I wonder what the best way to go about it would be? It can’t just work the same way as regular multiclassing since you’d effectively get no base class features for your second subclass
Pretty simple, just treat it like spellcaster multiclassing. Wizard/sorcerer/cleric/ 1/1/1 translates to a level 3 spellcaster for the sake of spell slots. Rogue 3/3 translates to class features level 6 and archetype feature level 3/3
A Spellcaster multiclassing always gets something on level up, be it a feature, more spell slots, or higher level slots.
A rogue multiclassing into rogue and splitting the levels would have dead levels at each subclass level.
To explain what I mean: a Rogue gets its subclass features at 3rd, 9th, 13th and 17th level. By going with your math, a 9th level rogue would classify as a 4/4 rogue (by rounding down) as far as the subclass is concerned, which means that the rogue gets nothing at 9th level.
Not only that. A 50/50 split for the multiclass progression would imply that a multiclassed rogue is precluded from getting any subclass feature higher than the 9th level one. By comparison, a Wizard/Sorcerer/Cleric multiclassed character can absolutely attain 9th level spell slots (although not 9th level spells, confusingly enough).
That still sounds balanced-ish. If anything, it’s too front-loaded. A 9th level rogue would still have its typical kit of sneakiness, skill proficiencies, and sneak attack at 9th level, but it wouldn’t have a 9th level bump via archetype because it received a 6th level bump via archetype.
A more typical example- a level 3 fighter/level 2 paladin wouldn’t get a second attack despite being a level 5 martial character, and they have to live with that mechanically poor decision. But they can instead choose to play until they become a level 5 fighter and then branch out instead, if they care to min/max.
And what gives you the impression it has to be 50/50? A sportsman can be great at throwing or hitting a ball, but it’s vastly different between one sport and another. You can be an incredible baseball pitcher and a garbage basketball player. Level 3 arcane trickster/level 17 assassin makes perfect sense to me.
Level 3 arcane trickster/level 17 assassin makes perfect sense to me.
That’s not a multiclass as intended in 5e rules. That’s just a 20th level rogue that got all the features from one subclass and the first feature of a second subclass for free.
I’ve never touched anything beyond level 20. I thought that’s what the epic stuff was? Are there regular class features and such published for those levels too, or were you homebrewing by then?
There’s published stuff for after 20. We didn’t use any of it, because the campaign was winding down. It all came to a nice ending, so wrapping up was more a matter of mutual storytelling than any dice rolls or challenges.
I’ve actually done it! I started at level 4, so I didn’t quite do the full 1-20 journey, but I did indeed go to 20 on xp per enemy killed and not milestone levelling
About a year and a half. It was a game explicitly intended to just be full of difficult combat encounters all of the time, so it was pretty much the ideal circumstances for levelling quickly. Her last encounter had about 60,000 xp worth of enemies in it per player, without using the multipliers for multiple enemies
I suppose an approach that takes the general intention of your design but is a bit omre mechanically rigorous could be to separate out subclass levels? You level up in one class as always, and every few levels the thing you get on levelling up is a subclass level. Subclasses then only get four or so levels, so you could be a warlock 11 (archfey 1 / fiend 2)
The difference is pretty minor either way. I’ve never had more issues balancing this than I have with sorcerer burst damage or creation bards collapsing economies.
I have actually personally done a subclass multiclass for a player in a game I run, but it was a very ad-hoc “okay it makes sense for your character to do this, so you’re just getting the level 3 feature from that subclass and the level 7 from this one, and this is how they interact” deal
I suppose I’m trying to think of how I would present it for games that I’m not involved in or don’t know the other players in. Something worded cleanly enough to stand up (at least a little) to situations when you can’t necessarily fall back on trust between the people at the table
(Assuming D&D 5E here)
I wonder what the best way to go about it would be? It can’t just work the same way as regular multiclassing since you’d effectively get no base class features for your second subclass
Pretty simple, just treat it like spellcaster multiclassing. Wizard/sorcerer/cleric/ 1/1/1 translates to a level 3 spellcaster for the sake of spell slots. Rogue 3/3 translates to class features level 6 and archetype feature level 3/3
That doesn’t work.
A Spellcaster multiclassing always gets something on level up, be it a feature, more spell slots, or higher level slots.
A rogue multiclassing into rogue and splitting the levels would have dead levels at each subclass level.
To explain what I mean: a Rogue gets its subclass features at 3rd, 9th, 13th and 17th level. By going with your math, a 9th level rogue would classify as a 4/4 rogue (by rounding down) as far as the subclass is concerned, which means that the rogue gets nothing at 9th level.
Not only that. A 50/50 split for the multiclass progression would imply that a multiclassed rogue is precluded from getting any subclass feature higher than the 9th level one. By comparison, a Wizard/Sorcerer/Cleric multiclassed character can absolutely attain 9th level spell slots (although not 9th level spells, confusingly enough).
That still sounds balanced-ish. If anything, it’s too front-loaded. A 9th level rogue would still have its typical kit of sneakiness, skill proficiencies, and sneak attack at 9th level, but it wouldn’t have a 9th level bump via archetype because it received a 6th level bump via archetype.
A more typical example- a level 3 fighter/level 2 paladin wouldn’t get a second attack despite being a level 5 martial character, and they have to live with that mechanically poor decision. But they can instead choose to play until they become a level 5 fighter and then branch out instead, if they care to min/max.
And what gives you the impression it has to be 50/50? A sportsman can be great at throwing or hitting a ball, but it’s vastly different between one sport and another. You can be an incredible baseball pitcher and a garbage basketball player. Level 3 arcane trickster/level 17 assassin makes perfect sense to me.
That’s not a multiclass as intended in 5e rules. That’s just a 20th level rogue that got all the features from one subclass and the first feature of a second subclass for free.
If you know anyone who has actually reached 20th level in a campaign, it might make a difference. I’ll put you in my will if I hit the lottery.
I ran a campaign that lasted several years and everyone went to 20. Technically past 20, though we never did any of the epic stuff.
It was 3.5 though
I’ve never touched anything beyond level 20. I thought that’s what the epic stuff was? Are there regular class features and such published for those levels too, or were you homebrewing by then?
There’s published stuff for after 20. We didn’t use any of it, because the campaign was winding down. It all came to a nice ending, so wrapping up was more a matter of mutual storytelling than any dice rolls or challenges.
I’ve actually done it! I started at level 4, so I didn’t quite do the full 1-20 journey, but I did indeed go to 20 on xp per enemy killed and not milestone levelling
How long did that take you?? The highest we’ve ever gone is level 11, and that took a couple of years.
About a year and a half. It was a game explicitly intended to just be full of difficult combat encounters all of the time, so it was pretty much the ideal circumstances for levelling quickly. Her last encounter had about 60,000 xp worth of enemies in it per player, without using the multipliers for multiple enemies
I suppose an approach that takes the general intention of your design but is a bit omre mechanically rigorous could be to separate out subclass levels? You level up in one class as always, and every few levels the thing you get on levelling up is a subclass level. Subclasses then only get four or so levels, so you could be a warlock 11 (archfey 1 / fiend 2)
The difference is pretty minor either way. I’ve never had more issues balancing this than I have with sorcerer burst damage or creation bards collapsing economies.
I have actually personally done a subclass multiclass for a player in a game I run, but it was a very ad-hoc “okay it makes sense for your character to do this, so you’re just getting the level 3 feature from that subclass and the level 7 from this one, and this is how they interact” deal
I suppose I’m trying to think of how I would present it for games that I’m not involved in or don’t know the other players in. Something worded cleanly enough to stand up (at least a little) to situations when you can’t necessarily fall back on trust between the people at the table