In 1e demihumans we’re the only ones to “ multiclass” . They could run 2-3 classes at the same time splitting the xperience. On the other hand they had class limits in everything but rogue. Humans couldn’t multiclass but they could switch class - you had to have the stats and you couldn’t use any of the features of your original class except your HPs until you had exceeded the original class level in the new class. 5e’s multiclassing has limited rules to allow players greater freedom in what they can create. Rules like those above limit the creativity.
You are correct Multiclassing is the problem, which is why it was only allowed for non-human PC's back in the early days of D&D. Mostly because they already had level limits.
Humans were forbidden to multiclass, they could only dual class.
The fact that there were two mechanically different ways to multiclass in AD&D was not one of the finest design decisions ever made.
(The entirety of the handling of non-human characters' classes was just a weird mess.)
But AD&D's leveling systems were completely different from 5e's, so even if you think it was a good idea for that game, it says nothing about whether it works for 5e.
I find the issue with multiclassing isn't that class combinations are overpowered, its that there are very few rules about how you can multiclass (Which is my biggest issue with 5e in general).
I find no issue with paladin/warlock combos, but the rules allow you to only take as many levels as you want from either class whenever you wanted to take them. If there were a few more solid rules like "If you multiclass, you need to take x levels before taking a different class level" or "You can multiclass if your subclass shares a feature with your new class" (ie. Arcane Trickster multiclassing into wizard or something)
Putting arbitrary limits like these on multiclassing gets in the way of a wide variety of character ideas. (For instance, the wizard apprentice who washed out right at the beginning of their career, and has since made a warlock pact to get the power they wanted.) You'd also prevent the alternating-levels balanced approach.
If the GM wants to put limits on multiclassing, they can impose the limits that make sense for their table and world.
I was giving examples of limitations that could be put in place to be easily understood; not what should be put in place. Multiclassing as a system isn't something that WOTC really put thought into balancing in any way, shape or form, but with a little balancing by applying rules to it, it could be a much more robust and rewarding system.
As it is, it's a vestigial system that WotC kept from earlier editions but didn't want to actually do anything with and so you get game breaking combinations.
Some are more arbitrary than others. If they'd wanted to reduce one-level dips, they could have put a three-level requirement on multiclassing, or they could have rearranged class benefits so you don't get as much right off the top by dipping into a class for one level. Which they did, and it also reduces decision load for new players. (Which I suspect was the primary reason they did it.)
I was giving examples of limitations that could be put in place to be easily understood; not what should be put in place. Multiclassing as a system isn't something that WOTC really put thought into balancing in any way, shape or form, but with a little balancing by applying rules to it, it could be a much more robust and rewarding system.
As it is, it's a vestigial system that WotC kept from earlier editions but didn't want to actually do anything with and so you get game breaking combinations.
It's not vestigial. It's an entirely new system, unlike those of 1e/2e, 4e, and only cosmetically similar to that of 3e. (The level cap matters a lot.)
There is no way to design a multiclassing system so that people cannot find synergies that will "break" the game, except by making it so weak or awkward to use that people don't.
There is also no way to design a non-multiclassing system so that people won't find synergies that will "break" the game. If you give people choices to make, they're gonna figure out what the "good" choices are.
But that's fine. Multiclassing needs to exist for roleplaying reasons. That's its primary reason to be. If somebody goes at it in actual play with the approach of "I will play the thing that gives me the most plusses", and that interferes with the other players' fun, that's a social problem, and should be solved with social solutions.
But the multiclassing rules cannot actually distinguish between "I took a warlock dip for access to eldritch blast and hex so now I have better ranged cantrips", and "a mysterious entity tempted me with offers of new power, and I made the deal, but will probably regret the cost once I learn what it is". Mechanically, they're identical.
The problem is that there's a cottage industry on social media talking about "broken" stuff that over-emphasizes the prevalence & importance of such at 90% of tables.
My personal rules for multiclassing: 1. Does it serve a strong narrative purpose in the game we're playing, in its own multiverse? Please explain how via backstory & handed-out lore. 2. Are you using it to cheese something? Not happening. 3. Did you learn it from social media? Congrats, it's banned. 4. Does it have a name? Banhammer incoming.
Rollback Post to RevisionRollBack
DM, player & homebrewer(Current homebrew project is an unofficial conversion of SBURB/SGRUB from Homestuck into DND 5e)
Once made Maxwell's Silver Hammer come down upon Strahd's head to make sure he was dead.
Always study & sharpen philosophical razors. They save a lot of trouble.
The problem is that there's a cottage industry on social media talking about "broken" stuff that over-emphasizes the prevalence & importance of such at 90% of tables.
Character optimization is a hobby all its own. I tend to liken it to speedrunning video games -- clearly a hobby with extremely dedicated adherents, but absolutely nothing like the normal gameplay experience, and shaping normal D&D play to do something about it is like trying to stop people from clipping through the walls and running on top of your rooms -- if it's something that happens easily, without trying, you have a problem. If it takes deliberate effort, you don't.
My personal rules for multiclassing: 1. Does it serve a strong narrative purpose in the game we're playing, in its own multiverse? Please explain how via backstory & handed-out lore.
This is probably the only one truly needed IMO. If it's justified by actual roleplay, then it's probably fine. I had a player who was thinking of taking a hexblade dip on their sorcerer. So I gave them the opportunity to make a deal. It's not just a minor powerup -- it's an entanglement with a supernatural force that has its own agendas that don't necessarily align with the character's. The actual combat boost is not a big deal in practice, but the extra supernatural entanglement for a character who already had several adds to the fun.
The problem is that there's a cottage industry on social media talking about "broken" stuff that over-emphasizes the prevalence & importance of such at 90% of tables.
My personal rules for multiclassing: 1. Does it serve a strong narrative purpose in the game we're playing, in its own multiverse? Please explain how via backstory & handed-out lore. 2. Are you using it to cheese something? Not happening. 3. Did you learn it from social media? Congrats, it's banned. 4. Does it have a name? Banhammer incoming.
I would have liked to see Multiclass removed from the game after WoTC bought TSR. I would have kept the Dual Class option, since that was far more balanced.
The problem is that there's a cottage industry on social media talking about "broken" stuff that over-emphasizes the prevalence & importance of such at 90% of tables.
My personal rules for multiclassing: 1. Does it serve a strong narrative purpose in the game we're playing, in its own multiverse? Please explain how via backstory & handed-out lore. 2. Are you using it to cheese something? Not happening. 3. Did you learn it from social media? Congrats, it's banned. 4. Does it have a name? Banhammer incoming.
I would have liked to see Multiclass removed from the game after WoTC bought TSR. I would have kept the Dual Class option, since that was far more balanced.
The current experience table doesn't work like that and it wasn't balanced, it was a pain in the keister.
Rollback Post to RevisionRollBack
"Sooner or later, your Players are going to smash your railroad into a sandbox."
-Vedexent
"real life is a super high CR."
-OboeLauren
"............anybody got any potatoes? We could drop a potato in each hole an' see which ones get viciously mauled by horrible monsters?"
Multiclassing in Dungeons & Dragons 5e offers players the flexibility to combine different class features and spellcasting mechanics to create unique characters. However, this flexibility often leads to significant imbalance and optimization abuses that go well beyond the intent of individual class design. The system’s lack of scaling control across multiple classes enables combinations that are disproportionately powerful and frequently exploited by power gamers.
I agree with this, but there are other players that enjoy the character optimization aspect of D&D. The easy solution would be for the DM to rule that player characters can only have a single class.
One of my favorite characters was/is a dual/switch class human. While I have had great fun playing him thru 3 switches it’s a freaking pain in the … to forgo the powers and abilities of earlier classes until you surpass them in the present one. I’m quite happy overall with the 5e multiclassing. It’s problems look worse from the outside than from actual lay. I find this to be typical of a lot of the power building that goes on. 5e’s , especially 2024, rules may not b “perfect” but they are easy, workable and don’t generally produce overpowered ( for their level) characters.
My main issue with multiclassing is that it tends to be either OP or a trap. In a game where characters tend to be overly resilient a lot of the time anyway, another power boost just makes it worse. On the other hand, most multiclasses (in terms of how many options there are) tend to be traps and often people pick them without thinking about wide Vs tall, for example. Either way, they're a pain to account for in balancing.
That said, I've only seen one 2024e multiclass (Barb6/Moon Druid4), so they might have improved the situation. The OP aspects I think may have been reined in by making dips less potent. It's easier to help players understand why weak MCs are a bad idea (so at least they're going in with their eyes open) than convincing minmaxxers not to minmax, so it's a good direction.
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
The problem is that classes get too many good things at level 1, as they require them to work. Probably there would not be DIP issue if characters would start at level 3 like 2024 version invites to do and distributing all those features through the 3 levels. I.e. fighters getting light armor at lvl 1, medium and shields at lvl 2, and heavy armor at lvl 3. Then if you are a fighter as main class you start at lvl 3 so you have all its starting features, but if you multiclass with a spellcaster you would get them progressively.
So the Warlock could make the Pact at level 3, and so on with all the classes.
What about ability boosts and/or feats that 1st occur at level 4. If I have a 9th level non fighter single class character, at level 4 and 8 I was able to do 4 point boost in abilities or less boost but add up to two feats.
But if I do x3/y3/z3 I am 9th level and never got any feats or ability boosts. Is that the correct translation of the PHB?
In addition, starting at high levels means some of the pain while multiclassing at low levels is not experienced. It was mentioned that in 1e you had to split XPs among classes, but what was not mentioned was that each class had a progression level different from another class. With every level having the same progression, it does eliminate some of issues & pain to multi class. However, not sure you quantify the pain of spending a lot of XPs to get a low level in a different class.
Warlock is pretty easy to limit via roleplay what can be picked up at low levels. The Patron who granted that level of warlock as part of the deal insists X/Y/Z be picked up first. For example, Pact of Blade can not be picked up until 5th level because the Patron demands the first 3 invocations be picked 1st. That eliminates doing a 1 level dip to get Pact of the blade.
I would have liked to see Multiclass removed from the game after WoTC bought TSR. I would have kept the Dual Class option, since that was far more balanced.
Good news: they did remove multiclassing.
Specifically, if you mean "multiclassing like it was in AD&D", instead of "multiclassing: the rules by which a character can have more than one class", it's gone. Since you differentiate between multiclassing and dual-classing (which is just another kind of multiclassing by the second definition), then yes, they got rid of it. Also they got rid of dual-classing.
Then they put in a brand-new, unified multiclass system in 3e. Then a different one in 4e. Then yet another one in 5e.
4e is its own thing, but the 3e/5e systems are more like old school dual-classing than they are like old-school multiclassing, in that you level your classes separately, rather than in parallel. Are they a lot like the old, cumbersome, dual-class system? No, not really. The mechanics of dual-classing only made sense, inasmuch as they actually made sense, in the context of the leveling system of AD&D. Since 3e and 5e have substantially different leveling systems, they need multiclass mechanics that make sense in their context, and trying to shoehorn in the old dual-classing mechanics would likely either be a bad design or end up changing them into unrecognizability.
What about ability boosts and/or feats that 1st occur at level 4. If I have a 9th level non fighter single class character, at level 4 and 8 I was able to do 4 point boost in abilities or less boost but add up to two feats.
But if I do x3/y3/z3 I am 9th level and never got any feats or ability boosts. Is that the correct translation of the PHB?
Yes, it's actually a significant balancing factor in multiclassing. Your multiclass levels delay your higher-level features, including fundamental stuff like feats and stat bumps.
And, whatever levels you're playing to, there are going to be features from your primary class that you never get because you multiclassed. You can get away with it and still be effective mostly because the classes are front-loaded, giving lots of stuff early on, which can make up for the lack of your higher-level features you'd otherwise have. Class feature delay is most noticeable up to level five, then multiclass can gain some ground into the mid-levels, and it starts losing ground again as you head for the level cap.
Multiclassing in Dungeons & Dragons 5e offers players the flexibility to combine different class features and spellcasting mechanics to create unique characters. However, this flexibility often leads to significant imbalance and optimization abuses that go well beyond the intent of individual class design. The system’s lack of scaling control across multiple classes enables combinations that are disproportionately powerful and frequently exploited by power gamers. Here’s why multiclassing is at the root of most “broken” abilities and spells:
TL;DR Multi-classing isn't a problem. What is "abuse" or "broken" is subjective and other DMs and players may not agree. Almost all of the examples given below as concerns were actually eliminated with the 2024 rules revision.
Personally, I have had no issues with multi-classing. There are a few combinations that had particular uses or capabilities (describing them as "abuse" already implies that there is something wrong/bad about the combinations) but most of those were reduced or eliminated with the 2024 rule set. In fact, the biggest "abuse" in the 2024 rules was the spell Conjure Minor Elementals which was broken from the beginning and should never have seen the light of day. At least, they have errata'ed it.
In addition, different people enjoy making different types of characters. Some players like characters with a bad stat or two and critical flaws. Others like playing heroes without notable weaknesses. Some like to create characters that seem mechanically powerful to them. As long as everyone at the table is having fun, how you play doesn't matter - including whether multi-classing is allowed or not.
Finally, any DM that finds multi-classing to be unbalanced can simply decide to not allow it in their game. For anyone who doesn't share these concerns or have a few specific ones, they can choose to allow multi-classing or disallow specific combinations if they are considered problematic.
1. Front-Loaded Power and Ability Spikes
Many classes in 5e have highly potent features at early levels—especially at level 1. For example:
Hexblade Warlocks get Hexblade’s Curse, Hex Warrior, and Charisma-based weapon attacks at level 1.
Paladins gain Divine Smite at level 2, allowing them to burn spell slots for massive burst damage—regardless of the spellcasting class those slots come from.
Sorcerers get Metamagic at level 3, letting them twin Hold Person, Haste, or Disintegrate—abilities that can single-handedly break encounters.
By dipping into one or two levels of these classes, characters gain outsize benefits that stack with their primary class features without meaningful trade-offs.
These are all 2014 issues and are all removed in 2024.
1) All subclasses are only available at level 3. This means that anyone trying for some combination based on subclass ability like the Hexblade, must wait until level 3.
However, even one level in a different class will slow progression in whatever your main class might be. A wizard which starts with a level of artificer or fighter for proficiency in constitution saving throws and the ability to wear armor and a shield ... also slows their spell progression. 3rd level spells won't be available until level 6 for this character. This is a significant consideration in actual play since the wizard with the 1 level dip is always a level behind where they could be and where other characters in the party might be.
2) Looking at your list specifically -
- a 1st level warlock can choose pact of the blade to attack with charisma and create a weapon. That is all. You also left off the other main attraction of a hexblade dip which was proficiency with martial weapons, medium armor and shields. The only thing now available as a feature is weapon attacks with ONE specific weapon using charisma.
- Smite is now totally different - requires a bonus action and is only 1 time/turn. Problem eliminated.
- Twin spell metamagic now only lets you cast a spell that can be upcast by one higher level. This does still work with hold person - but this works for a pure classed sorcerer so I don't see how it is a balance issue due to multi-classing. Twinned Haste and Disintegrate are no longer options.
2. Spellcasting Progression Exploits
The way 5e handles multiclass spellcasting through the shared spell slot table allows for disproportionately powerful spell slot access without comparable spell selection or feature scaling.
A Sorcerer/Paladin build might only have a few sorcerer levels but gains high-level spell slots—then uses those to supercharge Divine Smites, essentially turning every hit into a potential nova explosion.
The classic Coffee Lock (Warlock/Sorcerer) abuses the Warlock’s short rest spell slots combined with Sorcerer’s Flexible Casting to generate an infinite supply of spell slots, breaking the entire rest economy.
This is only possible because of how multiclassed spell slot progression interacts across classes—something that was clearly not balanced with these combinations in mind.
I'm not sure you understand the multi-class spellcasting table.
A level 17 paladin/3 sorcerer is considered a level 12 spell caster ... compared to a level 10 spell caster for a level 20 paladin. Comparing these two ... the level 12 spellcaster has just ONE 6th level spell that the level 10 does not have - they are otherwise identical. I think it unlikely that one additional level 6 spell (while missing out on the level 18-20 paladin features) would be at all unbalancing. At lower levels the effect is equally as insignificant.
In addition, smite in the 2014 version was capped at 5d8 damage no matter what spell slot was used so the effect is even smaller since a 6th level slot does the same as a 5th level slot when used for smiting. (Perhaps, I don't need to mention that there are likely a lot more valuable things to use a 6th level slot on than smiting).
Finally, the 2024 smite uses a bonus action to cast the Divine Smite spell. In this case, there is no cap to possible damage so a higher level spell slot has some benefit. However, since it is a spell, it can be counterspelled and is subject to the usual limitations imposed by casting a spell.
As far as the coffeelock goes ... it has a very simple answer ... "DM says No." Utilizing a coffeelock requires the DM to agree to it. They have to allow the character the ability to take multiple short rests instead of a long rest AND allow the character to go without long rests so that the spell slot count never resets. Otherwise, the character is limited to whatever they can obtain from a couple of short rests in a typical adventuring day which doesn't unbalance anything in my opinion.
3. Stacked Action Economy and Bonus Actions
Multiclassing enables combinations that exploit the action economy more effectively than any single class:
Monk/Rogue builds allow massive mobility, bonus action disengage, bonus action attacks, and Sneak Attack damage all rolled into one.
Fighter/Warlock combinations can use Eldritch Blast with Agonizing Blast, Hexblade’s Curse, and Action Surge to unleash a barrage of high-damage attacks in one turn.
Certain Bard builds can hand out Inspiration, cast spells, and use cunning support abilities all in the same round thanks to multiclass synergy.
This stacking leads to turns that feel unfair to both the DM and other players who chose a single-class path.
I'm not sure what you are getting at here. Characters are limited to ONE bonus action. Multiclassing may give them more options with what to do with that bonus action but that is all. In the 2024 rules, a monk can ALREADY dash or disengage as a bonus action for free ... there is no need to multiclass to a rogue to obtain the ability. Even the 2014 monk could ALREADY do this but it required the expenditure of a ki point for Step of the Wind. Multi-classing monk and rogue is arguably a waste of levels since it delays the monks progression in ki/focus points ... unless the character wants to pick up the expertise in skills offered by the rogue. Cunning action by itself adds very little for a monk.
A hexblade warlock can already use agonizing blast with hexblades curse (hexbades curse requires a level 3 hexblade warlock in the 2024 rules). In the 2014 rules, a multi-classed fighter could use action surge to cast it again which was pretty cool/powerful but it was possible only once/short rest since action surge is a short rest ability. Arguably, in the 2014 rules, the better combination was 2 fighter/X wizard so that they could cast 2 fireballs or something equally powerful.
However, the 2024 rules solve this issue ... the action from action surge can not be used for a Magic action. "On your turn, you can take one additional action, except the Magic action." ... so the 2024 rules prevent any use of action surge to cast multiple spells or use magic items which require the Magic action multiple times.
How is any of this "unfair to the DM"? The DM adjudicates the game, runs NPCs, sets up situations and provides fun for the players - the rules are just the rules. Since the DM is implementing the rules and running the game ... how can they be in any way "unfair to the DM"?
"Unfair to other players" is also a misunderstanding - if you have a group of players with different levels of understanding of the game and one of the players is using their knowledge to roflstomp enemies and make other players feel bad - that isn't a game problem, it is a PLAYER problem. The DM needs to have a chat with the player, possibly all the players and resolve the issue out of game. It isn't a rules problem ... it is a people problem.
4. Negating Intended Weaknesses
Each class in 5e is designed with strengths and weaknesses. Multiclassing allows players to patch over a class’s drawbacks with another class’s strengths, often leading to over-tuned “best of all worlds” characters.
A Barbarian/Paladin gets both Rage resistance and Divine Smite burst, combining tankiness and massive offense in one package.
A Wizard/Cleric can get Shield, Counterspell, Fireball, and Healing Word—turning one character into the Swiss Army knife of spellcasting.
A Rogue/Fighter/Warlock can deliver sneak attacks, multiple attacks, and spell-like abilities while maintaining high AC, mobility, and sustained damage.
These combinations circumvent the balance framework that single-class characters are constrained by.
It isn't clear to me that you have actually played these characters through a campaign (as opposed to a level 20 one shot) or are just theory crafting.
Barbarian/paladin - in the 2024 rules, smiting now requires casting a spell which is not compatible with raging since it prevents spellcasting. Every level of barbarian slows the progression in paladin or vice versa. Even just looking at the 2014 rules, extra attack would normally be at level 5 .. taking barbarian levels early slows this until level 7 and slows spell progression for smiting. Waiting to take the barbarian levels until later means that the combination doesn't really become available until level 7 or 8 depending on whether you want the aura or not. In addition, paladin already needs a high strength and charisma, raging is limited to medium armor ... so the paladin can't wear heavy armor and rage. This means requiring at least a 14 dex to optimize AC and a decent constitution. Unless your table uses rolled stats AND rolls incredibly well - a barbarian/paladin character is horrifically MAD :) (Multiple Attribute Dependent).
Wizard/cleric - yes a very good combination if you don't mind delaying wizard spells. Armor proficiencies are really useful for a wizard since their hit points are low. However, this delays wizard spell acquisition by at least a level. It also requires at least a 13 wisdom ... likely 14 would be better for skills/saves/modifier for spells. The character does get an emergency healing word and/or cure wounds, guidance (which has changed in 2024), bless and a couple other useful level 1 cleric spells. However, the focus is wizard and the character will always be a worse wizard than they would be without the level of cleric. Personally, I don't find it unbalanced and if a player likes the character backstory of a Cleric of Mystra deciding to practice magic instead of just worship it then awesome :)
Rogue/fighter/warlock - lol - "jack of all trades, master of none" - sneak attack goes up with rogue level, extra attack goes up with fighter or warlock level. Rogue and fighter need a good dex, warlock needs charisma, wisdom is useful for a rogue .. progression in each class is slow because it is spread over so many classes. I have seen VERY few triple classed characters in actual play and they are almost always "weaker" from a "power" perspective than a single classed character. You mention high AC ... but if the rogue is going to have a high dex then they don't benefit from medium or heavy armor anyway - if wearing light armor, the rogue will have the same AC as a single class. Go Arcane trickster and they will have the shield spell if they want it too (2024 rules). The rogue also has the mobility from cunning action and if you leave 2014 sharpshooter/great weapon master out of the discussion, the rogue sneak attack damage scaling keeps up reasonably with fighter and other multi-attack weapon users by itself - especially with features like steady aim or an arcane trickster using a familiar to provide the help action to obtain advantage on their attack. Basically, there is nothing unbalanced about a rogue/fighter/warlock character if someone wanted to play one.
5. 2024 Revisions Helped—But the Core Problems Persist
With the release of the 2024 One D&D updates, some of the most abusable multiclass combinations have been toned down. For example:
Smite no longer scales so freely with higher-level spell slots unless you actually have the Paladin levels to support it.
Warlock spellcasting has been reworked to reduce short-rest slot abuse.
Some class features have been restructured to scale more cleanly or require deeper investment.
However, the core problem remains: the rules still allow mixing powerful, front-loaded class features in a way that bypasses intentional class balance. Many broken builds are still possible because the multiclass system continues to lack proper safeguards against stacking unintended synergies. Until core mechanics—like spell slot progression, bonus action economy, and scaling feature gates—are fundamentally restructured, multiclassing will remain the go-to path for players seeking to break the system wide open.
In my opinion, 2024 rules actually addressed most of your concerns from 2014 ... concerns which many folks, myself included, did not have any issue with to start with. I don't think multi-classing is unbalanced.
As for your comments above ... I'm not sure you've read the 2024 rules ... paladin smite now scales with 1d8/spell level with no level limit ... so if they happen to have a level 7 spell slot then it can do 8d8 damage to one target, requiring a bonus action. Since a level 5 cone of cold will do 8d8 cold damage, con save for half with a level 5 slot - that seems like a balanced trade off. There is no requirement to use paladin spell slots. This is also incredibly bad compared to a spell like Conjure Minor Elementals which will do 5d8 on EACH attack with a 7th level slot (it used to be 8d8 but they changed it from totally broken to less broken ... it is still strictly better than Spirit Shroud).
Warlock pact magic hasn't changed - they still get all spell slots back on a short rest but have an additional feature that allows them to restore some spell slots in 1 minute once/long rest. If someone had some way to misuse warlock short rest spell slots before - it doesn't seem to have changed now.
Requiring 3 levels for all subclass abilities has reduced the combinations where one or two levels of a class gives access to some subclass defining features that could be used in combination with other class features. Three levels is a lot in terms of character development which is a good trade off for someone wanting to combine elements from different subclasses.
In conclusion:
1) Multi-classing is not the problem. It just increases options available - how those are used are up to the DM and the players at the table. A DM can always choose what to allow at their table.
2) A lot of players and DMs like multi-classing - what a DM thinks is "abuse" or "broken" is subjective and different DMs may not agree.
3) Most of the biggest concerns outlined were eliminated in whole or in part by the 2024 rules revision.
Not many people know the pain of needed 5000 exp in order to get lvl 2 on your F/MU(wizard)
That would be specifically 2 MU. You get to 2 Fighter at 4000xp ... so you'd be a 2 fighter/1 MU from 4000->5000xp. :) ... you'd also be either an elf or a half-elf ... the only non-humans that could be an MU ... though a gnome could be an Illusionist :)
Enough of us do, if your going to multiclassing get it down early when the costs are lower, but realize your going to pay costs for the life of the character. I have a suspicion that much of the multiclassing is really “whiteroom” exploration not actual game play over time. That’s why I think it’s probably a much smaller problem than the discussion makes it seem to be.
Tell me you never play martials without telling me you never play martials... none of these are better than straight classes because they all give up at least as much as they gain. The only real problematic thing in OP's post was Coffeelock, and I have literally never seen one in play because even if they didn't get shut down (which they did) they just became boring. The 2014 Hexblade dip was not ideal just because it was such a good choice, but it still didn't make you Superman or anything.
Certain Bard builds can hand out Inspiration, cast spells, and use cunning support abilities all in the same round thanks to multiclass synergy.
And? Straight bards can already do 2/3 of that, and there's no "cunning support abilities" you can MC for that are better than the benefits of just staying in bard to progress your features and get higher level spells. This is the kind of build that is absolutely multiclassing working exactly as intended. Not to mention that the table's powergamer optimizing their ability to make everyone else awesome is basically a best-case scenario.
Multiclassing is honestly very well done in 5e+. It's at the point where being more strict would shut down way more legitimate builds than "broken" ones.
In 1e demihumans we’re the only ones to “ multiclass” . They could run 2-3 classes at the same time splitting the xperience. On the other hand they had class limits in everything but rogue. Humans couldn’t multiclass but they could switch class - you had to have the stats and you couldn’t use any of the features of your original class except your HPs until you had exceeded the original class level in the new class.
5e’s multiclassing has limited rules to allow players greater freedom in what they can create. Rules like those above limit the creativity.
Wisea$$ DM and Player since 1979.
The fact that there were two mechanically different ways to multiclass in AD&D was not one of the finest design decisions ever made.
(The entirety of the handling of non-human characters' classes was just a weird mess.)
But AD&D's leveling systems were completely different from 5e's, so even if you think it was a good idea for that game, it says nothing about whether it works for 5e.
Putting arbitrary limits like these on multiclassing gets in the way of a wide variety of character ideas. (For instance, the wizard apprentice who washed out right at the beginning of their career, and has since made a warlock pact to get the power they wanted.) You'd also prevent the alternating-levels balanced approach.
If the GM wants to put limits on multiclassing, they can impose the limits that make sense for their table and world.
Every rule in the game is arbitrary.
I was giving examples of limitations that could be put in place to be easily understood; not what should be put in place. Multiclassing as a system isn't something that WOTC really put thought into balancing in any way, shape or form, but with a little balancing by applying rules to it, it could be a much more robust and rewarding system.
As it is, it's a vestigial system that WotC kept from earlier editions but didn't want to actually do anything with and so you get game breaking combinations.
Some are more arbitrary than others. If they'd wanted to reduce one-level dips, they could have put a three-level requirement on multiclassing, or they could have rearranged class benefits so you don't get as much right off the top by dipping into a class for one level. Which they did, and it also reduces decision load for new players. (Which I suspect was the primary reason they did it.)
It's not vestigial. It's an entirely new system, unlike those of 1e/2e, 4e, and only cosmetically similar to that of 3e. (The level cap matters a lot.)
There is no way to design a multiclassing system so that people cannot find synergies that will "break" the game, except by making it so weak or awkward to use that people don't.
There is also no way to design a non-multiclassing system so that people won't find synergies that will "break" the game. If you give people choices to make, they're gonna figure out what the "good" choices are.
But that's fine. Multiclassing needs to exist for roleplaying reasons. That's its primary reason to be. If somebody goes at it in actual play with the approach of "I will play the thing that gives me the most plusses", and that interferes with the other players' fun, that's a social problem, and should be solved with social solutions.
But the multiclassing rules cannot actually distinguish between "I took a warlock dip for access to eldritch blast and hex so now I have better ranged cantrips", and "a mysterious entity tempted me with offers of new power, and I made the deal, but will probably regret the cost once I learn what it is". Mechanically, they're identical.
The problem is that there's a cottage industry on social media talking about "broken" stuff that over-emphasizes the prevalence & importance of such at 90% of tables.
My personal rules for multiclassing:
1. Does it serve a strong narrative purpose in the game we're playing, in its own multiverse? Please explain how via backstory & handed-out lore.
2. Are you using it to cheese something? Not happening.
3. Did you learn it from social media? Congrats, it's banned.
4. Does it have a name? Banhammer incoming.
DM, player & homebrewer(Current homebrew project is an unofficial conversion of SBURB/SGRUB from Homestuck into DND 5e)
Once made Maxwell's Silver Hammer come down upon Strahd's head to make sure he was dead.
Always study & sharpen philosophical razors. They save a lot of trouble.
Character optimization is a hobby all its own. I tend to liken it to speedrunning video games -- clearly a hobby with extremely dedicated adherents, but absolutely nothing like the normal gameplay experience, and shaping normal D&D play to do something about it is like trying to stop people from clipping through the walls and running on top of your rooms -- if it's something that happens easily, without trying, you have a problem. If it takes deliberate effort, you don't.
This is probably the only one truly needed IMO. If it's justified by actual roleplay, then it's probably fine. I had a player who was thinking of taking a hexblade dip on their sorcerer. So I gave them the opportunity to make a deal. It's not just a minor powerup -- it's an entanglement with a supernatural force that has its own agendas that don't necessarily align with the character's. The actual combat boost is not a big deal in practice, but the extra supernatural entanglement for a character who already had several adds to the fun.
I would have liked to see Multiclass removed from the game after WoTC bought TSR. I would have kept the Dual Class option, since that was far more balanced.
The current experience table doesn't work like that and it wasn't balanced, it was a pain in the keister.
"Sooner or later, your Players are going to smash your railroad into a sandbox."
-Vedexent
"real life is a super high CR."
-OboeLauren
"............anybody got any potatoes? We could drop a potato in each hole an' see which ones get viciously mauled by horrible monsters?"
-Ilyara Thundertale
I agree with this, but there are other players that enjoy the character optimization aspect of D&D. The easy solution would be for the DM to rule that player characters can only have a single class.
One of my favorite characters was/is a dual/switch class human. While I have had great fun playing him thru 3 switches it’s a freaking pain in the … to forgo the powers and abilities of earlier classes until you surpass them in the present one. I’m quite happy overall with the 5e multiclassing. It’s problems look worse from the outside than from actual lay. I find this to be typical of a lot of the power building that goes on. 5e’s , especially 2024, rules may not b “perfect” but they are easy, workable and don’t generally produce overpowered ( for their level) characters.
Wisea$$ DM and Player since 1979.
My main issue with multiclassing is that it tends to be either OP or a trap. In a game where characters tend to be overly resilient a lot of the time anyway, another power boost just makes it worse. On the other hand, most multiclasses (in terms of how many options there are) tend to be traps and often people pick them without thinking about wide Vs tall, for example. Either way, they're a pain to account for in balancing.
That said, I've only seen one 2024e multiclass (Barb6/Moon Druid4), so they might have improved the situation. The OP aspects I think may have been reined in by making dips less potent. It's easier to help players understand why weak MCs are a bad idea (so at least they're going in with their eyes open) than convincing minmaxxers not to minmax, so it's a good direction.
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
The problem is that classes get too many good things at level 1, as they require them to work. Probably there would not be DIP issue if characters would start at level 3 like 2024 version invites to do and distributing all those features through the 3 levels. I.e. fighters getting light armor at lvl 1, medium and shields at lvl 2, and heavy armor at lvl 3. Then if you are a fighter as main class you start at lvl 3 so you have all its starting features, but if you multiclass with a spellcaster you would get them progressively.
So the Warlock could make the Pact at level 3, and so on with all the classes.
What about ability boosts and/or feats that 1st occur at level 4.
If I have a 9th level non fighter single class character, at level 4 and 8 I was able to do 4 point boost in abilities or less boost but add up to two feats.
But if I do x3/y3/z3 I am 9th level and never got any feats or ability boosts.
Is that the correct translation of the PHB?
In addition, starting at high levels means some of the pain while multiclassing at low levels is not experienced.
It was mentioned that in 1e you had to split XPs among classes, but what was not mentioned was that each class had a progression level different from another class. With every level having the same progression, it does eliminate some of issues & pain to multi class. However, not sure you quantify the pain of spending a lot of XPs to get a low level in a different class.
Warlock is pretty easy to limit via roleplay what can be picked up at low levels. The Patron who granted that level of warlock as part of the deal insists X/Y/Z be picked up first. For example, Pact of Blade can not be picked up until 5th level because the Patron demands the first 3 invocations be picked 1st. That eliminates doing a 1 level dip to get Pact of the blade.
Good news: they did remove multiclassing.
Specifically, if you mean "multiclassing like it was in AD&D", instead of "multiclassing: the rules by which a character can have more than one class", it's gone. Since you differentiate between multiclassing and dual-classing (which is just another kind of multiclassing by the second definition), then yes, they got rid of it. Also they got rid of dual-classing.
Then they put in a brand-new, unified multiclass system in 3e. Then a different one in 4e. Then yet another one in 5e.
4e is its own thing, but the 3e/5e systems are more like old school dual-classing than they are like old-school multiclassing, in that you level your classes separately, rather than in parallel. Are they a lot like the old, cumbersome, dual-class system? No, not really. The mechanics of dual-classing only made sense, inasmuch as they actually made sense, in the context of the leveling system of AD&D. Since 3e and 5e have substantially different leveling systems, they need multiclass mechanics that make sense in their context, and trying to shoehorn in the old dual-classing mechanics would likely either be a bad design or end up changing them into unrecognizability.
Yes, it's actually a significant balancing factor in multiclassing. Your multiclass levels delay your higher-level features, including fundamental stuff like feats and stat bumps.
And, whatever levels you're playing to, there are going to be features from your primary class that you never get because you multiclassed. You can get away with it and still be effective mostly because the classes are front-loaded, giving lots of stuff early on, which can make up for the lack of your higher-level features you'd otherwise have. Class feature delay is most noticeable up to level five, then multiclass can gain some ground into the mid-levels, and it starts losing ground again as you head for the level cap.
TL;DR Multi-classing isn't a problem. What is "abuse" or "broken" is subjective and other DMs and players may not agree. Almost all of the examples given below as concerns were actually eliminated with the 2024 rules revision.
Personally, I have had no issues with multi-classing. There are a few combinations that had particular uses or capabilities (describing them as "abuse" already implies that there is something wrong/bad about the combinations) but most of those were reduced or eliminated with the 2024 rule set. In fact, the biggest "abuse" in the 2024 rules was the spell Conjure Minor Elementals which was broken from the beginning and should never have seen the light of day. At least, they have errata'ed it.
In addition, different people enjoy making different types of characters. Some players like characters with a bad stat or two and critical flaws. Others like playing heroes without notable weaknesses. Some like to create characters that seem mechanically powerful to them. As long as everyone at the table is having fun, how you play doesn't matter - including whether multi-classing is allowed or not.
Finally, any DM that finds multi-classing to be unbalanced can simply decide to not allow it in their game. For anyone who doesn't share these concerns or have a few specific ones, they can choose to allow multi-classing or disallow specific combinations if they are considered problematic.
These are all 2014 issues and are all removed in 2024.
1) All subclasses are only available at level 3. This means that anyone trying for some combination based on subclass ability like the Hexblade, must wait until level 3.
However, even one level in a different class will slow progression in whatever your main class might be. A wizard which starts with a level of artificer or fighter for proficiency in constitution saving throws and the ability to wear armor and a shield ... also slows their spell progression. 3rd level spells won't be available until level 6 for this character. This is a significant consideration in actual play since the wizard with the 1 level dip is always a level behind where they could be and where other characters in the party might be.
2) Looking at your list specifically -
- a 1st level warlock can choose pact of the blade to attack with charisma and create a weapon. That is all. You also left off the other main attraction of a hexblade dip which was proficiency with martial weapons, medium armor and shields. The only thing now available as a feature is weapon attacks with ONE specific weapon using charisma.
- Smite is now totally different - requires a bonus action and is only 1 time/turn. Problem eliminated.
- Twin spell metamagic now only lets you cast a spell that can be upcast by one higher level. This does still work with hold person - but this works for a pure classed sorcerer so I don't see how it is a balance issue due to multi-classing. Twinned Haste and Disintegrate are no longer options.
I'm not sure you understand the multi-class spellcasting table.
A level 17 paladin/3 sorcerer is considered a level 12 spell caster ... compared to a level 10 spell caster for a level 20 paladin. Comparing these two ... the level 12 spellcaster has just ONE 6th level spell that the level 10 does not have - they are otherwise identical. I think it unlikely that one additional level 6 spell (while missing out on the level 18-20 paladin features) would be at all unbalancing. At lower levels the effect is equally as insignificant.
In addition, smite in the 2014 version was capped at 5d8 damage no matter what spell slot was used so the effect is even smaller since a 6th level slot does the same as a 5th level slot when used for smiting. (Perhaps, I don't need to mention that there are likely a lot more valuable things to use a 6th level slot on than smiting).
Finally, the 2024 smite uses a bonus action to cast the Divine Smite spell. In this case, there is no cap to possible damage so a higher level spell slot has some benefit. However, since it is a spell, it can be counterspelled and is subject to the usual limitations imposed by casting a spell.
As far as the coffeelock goes ... it has a very simple answer ... "DM says No." Utilizing a coffeelock requires the DM to agree to it. They have to allow the character the ability to take multiple short rests instead of a long rest AND allow the character to go without long rests so that the spell slot count never resets. Otherwise, the character is limited to whatever they can obtain from a couple of short rests in a typical adventuring day which doesn't unbalance anything in my opinion.
I'm not sure what you are getting at here. Characters are limited to ONE bonus action. Multiclassing may give them more options with what to do with that bonus action but that is all. In the 2024 rules, a monk can ALREADY dash or disengage as a bonus action for free ... there is no need to multiclass to a rogue to obtain the ability. Even the 2014 monk could ALREADY do this but it required the expenditure of a ki point for Step of the Wind. Multi-classing monk and rogue is arguably a waste of levels since it delays the monks progression in ki/focus points ... unless the character wants to pick up the expertise in skills offered by the rogue. Cunning action by itself adds very little for a monk.
A hexblade warlock can already use agonizing blast with hexblades curse (hexbades curse requires a level 3 hexblade warlock in the 2024 rules). In the 2014 rules, a multi-classed fighter could use action surge to cast it again which was pretty cool/powerful but it was possible only once/short rest since action surge is a short rest ability. Arguably, in the 2014 rules, the better combination was 2 fighter/X wizard so that they could cast 2 fireballs or something equally powerful.
However, the 2024 rules solve this issue ... the action from action surge can not be used for a Magic action. "On your turn, you can take one additional action, except the Magic action." ... so the 2024 rules prevent any use of action surge to cast multiple spells or use magic items which require the Magic action multiple times.
How is any of this "unfair to the DM"? The DM adjudicates the game, runs NPCs, sets up situations and provides fun for the players - the rules are just the rules. Since the DM is implementing the rules and running the game ... how can they be in any way "unfair to the DM"?
"Unfair to other players" is also a misunderstanding - if you have a group of players with different levels of understanding of the game and one of the players is using their knowledge to roflstomp enemies and make other players feel bad - that isn't a game problem, it is a PLAYER problem. The DM needs to have a chat with the player, possibly all the players and resolve the issue out of game. It isn't a rules problem ... it is a people problem.
It isn't clear to me that you have actually played these characters through a campaign (as opposed to a level 20 one shot) or are just theory crafting.
Barbarian/paladin - in the 2024 rules, smiting now requires casting a spell which is not compatible with raging since it prevents spellcasting. Every level of barbarian slows the progression in paladin or vice versa. Even just looking at the 2014 rules, extra attack would normally be at level 5 .. taking barbarian levels early slows this until level 7 and slows spell progression for smiting. Waiting to take the barbarian levels until later means that the combination doesn't really become available until level 7 or 8 depending on whether you want the aura or not. In addition, paladin already needs a high strength and charisma, raging is limited to medium armor ... so the paladin can't wear heavy armor and rage. This means requiring at least a 14 dex to optimize AC and a decent constitution. Unless your table uses rolled stats AND rolls incredibly well - a barbarian/paladin character is horrifically MAD :) (Multiple Attribute Dependent).
Wizard/cleric - yes a very good combination if you don't mind delaying wizard spells. Armor proficiencies are really useful for a wizard since their hit points are low. However, this delays wizard spell acquisition by at least a level. It also requires at least a 13 wisdom ... likely 14 would be better for skills/saves/modifier for spells. The character does get an emergency healing word and/or cure wounds, guidance (which has changed in 2024), bless and a couple other useful level 1 cleric spells. However, the focus is wizard and the character will always be a worse wizard than they would be without the level of cleric. Personally, I don't find it unbalanced and if a player likes the character backstory of a Cleric of Mystra deciding to practice magic instead of just worship it then awesome :)
Rogue/fighter/warlock - lol - "jack of all trades, master of none" - sneak attack goes up with rogue level, extra attack goes up with fighter or warlock level. Rogue and fighter need a good dex, warlock needs charisma, wisdom is useful for a rogue .. progression in each class is slow because it is spread over so many classes. I have seen VERY few triple classed characters in actual play and they are almost always "weaker" from a "power" perspective than a single classed character. You mention high AC ... but if the rogue is going to have a high dex then they don't benefit from medium or heavy armor anyway - if wearing light armor, the rogue will have the same AC as a single class. Go Arcane trickster and they will have the shield spell if they want it too (2024 rules). The rogue also has the mobility from cunning action and if you leave 2014 sharpshooter/great weapon master out of the discussion, the rogue sneak attack damage scaling keeps up reasonably with fighter and other multi-attack weapon users by itself - especially with features like steady aim or an arcane trickster using a familiar to provide the help action to obtain advantage on their attack. Basically, there is nothing unbalanced about a rogue/fighter/warlock character if someone wanted to play one.
In my opinion, 2024 rules actually addressed most of your concerns from 2014 ... concerns which many folks, myself included, did not have any issue with to start with. I don't think multi-classing is unbalanced.
As for your comments above ... I'm not sure you've read the 2024 rules ... paladin smite now scales with 1d8/spell level with no level limit ... so if they happen to have a level 7 spell slot then it can do 8d8 damage to one target, requiring a bonus action. Since a level 5 cone of cold will do 8d8 cold damage, con save for half with a level 5 slot - that seems like a balanced trade off. There is no requirement to use paladin spell slots. This is also incredibly bad compared to a spell like Conjure Minor Elementals which will do 5d8 on EACH attack with a 7th level slot (it used to be 8d8 but they changed it from totally broken to less broken ... it is still strictly better than Spirit Shroud).
Warlock pact magic hasn't changed - they still get all spell slots back on a short rest but have an additional feature that allows them to restore some spell slots in 1 minute once/long rest. If someone had some way to misuse warlock short rest spell slots before - it doesn't seem to have changed now.
Requiring 3 levels for all subclass abilities has reduced the combinations where one or two levels of a class gives access to some subclass defining features that could be used in combination with other class features. Three levels is a lot in terms of character development which is a good trade off for someone wanting to combine elements from different subclasses.
In conclusion:
1) Multi-classing is not the problem. It just increases options available - how those are used are up to the DM and the players at the table. A DM can always choose what to allow at their table.
2) A lot of players and DMs like multi-classing - what a DM thinks is "abuse" or "broken" is subjective and different DMs may not agree.
3) Most of the biggest concerns outlined were eliminated in whole or in part by the 2024 rules revision.
Not many people know the pain of needed 5000 exp in order to get lvl 2 on your F/MU(wizard)
"Sooner or later, your Players are going to smash your railroad into a sandbox."
-Vedexent
"real life is a super high CR."
-OboeLauren
"............anybody got any potatoes? We could drop a potato in each hole an' see which ones get viciously mauled by horrible monsters?"
-Ilyara Thundertale
That would be specifically 2 MU. You get to 2 Fighter at 4000xp ... so you'd be a 2 fighter/1 MU from 4000->5000xp. :) ... you'd also be either an elf or a half-elf ... the only non-humans that could be an MU ... though a gnome could be an Illusionist :)
Enough of us do, if your going to multiclassing get it down early when the costs are lower, but realize your going to pay costs for the life of the character. I have a suspicion that much of the multiclassing is really “whiteroom” exploration not actual game play over time. That’s why I think it’s probably a much smaller problem than the discussion makes it seem to be.
Wisea$$ DM and Player since 1979.
Tell me you never play martials without telling me you never play martials... none of these are better than straight classes because they all give up at least as much as they gain. The only real problematic thing in OP's post was Coffeelock, and I have literally never seen one in play because even if they didn't get shut down (which they did) they just became boring. The 2014 Hexblade dip was not ideal just because it was such a good choice, but it still didn't make you Superman or anything.
And? Straight bards can already do 2/3 of that, and there's no "cunning support abilities" you can MC for that are better than the benefits of just staying in bard to progress your features and get higher level spells. This is the kind of build that is absolutely multiclassing working exactly as intended. Not to mention that the table's powergamer optimizing their ability to make everyone else awesome is basically a best-case scenario.
Multiclassing is honestly very well done in 5e+. It's at the point where being more strict would shut down way more legitimate builds than "broken" ones.
My homebrew subclasses (full list here)
(Artificer) Swordmage | Glasswright | (Barbarian) Path of the Savage Embrace
(Bard) College of Dance | (Fighter) Warlord | Cannoneer
(Monk) Way of the Elements | (Ranger) Blade Dancer
(Rogue) DaggerMaster | Inquisitor | (Sorcerer) Riftwalker | Spellfist
(Warlock) The Swarm