With that rather… quotidian title, I’m launching into what promises to be the biggest midstream change in all of 5e. (We’re not really looking for something else to top this.) During the Future of D&D panel, they discussed the fact that NPC stat blocks are losing the big beefy spell lists that you see in, for example, the war priest, the mage, the archmage, and the lich. In place of lots of spells, they’re getting custom actions that do some kind of flashy magic stuff, but they’re specifically, explicitly not spells.

In this post, I’m going into the reasoning behind all of those (from what I know – please note that I don’t work for WotC), what I think it’s doing right, and what I think we’re losing in the exchange. To get this out up-front, I’m in favor of some change, but I don’t think this is the right change. To shed some light on this, I’ll be talking about LARP design issues as well, because the boffer LARP world has solved various aspects of this problem for years. Strap in.

This is always true, but I want to call it out here up front. It’s completely okay to disagree with every word of opinion in this. I’m setting myself up for a lot of people to disagree – please do me the favor of avoiding vituperation to the same degree that I have done here.

What’s Even Going on Here

Let’s start with a visual reference: the war priest from Volo’s and the war priest from Mordenkainen Presents: Monsters of the Multiverse. (And you thought Volo went in for grandiloquent titles.)

What you notice right off is that the Spellcasting trait has been chopped way, way down, from 26 spells to 10, and those spells are cantrips or “1/day each,” without spell slot levels. That’s a lot less spellcasting, though NPCs are generally only expected to survive for three rounds of combat.

Those three rounds of combat are a decision-making problem, from a design standpoint: the stat block doesn’t tell DMs how to make good decisions in those three rounds to hit the target damage output. Should they wade in with their maul? Probably not – an average of 20 damage per round (I’m oversimplifying what “average” means, but stick with me a minute) is about a third of the expected damage for a CR 9 (that’s also oversimplifying considerably, but I don’t have the internal calculations to get this one right). What I’m saying is, something needs to change here, no question.

In the new stat block, the war priest picks up a new action: Holy Fire. Holy Fire isn’t a spell in any formal sense.

• It doesn’t have a level.
• It isn’t called a spell.
• It isn’t recognizably the same as any existing spell. (We would generally call a cantrip that blinded its target substantially overpowered.)
• It doesn’t have V, S, or M spell components.
• It’s not called out as magical, it doesn’t imitate a particular spell, and it doesn’t use a spell attack; by the reasoning of Sage Advice, it is therefore not magical.
  • I am aware that this might be seen as a quasi-malicious reading; I’m trying to make a point that tagging things as either spells or magic gives the rules a much better understanding of how to handle effects.

Okay, so it’s not a spell. Does that matter? Well… yeah. It has a ton of knock-on effects.

• By not having a level or being called a spell, it can’t be countered with counterspell (I’m going to get into this a lot more in a moment), and an Ancients paladin’s Oath of Warding doesn’t halve its damage.
• By not having a level, it ignores globe of invulnerability and the Spell Thief feature of the Arcane Trickster. It incidentally also ignores a rakshasa’s immunity to spells below a certain level, though we can safely say that doesn’t super matter.
• By not being magical, it ignores antimagic field and any areas of dead magic, while armor of invulnerability does apply to this damage.
• By not being magical or a ranged attack, Holy Fire ignores wall of force – there’s nothing indicating that it passes through the intervening space.
  • When we get into special actions that have no “is magic” signifier and don’t require sight, such as the Psychic Crush of the Draconic Shard (FToD), I’m pretty sure Leomund’s tiny hut and prismatic wall are casually circumvented. That’s a CR 17 creature, so maybe that’s fine, but my point is that there are a LOT of actions like this now, and there are about to be a lot more. Nerf LTH all you want, but uh. Prismatic wall should stay pretty cool.
  • Edited to add: It has been brought to my attention that Psychic Crush, as a sphere, is stopped by prismatic wall, through its nature as a wall. It’s super easy to miss this rule interaction, so I don’t feel that bad about it.
• By not having a V component, it can’t be blocked with a silence spell, a gag, or anything else that blocks V components.
• By not having an S component, it doesn’t require empty hands.
• By not having an M component, you can’t stop the war priest from using it even when you’ve taken away every single thing they have.

That’s the mechanical side. I’ll absolutely grant that in most combats, the M component doesn’t matter – plenty of spells don’t have M components. I suspect that a lot of designers (sometimes including me) don’t think about the potential to be taken away if an NPC becomes a prisoner as a narrative wrinkle in their spell’s mechanics.

For the record, I think it’s very important to be able to take prisoners, since “we have no way to handle you except killing you” isn’t the style that D&D wants anymore.

There’s also a deeper level where mechanics inform the narrative. Tl;dr: spells aren’t just verbs, they’re a presence in the narrative that delivers nuanced meaning.

• The implicit narrative of the war priest is that it’s the NPC answer to a PC with the War domain. That is, the War domain is something that exists in the setting and can be discussed and understood. For ease of use, let’s call any mechanical thing that can also be discussed and understood by the characters a narrative object.
  • I anticipate nitpicking: no, the VGTM war priest isn’t an NPC 1:1 of a PC War domain cleric of any level. It’s sort of an impressionistic, simplified-for-play version. It’s not going to have as many use-per-day features – if it would get those things 3+ times per day, it probably gets those all the time instead. NPC versions of a lot of subclasses exist, though – each of the schools of wizardry, several of the roguish archetypes, and so on.
  • So yes, the new war priest is “just” several more steps simplified from the PC-facing War domain. The problem is that its gameplay loop no longer has anything in common with a PC, and if you’re matched up against one (it’s pretty likely! War is in the name!), you’re not going to see anything recognizable. If the DM doesn’t announce “this is a war priest,” you’re not going to figure it out by seeing someone similar to you. I think that moment of “oh you’re like me!” is powerful in games.
• Spells are also narrative objects, and in D&D’s Vancian magic system, they’re quite significant ones. In D&D as in the Dying Earth stories, if you see someone cast a spell, it follows that the spell exists and, if you belong to the same class, you could learn that spell. If that NPC and you are both wizards specifically, you can hope to take it off their soon-to-be-cooling corpse.
• There are spells – especially when you get into the late game – that are hard to see a PC ever using. Mislead comes to mind. My point here is that I love it when NPCs build a whole plan around a spell that has been on the books for the last roughly 75 editions of D&D but no one has ever seen used. The point I’m making here is not, of course, that you can’t just make an NPC use that weird spell – I’m just trying to establish for later use that NPC stat are erasing the “is an Xth level spellcaster” to suggest other spells they might sub in.
• Because spells are narrative objects, they can also be part of the players’ tactical and strategic predictions. If you know that you’re fighting a sorcerer, wizard, or Fiend warlock, you don’t cluster for the fireball. That kind of thing. Especially as you move into tier 3, I think the ideal is to face villains that you increasingly understand as people. You have some idea of what they can do and plan accordingly. This can still work in the new model – you’re just going to find yourself ignoring the stat block as written and granting the NPC additional spells.

 Counterspell

“Everyone hates playing the blue deck, everyone loves playing with the blue deck.”

–Will “Thagomizer” Kotas

Twitter has been abuzz of late in conversations about counterspell – that is, the specific awareness that cutting counterspell-focused play off at the knees is either a goal or an acceptable cost here. (It’s a goal. Olé olé olé.) To keep this pretty brief, the problem with counterspell really comes out when one side of a fight has an advantage in the number of characters who can cast it, and enough spell slots available to just keep doing so.

Nuances of the slot-level-bidding game aside – and they’re not nothing – the outcome is that the side with the greater number of counter-casters can lock down the other side’s casters. One side gets all (or almost all) of what it wants, while the other gets none of what it wants. In an attrition-focused game like D&D, you’ve at least got slot attrition happening, but the savings in healing spells alone more than pays that back.

There are ways to avoid the counterspell dynamic – sticking to long range (counterspell’s range is only 60 feet), Subtle Spell, any amount of invisibility or other sight blockers (counterspell’s trigger specifies sight… but so do a lot of attack spells), and reaction denial generally. WotC doesn’t like seeing counterspell occupy that much metagame space, though, and I think they’re right about that. It’s not a huge thing at every table, of course, but if it isn’t big at your table yet, that might just be a matter of time.

For a personal note, in the most recent session of my homebrew campaign, the party’s multiclass cleric/wizard used some clutch counterspells when an enemy mage joined in a big throwdown. There were an equal number of counter-casters on each side, which I think was important. The PC doesn’t have enough slots to counter everything, so he picked his moments – that’s the ideal dynamic. He felt good about it, the tension of the scene was maintained, all was well. But you can see, I think, how changing to a mix of non-counterable not-spells and a very small number of counterable spells means the positive dynamic is unlikely to materialize.

Another ideal case, of course – Scanlan Shorthalt countering the Divine Mister V at the end of Critical Role’s Vox Machina campaign. Still one of the most moving single actions I’ve ever witnessed in a game. If you’re not familiar with CR or how that went down, you kind of owe it to yourself to at least get enough context to watch that moment.

The quick way to sum up what WotC doesn’t like: zeroing out the damage threat of spellcasting enemies by seizing an action-economy advantage. Changing something about how counterspell works in the 2024 release is essentially inevitable.

Previous Solutions

Baldur’s Gate: Descent into Avernus showcases a previous attempt to solve the data-management problems of spellcasting enemies. The black gauntlet of Bane (CR 6), skull lasher of Myrkul (CR 1) and master of souls (CR 4) are all good examples. For their Actions, each has one or more spells written out in a somewhat abbreviated format. For example:

Chill Touch (Cantrip). Ranged Spell Attack: +6 to hit, range 120 ft., one creature. Hit: 13 (2d8) necrotic damage, and the target can’t regain hit points until the start of the master of souls’ next turn. If the target is undead, it has disadvantage on attack rolls against the master of souls for the same duration.

The master of souls uses similar notation for ray of sickness and scorching ray, which incorporate At Higher Levels in, again, abbreviated notation. I can’t say specifically why WotC regards this as a failed solution, but overall I’m guessing that it’s a ton of page space in exchange for it still being hard to make the decisions the NPC’s designer expects you to make.

I’m not at all convinced that “what NPCs are using isn’t a spell or even magic” is an answer, though.

LARPs Have Solved This for Years

It’s okay if you don’t want to dig into this. The tl;dr of it is:

• Think about shifting counterspell out of being a spell and into being a feature or feat. Maybe its usage is “half your proficiency bonus per long rest” or something, I don’t know. Then NPC spellcasters might get one, legendaries 2-4.
  • The only reason not to chop counterspell outright is how damn good it feels to tell the enemy caster NO. Power word no is too perfect of a power fantasy to completely shut off.
• Keep the new special actions fairly recognizable as spells PCs can use, even if they have different damage expressions.
• Tag the new actions as spells with a placeholder spell level, so that they interact with features appropriately.
• Add spell level notation to each of the real spells in the Spellcasting trait, just for ease of use.
• I know this is rough on space, but if you have a tactical plan for the stat block, find a place to say it. We’ve relied for way too long on DMs to read the designer’s mind. Let’s try a brutally stripped-down notation – 1, 2, and 3 as superscript could work. If this reminds designers of non-spellcasting stat blocks to include tactical variation, so much the better.

Going back to NERO and SOLAR (1991 to present), games needed easy ways to send out NPCs as packet-throwers. This meant making it easy to teach NPCs to play spellcasters, without tons of practice in learning spell effects or special spell incants. NERO and SOLAR achieved this with a very narrow variety of standard spell verbals (“With mystic force I Pin you,” “I call forth [damaging spell name]”). This is narrowly solving the data burden of spell lookups. They didn’t have a counterspell cognate, but they did have some very significant buffs that blocked magic spells. Not important right now, I just don’t want to be accused of leaving stuff out.

Flash forward to Shattered Isles (1996-2006), which uses a tagline system for spells (a more complicated but still manageable approach to teaching players what they’ve just been hit with), requires custom incants for six of the seven types of magic, and introduces Countermagic. Spell taglines are things like Fire Dart, Stone Arrow, Rootfoot, or Rend Spirit. That winds up being a lot of taglines to learn, but there’s a set of maybe 10 or so you’re going to hear SO much more often than everything else that you learn them pretty quickly, and people try to be chill about mistakes.

The seventh type of magic, the one that doesn’t have incants? Yeah, unsurprisingly, that one is by far the most used for sending out NPCs. In a nice twist (functional design), that type of magic also has the majority of the really evil baddies. The spell verbals for THAT school are fast and simple: Rend Spirit Tal Shar. Cause Mortal Wounds Tal Shar. Fine family fun.

SI’s Countermagic is a skill that casters could buy once at Fourth Circle (very advanced, not yet a master), and that was good because every spell they countered cost them 5 mana. So probably more mana than it cost the original caster, but hey. When you’re countering a Rend Spirit (an instant death spell), who really cares about one measly point of mana difference? This is a game where 40 mana is a master spellcaster’s daily allotment, though, so you wanted to be conservative with countering unless you had a LOT of expendable mana recharging available. In principle, if you had your full daily allotment of mana and got up in someone’s face, you could lock them down with Countermagic while your buddies casually murder them. The only thing that stops that is the caster’s buddies acting like linebackers.

Oh, and there’s also a tagline modifier that means the spell can’t be countered or blocked with Shield, Ward, or anything else. That tagline is “Arcane.” It doesn’t show up much in SI, mostly in the hands of elder gods. The implication is that it’s top-end elder-god level stuff; that status is going to get diluted somewhat over time.

Next comes King’s Gate (2001-2011), which mostly shares SI’s rules, but adds a “spell preparation” rule for 5^th-circle spellcasters who would otherwise have incants. They can lock in a number of activations of particular spells, basically pre-spending their mana. If they do, instead of an incant, they can just deliver it as “Magic [tagline]” (Magic Exsanguinate was the first of those we saw.)

Digression

This is where I want to explain one of my deep biases. It’s a key conceit of SI, KG, Eclipse, and DtD that the rules present basically how people work. (Not humans – all of these have non-human options – but people.) Statting can work differently for NPCs than for PCs in a lot of ways, but if there’s a fourth Dodge, that means something. If you’re able to cast spells with a “Magic [tagline]” delivery, that tells the players something about the narrative of who that character is, and the players are expected (on some level, it’s not a bar exam) to pick up on those cues. If a character in SI or KG was observed to cast both Fire Lance and Heal Grievous Wounds Tal Elan, something extremely weird is happening and you need to take notice.

I want there to be payoffs for PCs who pay close attention to what NPCs do and how they do it. There aren’t nearly so many conflicts of magic in D&D as there are in SI/KG, and that’s fine – but if they’re using both high-end wizard-only spells and high-end cleric-only spells (for instance) that should deliver clues. Maybe it just means this NPC gets to be a 34^th-level character! But that’s still useful to know. Or maybe you know that it’s a cleric but they drop a fireball – oh hey, guess they’re probably a Light domain cleric.

Un-Digression

I could be wrong about this, but I want to say that KG was also a little more adventurous with slightly magical monsters that have a narrow range of magical effects… they just don’t use incants because the magic is inherent. This is reasonably equivalent to a special action, except that it’s still in a tagline system with a limited variety of possible effects.

Eclipse (2007-present) stripped out incants completely and changed up Countermagic. (I’m going to keep calling it Countermagic rather than the more correct Counterpsi so that my terminology forms a throughline.) In Eclipse, Countermagic is a defense that you can buy up to three times after you meet certain prereqs, and that defense resets on a 10-minute short rest. Stripping out incants and changing the taglines to the ultra-simple “[tagline] Ka’Tul/Ba’Tul” is a huge help in making it easy on NPCs who have very little time to practice for the role.

Dust to Dust’s core change to this dynamic is the distinction between magic and nonmagic. In SI, KG, and Eclipse, some creatures have Natural as a tagline modifier. For obvious reasons, a Natural tagline can’t be countered or warded, but it interacts a bit differently with armor, shields, and other taglines. One of the most common taglines of that type is Natural Entangle, used by giant spiders the world over. We weren’t into that – we wanted to make the point that there’s no separation between ordinary fire and magical fire, and so on through each Realm of Energy and Form of Matter. (Only weapons are different.)

As a result, DtD spiders used Fluid Entangles, and characters used Countermagic to fend off part of a cave-in. But we also didn’t have “this creature requires magic to hit” as a rules concept. I think we benefited from Natural not meaning “mostly better than magic.” I think 5e going all the way toward “magic/nonmagic is not a dichotomy” would abandon a lot of things that are essential to D&D and make some kinds of communication harder, so I don’t recommend that, but if you’re willing to think through it as a cosmological statement, it can work.

Conclusion

Most of the conclusion I want to get across is in the tl;drs above. I want to see tactical complexity, especially in the areas of preparation and counter-play, stay robust in D&D – I strongly believe that learning about an enemy so that you can prepare for them is part of the interest of the game, especially in Tier 3 and 4 play. I think mechanics deliver narrative, and some of that is fundamental cosmology. I don’t care about every NPC’s fireball being 8d6, +1d6/spell level, or even every NPC’s fireball being 20-ft radius. I’m more interested in how the effect being a fireball delivers narrative to the players, even if the rules tag it as:

Schmendrick’s Fireball (spell, fireball, 3^rd level). [rules block for Schmendrick’s fireball goes here]. That way, maybe it’s still stopped by a helmed horror’s immunity to the fireball spell, as it should be if it’s “supposed” to be a fireball.

To strip my expression of my thinking down even more: simplify presentation if you have to, but narrative comes before everything. Choose the narrative over getting the damage totals tuned just right, if that’s what it takes.

What they’re doing in Wild Beyond the Witchlight and Fizban’s Treasury of Dragons (and what they’re signaling for Multiverse of Monsters) points in a good direction. The problems that remain are problems, and they need to resolve those (or show us more about how they’re supposed to settle out) to get to a place where an ordinary DM can understand how the game should work in 2022, 2023, and the early part of 2024. “A bunch of defensive magic is just broken for the next few years” is not ideal, and it’s hard to see how some future core rules formulation would resolve everything here.

---

You can now buy Under the Seas of Vodari in PDF!

Under the Seas of Vodari is a 300-page book that provides 5E/Fifth Edition material for both gamemasters and players to dive into a world full of action and adventure set beneath the waves of Vodari or your own campaign world. This book is a follow-up to The Seas of Vodari, a setting book filled with swashbuckling action and magical adventure. Dive into your first adventure today!