating the amount of challenge a CR 8 monster is supposed to be for an 8th level party.

No, it's that a Big Dumb Bruiser monster has to have the raw meat to survive two or three rounds so it's an actual fight, which sets the design towards the durability end of the spectrum. How often do you see a Raging Barbarian who has more health than a same-CR melee beater Giant? How often do you see a Giant dealing more damage than that Barbarian? PCs tend to be glass cannons because there's a lot more damage content than defense content. Psychic Warriors accidentally have the tools to equalize with monster-math at staggering PP costs.

Who says those numbers have to stack with magic items? It's called a Belt of Giant Strength because it used to literally give you the strength of a giant.

Yes it used to, as a set to functionality, and this returned in 5e. This makes the character's base stats largely meaningless, with Orcs being actually disadvantaged at higher levels (even setting aside the level caps) because they payed a hefty opportunity cost for bonuses that completely stopped doing anything at all. The reason they end up stacking is to have the racial differences persist instead of going in the trash the moment the ability score items arrive. It's a result of commonplace magic items, an extremely fundamental premise of 3rd edition functionality, coexisting with your race always keeping its minor influences.

Well, sure. But so is every other solution. There isn't any easy fix in the context of the rules we have now. The only way to make monsters-as-PCs work in the system we currently have is to look at the monsters your players want to play and create bespoke homebrew to make it work. That scales horribly, but it's the only solution that works at all in the 3e rules framework. And maybe the way you choose to create your bespoke homebrew is based on LA. If you hammer at it enough in the context of the one to four specific characters it is going to need to work for in whatever limited level range your campaign covers, I'm sure you can get something that is no more problematic than Greg wanting to play a Monk while Sally wants to play a Druid. But there's no general solution.

LA is the easy solution. It's not rewriting huge chunks of the game to overhaul its assumptions, it's not making enormous piles of "technically-this-monster" content, it's looking at the monster, going over the numbers to see how many levels ahead or behind a PC in the same niche it is, and declaring its effective ECL in light of how its capabilities affect a PC. It is literally "What level PC does this most reasonably equal?"

Oh, and funny thing about the Druid/Monk comparison, the Monk is actually almost functional all over the place, but is cursed by the way the system splits its needed values among ability scores and having bad frontloading to compensate for its gear restrictions. The commonplace wide bonuses of monsters, including size increases and Natural Armor, do spectacular wonders to Monk viability. There are many monsters that'd love to be a straight-up Monk. Because everything Monks need to be viable is commonplace monster-math.

[Psyren]

LA buyoff has problems of its own, as it makes LA power-for-nothing in the long run. This is less bad than monster PCs being flatly unplayable, but such a system can clearly be improved upon. Last I checked (which was admittedly a while ago, perhaps this has been fixed), PF monsters-as-PCs had the same problem, with it being a suckers bet not to pick a synergistic monster as the base for your character if you expected the game to run long enough. LA is too punishing of a cost, but there does need to be some cost.

The maximum amount of automatic "LA payoff" in PF1 is half CR rounded down.

PF1 Monsters as NPC buys at a rate of 3 class levels per 2 ECL.
PF1 Monsters as PC buys at a rate of 4 class levels per 3 ECL.

As HouseRules said, the buyoff is capped, so there is still a tradeoff for many monsters. I agree it's not perfect however, e.g. if you pick a Lillend Bard nothing keeps them from getting full progression, so the GM still has to wade in, but it's a better starting point than both 3.5 and "no."

[RandomPeasant]

Monsters do not make good spellcasters, because monster-math virtually always needs to answer the question of how you handle fighting a party of four.

Monsters don't make good spellcasters because monsters that fight "like spellcasters" arbitrarily have mechanics that do not scale or transition neatly into any class. A Beholder is a caster. A Mind Flayer is a caster. Many fiends are casters (though obviously it varies a lot because there are many, many types of fiend). The idea that monsters can't fight like a Wizard or a Cleric is just not supported by the range of monsters that exist.

The fact that PCs are meant to be exceptional? It's why the Elite array even exists at all?

PCs are exceptional because they are higher level than other people. The fact that your 10th level Warlock has marginally better base stats than a 1st level Commoner is not the reason he's awesome, it's the fact that he can fly, shoots blasts of eldritch power at his enemies, and calls on dark magics to do his bidding.

An inherent poisonous bite is not how the extremely vast majority of PC classes work, because PC classes are nearly always learned abilities.

So your big example of "asymmetry between PCs and NPCs" is "what if you had a poisoned dagger that couldn't be sundered or disarmed, but also couldn't be upgraded to a mace or a sword"? If that's the problem we need to solve, I think we're okay.

The game was not designed from the start with monster PCs in mind. Monsters are not designed with being PCs in mind. Therefore, playability of monsters must be an add-on,

Therefore monster PCs are unplayable. They don't work in RAW 3e. They don't work if you tweak LA numbers. They don't work if you let people buy off small amounts of LA. They don't work if you do complicated XP accounting. All you can do in the 3e rules framework is get something that works okay for your game. There is no general fix, there are only bespoke fixes for the N~=4 version of the problem that happens at your table.

LA is the easy solution.

LA isn't a solution because it doesn't work. Monsters are universally unplayable for builds that do not directly synergize with them, and often unplayable anyway. There are monsters out there where the level at which LA says they are playable as PCs is higher than the level where ECL says they are no longer a challenge for PCs. It's true that you get something from WBL, but you're not closing that gap.

It is literally "What level PC does this most reasonably equal?"

Except that's not enough. Because PCs don't just stay one level. If you were doing a one-shot, then yeah I could see LA being workable. But people gain levels. And that changes the value of abilities they have. The pile of stats, SLAs, and random immunities that is worth being six or three or nine levels behind at 10th level is not the same as the pile that is worth being that many levels behind at 4th or 12th level.

As HouseRules said, the buyoff is capped, so there is still a tradeoff for many monsters. I agree it's not perfect however, e.g. if you pick a Lillend Bard nothing keeps them from getting full progression, so the GM still has to wade in, but it's a better starting point than both 3.5 and "no."

That's about what I remembered. I'm not saying you can't do better than the existing system, but there are structural problems with LA that mean any system will need that sort of DM intervention to a far higher degree than if you had a system that was good. That doesn't mean you need to write that system to have a functional game, but it's pretty obvious to me that a good (rather than merely "kind of okay, if the DM tunes it") monsters-as-PCs system would look absolutely nothing like LA.

[sreservoir]

The fact that PCs are meant to be exceptional? It's why the Elite array even exists at all? For a monster that doesn't properly work as a PC, look at the Aranea. Its racial hit dice are literally Warrior gestalted on Sorcerer with a side-order of Monstrous Spider and +4 to three relevant ability scores. An inherent poisonous bite is not how the extremely vast majority of PC classes work, because PC classes are nearly always learned abilities. A Giant does need outsized Strength if it is to compete with PCs, unless you give the Giant the same Standard Magic Items to get the same resultant to-hit bonus, or painstakingly design a system to have monster-math take up the same positions as PC-math, which is an extremely large workload of overhauling the actual system itself.

Tragically, they hit one something very close to a solution to both this and the NPC wealth problem midway through 3.5e's run, in the form of Incarnum ... and then wrote it up a self-contained niche subsystem with like three reverse-deps in later books.

[RandomPeasant]

Tragically, they hit one something very close to a solution to both this and the NPC wealth problem midway through 3.5e's run, in the form of Incarnum ... and then wrote it up a self-contained niche subsystem with like three reverse-deps in later books.

I don't think Incarnum really solved the problems, because while they made some noise about soulmelds conflicting with magic items, you could mostly avoid that if you dumpster-dived hard enough or used the optional rules for combining magic items from the MIC.

The big problem is, again, that 3e made a bad design choice: WBL. When 3.0 launched, WBL was an interesting experiment. But looking at it with the benefit of 20 years of playing the game, it seems pretty clear that random magic items was a better system. There are problems (particularly with the AD&D implementation), but overall they are lesser than the problems with WBL, and the constraints it puts on the game aren't worth the benefits it provides (incidentally, this is something you can fix with only minimal effort -- just put everyone on something like PF's ABP for their math fixes and hand out random magic items that do things players might care about as treasure).

There's a secondary problem where the magic item slots system rapidly became nonsensical, with slots proliferating in number, losing any meaningful identity, and slotless magic items making the whole thing a cruel joke. If you want to cap how many magic items people can have, just cap how many magic items they can have.