Five Takeaways From the 2025 Monster Manual

The 2025 Monster Manual is the missing puzzle piece for Dungeons & Dragons' recent Fifth Edition revisions, with reworked monsters that hit harder and make combat more exciting at every level. Released in February, the new Monster Manual drives home many of the design choices made in other parts of D&D's core rulebooks. Building off of a decade's worth of lessons about how DMs use statblocks and how players tend to handle combat, the Monster Manual features creatures with streamlined abilities meant to speed up combat without sacrificing the "fun" of fighting in the game. Plus, the book includes a ton of gorgeous new artwork that depicts D&D's iconic monsters at their most threatening. Here are five of my biggest takeaways from the new Monster Manual.

1) Revamped Legendary Actions, With More Power Than Before.

One of the big goals of the new Monster Manual was to redesign monsters to have them punch harder but simultaneously make them easier to run. This design ethos can be seen in many revamped monster statblocks, especially at higher Challenge Ratings. Lair actions are now incorporated into the statblock, with monsters typically gaining access to an additional Legendary Resistance and Legendary Action while in their lair. Additionally, many of the Legendary Actions are much more powerful than their 5E equivalents, with creatures usually gaining more dangerous options.

For instance, all of the dragons have lost their functionally worthless "Detect" action and instead have access to new spellcasting options or more powerful attacks. The Adult Blue Dragon, as an example, can cast Shatter as a Legendary Action or it can cast Invisibility on itself and then move up to half its speed. While not as strong as the dragon's standard actions, the Adult Blue Dragon can now do a lot more over the course of a round then simply deal moderate amounts of damage and soak up hits from opponents.

2) Either Attack Rolls or Saving Throws, Not Both

Another major streamlining within rulesets is that monster attacks with effects are either triggered with a failed saving throw OR a successful attack roll. This should significantly speed up combat by reducing the number of rolls made during a game. As an example, the Bearded Devil's 2014 statblock included a Beard attack that damaged on a successful hit and forced its target to make a Constitution saving throw or be Poisoned. In the 2025 Monster Manual, the Bearded Devil's Beard attack deals damage and automatically inflicts the Poisoned condition on a successful attack.

There's two major consequences to this. The first is that only one dice roll is needed to determine the success or failure of a certain attack or ability. The second is that a creature is more often able to threaten player characters at their intended level. By having a creature's full attack trigger based on a single success instead two successes (or I suppose a success combined with a separate creature's failure), it radically changes the dynamics of many D&D combats.

3) Yes, The Art Is Fantastic

Keeping with another theme of the 2024/2025 Core Rulebooks, the artwork in the new Monster Manual is frankly fantastic. There are a lot of D&D players, myself included, who love to look through the Monster Manual and other bestiaries primarily for the art and lore. Those players should be more than happy with this new book, which contains artwork for every single monster in the book. What's more, much of the artwork shows the monsters in action. The Chasme, for example, looks much more threatening in the 2025 Monster Manual, with art showing the demon hunched over an adventurer with its probiscus covered in blood. Compare that imagery to the 2014 Monster Manual, which just has the chasme standing in profile.

One comment made to me by Jeremy Crawford was that Wizards had found that monsters without art tended to be used less often, so I'm expecting the trend of more art to continue in future books.

4) A Handful of Interesting New Mechanics

While not found widely in the new Monster Manual, there are a handful of new (or at least very uncommon) mechanics. The Empyrean, for instance, has a Sacred Weapon attack that deals damage and Stuns its target. However, the target can choose to bypass the Stunned condition by taking additional damage. Meanwhile, the Arch Hag has multiple abilities that curse their opponent, taking away their ability to use Reactions or spells with verbal components. Additionally, the hag has a bonus action that deals automatic damage to anyone cursed by the witch.

Finding new mechanics in the Monster Manual is rare, but they represent some interesting innovation that hopefully will be incorporated with future statblocks. Not every creature needs stacking abilities, or "pick your poison" choices, but I love these and want to see them more often in the future.

5) Species-Free NPCs

Over the past few weeks, Wizards has revealed several monsters with new creature classification types. Goblins, aarakocra, lizardfolk, kobolds, and kenku are all now classified as non-humanoids. It's interesting that non-humanoid species often have multiple statblocks with unique abilities, but that the humanoid statblocks are meant to include elves, dwarves, orcs, humans, and more. I'm assuming (given that Eberron: Forge of the Artificer is bringing back the Warforged) that D&D won't remove non-humanoid species as playable species, but it feels like there's a deliberate push to make all humanoids interchangeable, at least when it comes to these NPC stats.

It's a shame that Wizards seems to have done away with templates in the new Monster Manual because they'd be useful for transforming a generic guard or scout into a Drow guard or a Dragonborn scout. I don't think these would be hard to homebrew if necessary, but I do feel like this is one of the bigger misses in the Monster Manual. Hopefully, we'll see more specialization in the future, and the Monster Manual opted to focus on monsters instead of highly specific statblocks.

 

[Reynard]

Has not been my experience. A Mind Flayer with several lower CR minions might pose a threat, but a single mind flayer against 3 level 5 PCs (caveat: my group knows what it is doing)? Nah, that's a dead mind flayer.

Yeah, I have experienced players and barring terrible dice, they steamroll even Double Deadly solo encounters all the time. Solo monsters in 5E DO NOT WORK.

 

[Ruin Explorer]

caveat: my group knows what it is doing

That can be a pretty big caveat depending on how much weight "knows what it is doing" is bearing!

In my experience people have used that sort of phrase to describe everything from a group that isn't actively incompetent/Three Stooges to a group that knows the rules backwards, knows every monster and its capabilities backwards, and plays well-optimized characters in a carefully-coordinated and expert way.

Solo monsters have literally never worked well in any edition of D&D though, I would assert. It's just more obvious now - and frankly a lot more players play well tactically now and make smart decisions, as a combination of aging and more experience of tactical games.

 

[Doctor Futurity]

There's a logic to it.

1. Wizards did polling on their players and found that a lot of their players play species and races that were in the past not playable because those species and races will only seen as villainous or evil.
2. Wizards also did research into their past and found that some third party or older books would paint these past only evils species in the cultures of real life cultures.
3. Once they decided to make those species playable, They would either have to include the other species as monsters to be killed in the Monster Manual or remove all playable species from the Monster Manning book as long as they're humanoid.

Wizards chose not to make a human or halfling stat block for you to stab in the Monster Manual. They decided that any humanoid stat block that reference a specific species would have to be setting dependent which matches how their current player base thinks.

Blame Warcraft and lazy artists of the past.

This is a very specific approach to the matter with WotC. To contrast, Pathfinder 2E has handled this exceedingly well and stands out for providing all the tools a GM could need for NPCs of various ancestries (and custom NPC stat blocks in the NPC Core, no less) while also meeting the expectations that your table will manage it as works best for you.

I wouldn't blame Warcraft, though; no one playing Warcraft who also plays D&D is thinking somehow that you can't realistically fight humanoid opponents; that's a core conceit of WoW, after all. Indeed, they might be rather peeved that D&D seems to have tried hard to make its orcs distinctly different and weird from both WoW and Tolkein's orcs.

 

[Reynard]

That can be a pretty big caveat depending on how much weight "knows what it is doing" is bearing!

In my experience people have used that sort of phrase to describe everything from a group that isn't actively incompetent/Three Stooges to a group that knows the rules backwards, knows every monster and its capabilities backwards, and plays well-optimized characters in a carefully-coordinated and expert way.

Solo monsters have literally never worked well in any edition of D&D though, I would assert. It's just more obvious now - and frankly a lot more players play well tactically now and make smart decisions, as a combination of aging and more experience of tactical games.

On the upside, since solo fights tend to be boring anyway, adding minions, side antagonists, traps, angry terrain, etc... just makes for a more fun experience anyway. Win win?

 

[Doctor Futurity]

That can be a pretty big caveat depending on how much weight "knows what it is doing" is bearing!

In my experience people have used that sort of phrase to describe everything from a group that isn't actively incompetent/Three Stooges to a group that knows the rules backwards, knows every monster and its capabilities backwards, and plays well-optimized characters in a carefully-coordinated and expert way.

Solo monsters have literally never worked well in any edition of D&D though, I would assert. It's just more obvious now - and frankly a lot more players play well tactically now and make smart decisions, as a combination of aging and more experience of tactical games.

True that. I guess what I was trying to say is: my group is old, experienced, and thinks tactically, and I have more than one rules lawyer in the group. They will find such an encounter trivial unless I stack the odds against them, and this has been an ongoing issue for most of the years I have run 5E, which is simply that it does not feel challenging to them.

For my purposes, I would not make the argument that a mind flayer by itself vs a similar group in Pathfinder 1E, D&D 3.5 or AD&D (1 or 2) would have been a push over; and indeed in the old days a single mind flayer against a group of level 5 PCs was definitely a challenge for them, even if the group was tactical-minded and experienced.

 

[Ruin Explorer]

On the upside, since solo fights tend to be boring anyway, adding minions, side antagonists, traps, angry terrain, etc... just makes for a more fun experience anyway. Win win?

I agree.

I think really the whole thing is a trope that people keep trying to replicate, that is really more of a videogame and movie trope than it is inherently a trope of fantasy fiction or TTRPGs. People feel like there should be one "boss monster", and hence we get stuff like Legendary Actions and so on to try and give a single monster more of a fighting chance.

Maybe if people want it, TTRPGs should keep trying to get it to work, maybe that's fair, but I don't think D&D has done great with it, in practice.

 

[Celebrim]

In Skerples's magnificent book The Monster Overhaul, he has something really insightful to say about orcs. Quoting, but edited:

I think that's brilliant, and very true. The only thing I would add is that either way Orcs are "humanoid" because literally just look up that damn word AUGHHH

It's strictly true, in that orcs are either not people or else they people. But it's also misleading.

To talk about it, I'm going to move from talking about orcs to goblins (and probably more Star Trek and sci-fiction).

The trouble is that being a person doesn't necessarily imply being human. For lack of referents, let's imagine that humans are no more capable of being evil than is average for a sentients. I think we will all agree humans have a variety of character flaws that led them to acts of evil. We might not agree entirely on what those acts are, and we might argue about the origin or correctability of those flaws, but we all agree that they are there. One problem that the "biological essentialism" crowd runs into is that it's completely plausible and reasonable to link our maladaptive behavior to our biological imperatives and origins. Greed was a successful survival strategy because it meant in situations of limited food stuffs, you and your descendants were more likely to have adequate nutrition. A lot of what we think of as evil can be justified biologically, either in a modern context or especially in the context of our own evolutionary history. Now, I don't mean to suggest that makes evil less real, but I do mean to suggest that think you can do away with something like "biological essentialism" in a species description is not thinking past the first step in your thinking.

If you look at Star Trek, you have a lot of things that are near human. These nearly human things have a lot of the qualities we associate with being human and as such are near enough to being human that morality should suggest that they deserve the sort of treatment we would give another human. "Vulcan rights" are going to be really close to "human rights", and there logically exists some sort of blanket "person rights" that all people near enough in nature to recognize each other as persons would possess. But within this framework we see various differences in the propensities for evil and how that evil is expressed. The biological essentials of Vulcan, Andorran, Human, and Klingon lead to different tendencies toward evil.

It's not hard to imagine that there exist among sentient species, species with sufficient biological differences and sufficiently different accompanying world views that they either can't successfully communicate or can't agree on what rights the other deserves. Indeed, it's easily possible to imagine a sentient species so biologically different that it can never recognize you as a person no matter how hard you try to treat it as one. The concept of shared personhood you are trying to communicate is not one it recognizes precisely because of its biological differences. Card refers to this concept as "Varelse", too alien to have any communion with because you lack any shared context.

But it's also possible to have "near Varelse", beings that while they are closer to you than a true Varelse are still so far from you that any shared framework or shared goals or mutual respect is basically or at least normally impossible. The vast majority of time the near Varelse treat you as non-persons, no matter how much you attempt to teach them otherwise, because their biological essentials and resulting culture tends to make them behave in a way that is evil. (Or if you don't believe in evil, then in ways that invariably don't involve mutually beneficial interaction, but which tend to treat you as prey to exploit.)

That's Goblins in my game. While they are "people" and you can have a relationship with them and some of them are even honorable, for a variety of reasons in their background including biological reasons (they are a product of selective breeding program to reinforce certain traits and propensities) most goblins do live up to their negative stereotype as vicious people eaters unconcerned with treating anyone else as a person. They are obligate carnivores that tend to not understand you as anything other than "food" and really don't care if you have feelings.

There are no orcs in my homebrew. There are only goblins. People fear them for a reason. You can treat them like people, and the better ones will respect that, but don't assume that the recognition is reciprocal. The vast majority of them are fully on board a campaign to genocidally wipe out and eat every other non-goblin person in the world.

And this doesn't even get into the fact of what does "biology" mean in a world that doesn't run on science. Having a supernatural origin in a fantasy world is the norm. No one has a purely "natural" origin in the sense that a person who doesn't believe in anything but a material cause would understand the term. Saying that a fire elemental can have an essential nature as inherently evil because it's not a natural biology is meaningless in the context of a fantasy universe. The persons in my homebrew game that have an essential nature that lets them choose between good and evil have that nature explicitly as the result of an accidental gift by the gods that gave them that nature. They have true freedom to choose right or wrong precisely because that is the way they were made. The vast majority of things in the setting lack that freedom in whole or in part, and are slaves to their biological essentialism that forces them to behave in a single way and gives them no emotional context for behaving in any other way. Player characters are part of a limited range of "free peoples" with perhaps in some sense more free will than the deities that made them. That's my setting and I find it coherent. I don't necessarily find other people's settings coherent or thoughtful, but on the other hand, it's their setting and they can do what they enjoy. But don't try to stuff down my throat the ill-considered insanity that you can't be a person and have a biologically dictated essential nature, because that's not how any of this works.

UPDATE: The trisolarians from "Three Body Problem" and the Ruml from "The Alien Way" are examples of near Varelse. In fact, both settings seem to suggest that on average, any two alien species will always be near Varelse and be unable because of biological imperatives to form stable relationship with each other. In fact, Liu Cixin basically suggest in this series that because of biological essentialism, any sentient species that fails to act and think like near-Varelse and accept this situation as normal goes extinct as a Dodo in short order for failing to recognize it is to everyone else just food (or a threat).

 

[occam]

Five is the one that bugs me the most, and I don't understand why you couldn't have both the generic templates and a few specific creatures representing each of the major species (a dwarf warrior, an elf scout, an orc shaman, a human priest, and so on). It's very, very strange to have a MM that doesn't include orcs or elves.

Just noting here that the 2014 MM also didn't include stat blocks for elves (or dwarves, halflings, gnomes, etc.), other than the more "monstrous" versions like drow, duergar, or svirfneblin. Not including orcs, now that they're in the PHB, is just being consistent.

 

[evilbob]

I guess what I was trying to say is: my group is old, experienced, and thinks tactically, and I have more than one rules lawyer in the group. They will find such an encounter trivial unless I stack the odds against them, and this has been an ongoing issue for most of the years I have run 5E, which is simply that it does not feel challenging to them.

These are good points; maybe the 5.5 MM was balanced against a new set of players, and that's why it seems easy to us old-timers? And that's fine, too! It might have been helpful to have a little aside that was like "hey folks, if your group has been around the block, -2 per tier from all CRs" or something, but you can't make a product that completely satisfies BOTH inexperienced AND experienced users, and it's much more important to get the new players' experiences right.

 

[Ruin Explorer]

For my purposes, I would not make the argument that a mind flayer by itself vs a similar group in Pathfinder 1E, D&D 3.5 or AD&D (1 or 2) would have been a push over; and indeed in the old days a single mind flayer against a group of level 5 PCs was definitely a challenge for them, even if the group was tactical-minded and experienced.

I feel like in an awful lot of those, a solo unbuffed Mind Flayer absolutely would get chunked/wrecked in a single round, just depending on initiative. You seem to be assuming that it would get initiative and get its Mind Blast off. Like in 3.XE, that's definitely all it comes down to (well, that an if the entire party is casters reliant on spells for damage/CC that's also a problem, but unlikely). It gets that off and hits multiple PCs? Then it's quite likely 1-3 PCs are completely out of the action for the entire fight and there's nothing that can be done about it, which improves the odds considerably. If the PCs are spread out, probably it's only one that it can get, it gets chunked. If the PCs win initiative, and focus on making it sure it can't do anything (and likely spread out too), it's going to get chunked.