D&D General D&D 6e ala Steampunkette: Structural thoughts

[dave2008]

So I've talked a few times about what I'd want in a 6e game. Often enough that at 3am this morning I was plagued with thoughts of game design and couldn't sleep. So here's some of the stuff I'm looking at. Putting them in quotes to be easier to isolate and parse.

1) 10th Level Classes


2) Make Magic Users More Magical and Less Casty


3) Day 1 Psionics


4) Warlords


5) Combat Maneuvers


6) Spellcasting Mechanics Variety


7) "Extra Attack" at 3rd with Caveats


8) Crit Protection as a Core Mechanic


9) Exploration and Social Mechanics as Core


10) Sensible HP Structures


What do you think about a system like this? Interesting package of alternate rules, overengineered nonsense, or nothingburger?

I agree with a lot of your ideas, but not as much with your implementation.

1. Disagree: I think the game should support 20+ levels of play. However, I would be fine if it was parceled out like BECMI was into tiered rules / guidance.
2. Agree: My implementation would be different, but I like general idea.
3. Disagree (sort of): I generally don't like psionics and I don't want in my game. However, I could be converted and I don't mind it if is something easy to ignore.
4. Agree I guess. I don't have any issue with warlords, but I also don't think they are gone really.
5. Agree (I think): I am not sure what you are going for, but I don't maneuvers for all to take from the fighter's niche. That being said I think there is a way to make this work.
6. Agree. I would make arcane casters the only ones that cast spells. Cleric, paladins, warlocks, etc. would use magic but not cast "spells."
7. Not sure: I am thinking extra attack should be a fighter only feature (or you can buy it with a feat maybe)
8. Agee: I may have different ideas on how to do this though.
9. Agree: Not sure about the how though.
10. Agree: I would probably go about it differently, but I agree with the general idea.

 

[Steampunkette]

this could get conversations, why do I only get +1 with my greatsword and that one gets +1 for the dagger?

"Why does someone with a dagger get +4 to their damage for having high Dex?" is the same argument, in the end. But yeah. I could see that thing happen, too.

That said... I might be inclined to do the no-mod design with the +1 per point over the target's AC because not only does that sound really cool? It also results in a Goblin rolling a Crit dealing, potentially, quite a bit of damage to a player... Which makes goblins dangerous even to high end characters!

 

[Distracted DM]

"Why does someone with a dagger get +4 to their damage for having high Dex?" is the same argument, in the end. But yeah. I could see that thing happen, too.

That said... I might be inclined to do the no-mod design with the +1 per point over the target's AC because not only does that sound really cool? It also results in a Goblin rolling a Crit dealing, potentially, quite a bit of damage to a player... Which makes goblins dangerous even to high end characters!

High end melee characters end up with an AC of like.. 25 or more. So that goblin won't be threatening them, if I understand it right, but on the other hand those high end casters with 18 AC could get hit decently hard.. in fact it'd give a reason to use Shield spell against attacks that hit, if I understand the idea correctly.

 

[Faolyn]

'Nother few ideas...

1) Less Heritages/Races
Yeeeah I get it. People like having a lot of variety in their heritages. But I find it mostly winds up in making the game a neverending cavalcade of animalfolks with a few planar entries along the way.

Here you could simply do a generic animal folk. Take a look at how Ebberon's Shifters are done. Instead of doing each individual type of animal as a different type of Shifter (wolf-shifter, deer-shifter), they're divided up into rough types (beasthide, wildhunt) and the player choses one. It's then up to the player to say if their wildhunt shifter looks like a wolf or a lion or hyena. Thus, your animal folk section could be divided in similar ways.

Likewise for planar creatures. Level Up, of course, has its Planetouched, with a simple heritage gift to differentiate between infernal, celestial, or elemental. When I put out my LU Handbook of Heritages, I included gifts to allow for law and chaos variants as well, as well as for various types of demons, daemons/yugoloths, and non-angelic celestials. Also, look at how 5.24 has done it. For tieflings, you pick Abyssal, Infernal, or Chthonic, and it's done in a table. Mind, I don't think the 5.24 way is particularly flavorful, since it's literally just spells, damage resistance, and darkvision, but it's simple, and the flavor can be in the description.

Doing these races like this gives people the options they want without cluttering up the races section or inviting that, yes, inevitable deluge of new options. I know people who really want to play animal-folk as a form of self-expression, and I myself generally prefer planetouched to any other type.

And since every heritage has their own unique language, it dilutes the languages into a soup of nonsense where no one bothers to speak anything but common unless they're secretly talking trash with the one other person in the party who shares their language, or they meet an Elf King or something and the barbarian starts spouting Shakespearean English with iambic pentameter.

Languages are -fun- to play around with, when there's only a few to learn and you can be reasonably sure several of them will come up at different times in a campaign.

One idea to take from GURPS is language proficiency. As in, a difference between broken, fluent, and native levels of understanding. In GURPS, they actually required this for both speaking and reading. When I ran Ravenloft in GURPS, we had a lot of fun with characters who had lower levels of understanding.

While I wouldn't necessarily go quite that far, you could divide understanding into poor and fluent, or do speaking and reading as different things. Instead of saying that culture or background gives you X number of languages, say it gives X number of points. One point for poor understanding, or either speaking or reading; two points for fluent understanding, or both speaking and reading.

Though, to be clear, I'm just talking the core book. Individual Settings can, and should, have unique heritages for the setting. But Warforged and Changelings shouldn't become a core race on every world just because they're nifty in Eberron, you know? And I don't envy WotC's incoming attempts to bust Dark Sun wide open so everyone can play their favorite kitchen sink species....

I kinda wanna do a core set of heritages and then do the wider subraces thing more like A5e does with...

2) Cultures
Yup. Elves in the Underdark are Elves with an Underdark culture that gives them the stuff that makes them mechanically Drow. And you can also do it to Dwarves and then they'll be Duergar. And Gnomes for Svirfneblin. Etc etc etc. But also stuff like High Elf Culture for a Human to be a part of, too. Smaller number of heritages, broader quantity of cultures, more mix n' match ability.

I'm less fond of this myself, but perhaps that's because the Adventurer's Guide racial cultures just a copy of D&D's various races, and because, well, there's a lot of them. Too many. I like LU, but it's become very bloated.

Now, while I've decided I'm not a huge fan of Draw Steel (nothing wrong with it; just not for me), I will say that the way they do cultures is pretty cool and worth taking inspiration from. In that game, culture is divided up into three types: environment, organization, and upbringing, each of which gives skills, and you mix-and-match as you please. Woodland Beaurocratic Noble, for instance. With this, you could then give several sample cultures. High Elf: Woodland, Beaurocratic, typically Academic, Creative, or Noble. Wood Elf: Woodland, Beaurocratic, typically Creative, Lawless, or Martial.

And a fun thing for culture? It helps determine your lifespan. Oh, sure. A human has a life span that's shorter than an elf. But a human raised in high elf culture is gonna live a lot longer than a human raised on a farm. Similarly, an elf in a cosmopolitan city will live longer than their orcish neighbors in the same town, but not so long as the high elves of old. Call it diet, medicine, magic, whatever you like. I just think that's a fun thing.

Mmm, I'm not sure. Honestly, the only real reason to even care about lifespans in a game like this is when you had supernatural aging, like from ghosts and spells. Unless you bring that back, there's really no point to having this other than as a note for flavor purposes: people in cosmopolitan cities live longer than people in oppressive tyrannies do; elves have lifespans of over a millennia, while orcs rarely live to the age of 60. Yes, there are people who use D&D to play games that span decades or even centuries, but that's vanishingly rare.

3) Languages
And, of course, the languages will be tied to the cultures! All the humans raised in Elf Cultures speak Elven. Or maybe something like "Ellowyn" as the elven language. And then the "Common" language is a regional/national thing. Wherever your game is set, that nation's language is the 'main' language, plus your cultural language. And, hey, however many points of Int mod, more languages.

I prefer languages to be regional or based around very specific ideals rather than on race or even culture (in my current LU game, I have the city-state's language, Albastilian, and also Contract (the trade/legal language), and Tower (the magic language). Plus languages for other countries elsewhere. But the different ancestries that make up the city of Albastile don't have their own languages.

But to me, it doesn't make sense that people in different Human cultures would all speak Human. While you could potentially say that there's very little drift in Elven languages (because of their lifespan or their trance-sleep which connects them to their past), that shouldn't be the case for shorter-lived people. And if there's going to be regional Human languages, there should also be regional languages for others. But that makes for a ginormous list of languages.

That being said, if you do have "cultural tongues" and choose to rename them into things like Ellowyn, don't do what Draw Steel did and give them completely unrelated names: Dwarfs (and engineers) use Zaliac; bugbears and fey speak Khelt. Huh? That's going to be easily forgettable and then people will just default to Dwarvish and Sylvan.

One other idea that's been floating around in my head (and is up for anyone else to use, if they want it) has been to have languages based on the creator god(s). Any sentient beings made by that god or pantheon share a language; there was no "Tower of Babel" event to cause them to have different languages. The key here is that this wouldn't be like most D&D where every sentient race has a god or pantheon. You could say that one pantheon made Humans, Halflings, Ogres, Giants, and other beings based on that sort of idea or body plan, and they all speak the same language. Another one created Lizardfolk, Dragonborn, Aarakocra, Merfolk, Sahuagin, Pterafolk, and other reptile/fish/avian beings, and they all share a language. At least that way it makes sense for you to be able to communicate from someone on the completely opposite side of the world!

5) Multiclassing and Feats
Rather than being a Rogue 4, Barbarian 6, you'll just grab feats that grant you class abilities in addition to what you've got from your current class. Maybe get 5 feats over the course of leveling, and you can choose between grabbing class-feats or general feats.

It means balancing the multiclass feats against each other, and then the general feats against them... but I think it's plausible. I hope.

Maybe you should make all the class features into feats. By which I mean, instead of buying a feat that gives you X abilities, you buy a feat that gives you access to abilities of a particular class. Or perhaps Feat 1 says you get the levels 1 & 2 abilities from the class of your choice, while Feat 2 (prereq: Feat 1) says you get the levels 3 & 4 abilities. This includes what spell levels you have access to and how many slots you get. If you multiclass into wizard and buy Feat 1, you get (using 5.24 as the guide here), Spellcasting Ritual Adept, Arcane Recovery, Scholar, and the ability to cast 1st and 2nd level spells. You can't cast 3rd-level spells or gain additional spell slots until you get Feat 2, which gives you access to level 3 & 4 abilities,. Then give Feat 1 a prereq of being level 5 or something.

This means you don't have to balance them as much--the classes themselves should already be balanced.

 

[Steampunkette]

Here you could simply do a generic animal folk. Take a look at how Ebberon's Shifters are done. Instead of doing each individual type of animal as a different type of Shifter (wolf-shifter, deer-shifter), they're divided up into rough types (beasthide, wildhunt) and the player choses one. It's then up to the player to say if their wildhunt shifter looks like a wolf or a lion or hyena. Thus, your animal folk section could be divided in similar ways.

I was thinking of that, yeah. Though, personally, I infinitely prefer a shifter to a tabaxi or leonine or whatever name someone wants to give 'catperson'.

Likewise for planar creatures. Level Up, of course, has its Planetouched, with a simple heritage gift to differentiate between infernal, celestial, or elemental. When I put out my LU Handbook of Heritages, I included gifts to allow for law and chaos variants as well, as well as for various types of demons, daemons/yugoloths, and non-angelic celestials. Also, look at how 5.24 has done it. For tieflings, you pick Abyssal, Infernal, or Chthonic, and it's done in a table. Mind, I don't think the 5.24 way is particularly flavorful, since it's literally just spells, damage resistance, and darkvision, but it's simple, and the flavor can be in the description.

Yeah, kinda. That sorta thing largely works.

Doing these races like this gives people the options they want without cluttering up the races section or inviting that, yes, inevitable deluge of new options. I know people who really want to play animal-folk as a form of self-expression, and I myself generally prefer planetouched to any other type.

It's an issue. Trying to find a balance point between giving people what they want and spending the rest of your life statting out every possible variety of animal-person.

One idea to take from GURPS is language proficiency. As in, a difference between broken, fluent, and native levels of understanding. In GURPS, they actually required this for both speaking and reading. When I ran Ravenloft in GURPS, we had a lot of fun with characters who had lower levels of understanding.

That -can- be fun, yeah.

While I wouldn't necessarily go quite that far, you could divide understanding into poor and fluent, or do speaking and reading as different things. Instead of saying that culture or background gives you X number of languages, say it gives X number of points. One point for poor understanding, or either speaking or reading; two points for fluent understanding, or both speaking and reading.

That's not bad, yeah. Though maybe fluency in one and a couple points to spend, then add in Intelligence as points rather than languages? That way you could speak Ellowyn and Common well, and poorly speak 2-4 other languages.

I'm less fond of this myself, but perhaps that's because the Adventurer's Guide racial cultures just a copy of D&D's various races, and because, well, there's a lot of them. Too many. I like LU, but it's become very bloated.

S'truth. I'd be aiming for narrowing it SIGNIFICANTLY.

There wouldn't be a Dark Elf, Dark Dwarf, Dark Gnome, and general Underdark culture as four separate versions of the same thing. There'd -just- be the Underdark culture.

Now, while I've decided I'm not a huge fan of Draw Steel (nothing wrong with it; just not for me), I will say that the way they do cultures is pretty cool and worth taking inspiration from. In that game, culture is divided up into three types: environment, organization, and upbringing, each of which gives skills, and you mix-and-match as you please. Woodland Beaurocratic Noble, for instance. With this, you could then give several sample cultures. High Elf: Woodland, Beaurocratic, typically Academic, Creative, or Noble. Wood Elf: Woodland, Beaurocratic, typically Creative, Lawless, or Martial.

How you do the culture doesn't super matter so long as you cover the bases you need to cover and don't go overboard...

Because trust me: Someone pedantic enough could write 30 environments, 60 organizations, and a thousand upbringings.

Mmm, I'm not sure. Honestly, the only real reason to even care about lifespans in a game like this is when you had supernatural aging, like from ghosts and spells. Unless you bring that back, there's really no point to having this other than as a note for flavor purposes: people in cosmopolitan cities live longer than people in oppressive tyrannies do; elves have lifespans of over a millennia, while orcs rarely live to the age of 60. Yes, there are people who use D&D to play games that span decades or even centuries, but that's vanishingly rare.

Three big reasons come to mind.

1) Breaking Expectations
A player who is playing an orc raised among elves gets to be over 120 years old while most of the rest of the party is in their 20s and shock the rest of the table when they find out she's a grandma widow who outlived 3 human husbands. Some players are gonna love that kinda thing.

2) Establishing Narrative Weight
If the Wartorn Kingdom's life expectancy is a REDUCTION of how old you can get, rather than an expansion, it helps to sell the threat and make the party member who grew up under the oppressive regime into a bit of a martyr when they kill that tyrant and die from old age only a couple of winters later.

3) Gameplay Mechanics
"I've just stolen one year of your life. Tell me, how do you feel? and remember: Be honest. This is for Posterity."
piteous whimpering noises
"Fascinating..."

And, later: "NOT TO FIFTY!"
the sound of a heart breaking

Yes, Ghosts. But also spells and rituals that take percentages off your max lifespan could be a fun thing to play with!

I prefer languages to be regional or based around very specific ideals rather than on race or even culture (in my current LU game, I have the city-state's language, Albastilian, and also Contract (the trade/legal language), and Tower (the magic language). Plus languages for other countries elsewhere. But the different ancestries that make up the city of Albastile don't have their own languages.

That's pretty fair, yeah. That's what I meant by regional languages in the above post. French being the dominant language of the region and a major language of it's neighbors, for example.

But to me, it doesn't make sense that people in different Human cultures would all speak Human. While you could potentially say that there's very little drift in Elven languages (because of their lifespan or their trance-sleep which connects them to their past), that shouldn't be the case for shorter-lived people. And if there's going to be regional Human languages, there should also be regional languages for others. But that makes for a ginormous list of languages.

On the one hand, absolutely agreed. On the other hand, having Elves and Dwarves be obnoxious pedants about keeping their languages "True" could also be fun. Like how there's clear lingual drift in English between the UK and the US after a couple hundred years, but it's largely the same language, and every few decades someone comes along promising to reform the language into something more sensible (often with thru, threw, and throo being important keystone words for their phonetic spellings) and people shout them down.

That being said, if you do have "cultural tongues" and choose to rename them into things like Ellowyn, don't do what Draw Steel did and give them completely unrelated names: Dwarfs (and engineers) use Zaliac; bugbears and fey speak Khelt. Huh? That's going to be easily forgettable and then people will just default to Dwarvish and Sylvan.

Oh, nah. I wouldn't be so daft as to intentionally misspell "Celt" as a way to make a language for fey creatures. I'd probably be more likely to go to the root words like Dwergi for Dwarven, Ylfen for the Elves. And then that'd be it for the "Racial" languages with everyone else using Nation-Specific languages.

You know, with the point-based languages I could also do things like make it so no one speaks fluent Celestial. You're only ever allowed to put 1 point into that one or Infernal... Treat them almost like dead languages where people can learn some of it, but never really "Get It" as it's commonly used by otherworldly beings without magical intervention like Tongues or Comprehend Languages. Would also help sell why prophecies are so damned -vague- all the time. The language is imprecise 'cause the mortals don't really understand it all that well.

One other idea that's been floating around in my head (and is up for anyone else to use, if they want it) has been to have languages based on the creator god(s). Any sentient beings made by that god or pantheon share a language; there was no "Tower of Babel" event to cause them to have different languages. The key here is that this wouldn't be like most D&D where every sentient race has a god or pantheon. You could say that one pantheon made Humans, Halflings, Ogres, Giants, and other beings based on that sort of idea or body plan, and they all speak the same language. Another one created Lizardfolk, Dragonborn, Aarakocra, Merfolk, Sahuagin, Pterafolk, and other reptile/fish/avian beings, and they all share a language. At least that way it makes sense for you to be able to communicate from someone on the completely opposite side of the world!

Oh, for sure. That'd make it make sense at least... but. It would also be world-specific since not every world has the same gods.

Maybe you should make all the class features into feats. By which I mean, instead of buying a feat that gives you X abilities, you buy a feat that gives you access to abilities of a particular class. Or perhaps Feat 1 says you get the levels 1 & 2 abilities from the class of your choice, while Feat 2 (prereq: Feat 1) says you get the levels 3 & 4 abilities. This includes what spell levels you have access to and how many slots you get. If you multiclass into wizard and buy Feat 1, you get (using 5.24 as the guide here), Spellcasting Ritual Adept, Arcane Recovery, Scholar, and the ability to cast 1st and 2nd level spells. You can't cast 3rd-level spells or gain additional spell slots until you get Feat 2, which gives you access to level 3 & 4 abilities,. Then give Feat 1 a prereq of being level 5 or something.

I don't like Pathfinder 2e. Why would I make 10 level Pathfinder 2e?

This means you don't have to balance them as much--the classes themselves should already be balanced.

wibble wobbles hand a bit

It's a matter of exponential growth of complexity, more than anything. I'm pretty good at eyeballing balance and then refining... but I don't think I'd want to try to balance every possible combination of "Barbarian" class features when they're each 2-3 "Feats" that need to be balanced against each other -and- against and with the rest of the class features moving up the tree.

 

[doctorbadwolf]

What if the whole game used Focus Points, and you get roughly 1 per level, spells cost 1 per spell level, and different classes use them and regain them differently?

 

[loverdrive]

5) Multiclassing and Feats
Rather than being a Rogue 4, Barbarian 6, you'll just grab feats that grant you class abilities in addition to what you've got from your current class. Maybe get 5 feats over the course of leveling, and you can choose between grabbing class-feats or general feats.

It means balancing the multiclass feats against each other, and then the general feats against them... but I think it's plausible. I hope.

I think multiclassing works mostly fine, the main issue is that non-casters don't have a uhhh bus to connect them. Paladin/Sorcerer, or Paladin/Warlock, or Warlock/Sorcerer work so nicely because they all rely on spell slots, and advancing in either of your classes benefits the other too.

In contrast, there's no such bus for rogue/barbarian, or fighter/barbarian, or barbarian/monk. I imagine if all martial classes relied on fighter's superiority dice or something along those lines, multiclassing would work much, much better.

 

[Steampunkette]

I think multiclassing works mostly fine, the main issue is that non-casters don't have a uhhh bus to connect them. Paladin/Sorcerer, or Paladin/Warlock, or Warlock/Sorcerer work so nicely because they all rely on spell slots, and advancing in either of your classes benefits the other too.

In contrast, there's no such bus for rogue/barbarian, or fighter/barbarian, or barbarian/monk. I imagine if all martial classes relied on fighter's superiority dice or something along those lines, multiclassing would work much, much better.

Multiclassing "Works".

As a designer, there's an impetus to front load classes with core abilities and features for two reasons.

1) Establish Fantasy and Identity.
2) You can't scale an ability with progression if you don't have it, already.

This foments the inevitable problem of Splashing. Where classes are taken for 1-2 levels just to grab hold of their core identifying abilities to get maximum function from interaction with another class' features. Sorcadins and Lockadins are both popular because the Paladin gets a specific number of spell slots to offset their smiting ability, and Sorcerers get WAY MORE plus Sorcery Points to invent more slots to dump into smites. Meanwhile Lockadins can fire off two 11d8 smites every short rest once you get them to level 11.

That's not just synergy. It's explicitly a matter of taking the Paladin class's damage structure and breaking it. And the Coffeelock does the same thing with the Sorcerer by providing more Sorcery Points every short rest instead of only on a long rest.

I'd have -loved- if they'd made the effort to fix that in 2024, but it's just too damned popular to do anything about. And so Paladins become a zesty splash you can add to Sorcerers and Warlocks. Hell. Now you don't even need to take TWO levels of Paladin, just the first one, since Divine Smite is a 1st level Paladin spell. Level 2 just gets it "Always Prepared".

From a design standpoint, multiclassing is a headache and a half.

There's a reason, after all, you mentioned Paladin/Sorcerer and Paladin/Warlock but didn't say Paladin/Druid or Paladin/Wizard. It's not the spell slots and ability to cast magic that makes those classes compatible, it's the class features that -particularly- interact well between the structures. Spell slots just make it nicer.

Also: Barbarian 2, Monk X is actually a great splash. You get your rages for when you need more damage (+2 to each hit) and less incoming damage (1/2 BSP), and you get Reckless Attack for all of your unarmed strikes which gets insane.

 

[loverdrive]

This foments the inevitable problem of Splashing. Where classes are taken for 1-2 levels just to grab hold of their core identifying abilities to get maximum function from interaction with another class' features. Sorcadins and Lockadins are both popular because the Paladin gets a specific number of spell slots to offset their smiting ability, and Sorcerers get WAY MORE plus Sorcery Points to invent more slots to dump into smites. Meanwhile Lockadins can fire off two 11d8 smites every short rest once you get them to level 11.

As a designer, I don't see a problem in splashing.

Sorcadin is fun! Coffeelock is kind of neat, but whatever. I haven't played Lockadin (and am not sure how it works? Doesn't warlock already have access to smites? I must be missing something) but I imagine it's pretty fun too.

Sure, they are strong, arguably busted, but at least Sorcadin is, at least, interesting. Both sorcerer and paladin as singleclasses are less rich in depth than sorcadin is: she has to constantly evaluate two largely orthogonal options that solve different problems — is raw damage of smites better than utility of spells? The answer is constantly changing depending on what's going on.

Paladin mostly spends her spell slots on smites, sorcerer mostly spends her spells slots on, well, spells. The only reason not to throw smites (or fireballs) every turn is that you might want to save them for later. Such resource management might be interesting, but paired with a test of valuation skill it becomes deeper.

There's a reason, after all, you mentioned Paladin/Sorcerer and Paladin/Warlock but didn't say Paladin/Druid or Paladin/Wizard. It's not the spell slots and ability to cast magic that makes those classes compatible, it's the class features that -particularly- interact well between the structures. Spell slots just make it nicer.

Maybe I'm wrong, but I imagine multiclass stat requirements just kill Paladin/Druids or Paladin/Wizards. I don't remember how the actual mechanic works (I think there are just number requirements? Like 15 CHA, 15 STR to multiclass to paladin?), but I'm sure it's just more feasible to combine two classes that share the primary stat. Having high INT to qualify for multiclassing from/into wizard and high STR and CHA to qualify for paladin is unlikely.

I don't see why a very high-stated character going Pal/Wiz would be weak. Especially since wizard is the only one who has access to many utility spells (which of higher priority — damage is handled by smites/fireballs) it must be perfectly workable. Necromancer/Paladin (Hell yeah! Death knight!) especially sounds like an amazing time.

Overall, I think there should be more of that, not less — elevate all other cool combinations instead of removing/nerfing sorcadins.

 

[Horwath]

Multiclassing "Works".

As a designer, there's an impetus to front load classes with core abilities and features for two reasons.

1) Establish Fantasy and Identity.
2) You can't scale an ability with progression if you don't have it, already.

features need to be tied with class level for power.
usage number can be tied to overall level(proficiency bonus)

As stated, barbarians rage damage bonus of +2 is no problem for MC, it would be if it would scale in power to +4/5 irrelevant to level of barbarian

paladin smites can be "normalized" in a way that you need 5th level paladin to spend a 2nd level slot on smite and so on, sure as a sorcerer you can juggle spell slots via sorcery points for extra 1st level slots but you would be still limited by paladin level in smite power. OFC it would be easier with spell points and say that you need 1 SP for +1d8 smite damage, amount limited by paladin level.