Mike Mearls Games
patreon
Improved 5e: Monster Levels
🕑 Added 2025-09-16 12:00:43 +0000 UTCI literally can't think of anything that challenge rating does well.
Let's back things up and think of how a good system for rating monster strength would work. I think it should do the following:
It should be driven by character level. The rating I use to measure a monster's strength should be easy to understand with respect to character level.
It should allow me to set up fights that are easy, hard, or somewhere in the middle.
It should be driven by the characters' power. Monsters exist to challenge the characters. There's a metagame between what PCs can do and what monsters can do to counter that. I don't want this to be a dull set of creatures that neutralize everything the characters can do. I do want monsters that have sensible defenses against what PCs can use.
CR does none of these things. Quick, what range of CRs are a good match for a 5th level party? How many monsters should I normally put into a fight to match up with a party of four characters? What if someone can't make it that night. Can I easily adjust to three characters?
4e got this right by organizing monsters into levels. AD&D did this too! The 1e DMG sorted monsters into by level, including guidelines for how many monsters to throw at the party (usually represented by a range, like 3 - 18 or 1 - 6). Even OD&D featured an early version of this system and advised DMs to adjust the number of monsters appearing based on the party's size.
Building Better Monsters
To start with, we need to understand what we're going up against. Monsters are defined by characters. My monster design can be elegant, but if they are unable to threaten the characters they are useless. So what makes a good monster? I think we need three things.
Durability. Monsters need to absorb what the characters and throw at them. They need the hit points to stand up to attacks, and they need an offense that threatens the characters. If a monster is supposed to threaten a 5th level character, its damage needs to back that up.
Scalability. Some monsters, like an orc or goblin, are meant to fight in groups. Others, like a dragon, need to fight alone. The system needs to account for this by understanding what a monster needs to survive when it is outnumbered. A monster's action economy, its defenses, and its attacks need to serve its roll as a member of a mob or a creature that fight alone.
Ease of Use. The DM needs to pilot several creatures, manage the environment, and keep track of the rules. That's a lot of work! The players have a built-in advantage in that they are a team of specialists taking on a time-strapped generalists. My goal is to design specialist monsters with fewer but more powerful abilities. I'd rather give you three dragons and then give you advice on which one is the best fit for your group's composition than one dragon with a full page of options. In my experience, that massively complicated dragon still fails to deliver.
Where to Begin
Let's start by giving a monster a level. I built a spreadsheet using a baseline fighter built with the following assumptions:
A +3 Strength bonus at 1st level, improved with feats to +5. After that point, I treated each feat as a +2 bonus to damage rolls.
I used the battlemaster subclass as the baseline for subclasses. I think that the battlemaster's ability to alpha strike (drop action surge and maneuver dice to spike its damage as high as possible) is a better model for how optimized groups function.
I created two damage tracks. One assumed that the fighter used no limited resources like maneuver dice or action surges. The second track assumed an alpha strike, with the fighter burning through resources to throw as much damage down range as possible.
I used the prototype fighter's AC and attack bonus as the baseline, with a rolling bonuses added in based on tier to reflect magic items.
I think used that character's stats as the baseline to generate what monster stats needed to look like to hit the following benchmarks:
A monster that, alone, could provide a challenge to the PC. In this one on one fight, the monster has enough HP to survive 2.2 attacks and deals enough damage in two actions to reduce the fighter's hit points by 2/3rds.
I used that baseline stat block to generate two other baseline stat blocks:
A monster that could outnumber the fighter 2:1 and pose a threat equal to the standard monster. These monsters have 50% of baseline HP and 60% baseline damage.
A monster that could outnumber the fighter 4:1 and pose an equal threat. These guys have 20% of the baseline hit points and 40% of the damage.
The variation comes from the drop off in monsters as a fight continues. Simply spreading the damage out makes things too easy, as the PC is likely to drop a foe in the first round of combat.
What about monsters meant to take one more than one character? We'll get to those next time, but their stats require a different approach to account for the likelihood that the party will try to alpha strike them.
So what does that look like? Here's a bugbear's stat block:
This is a 3rd level creature meant to fight the party in even numbers. By design, I tried to make this creature simple to run. A few notes:
I'm using static initiative. A monster's place in the initiative order has a huge effect on its combat effectiveness. It's a knob that we need to be able to turn in design.
I settled on giving the bugbear two simple abilities, a bonus to Stealth and advantage against bloodied targets. I want them to feel like bullies who are good at setting up ambushes. Since these are creatures meant to show up in numbers, I want them to be streamlined.
Note the lack of saves and skills, other than the big bonus for Stealth. I created baseline stat bonuses for the creature than added proficiency to them to get the final scores. It makes things easier to reference at the table and ensures that monster bonuses scale up with level.
Damage is a die expression, created without worrying about finding the right weapon and adding up abilities. I calculated attacks based on the monster's flavor and intended play experience rather than treating them like characters and trying to line up the right stat bonus, proficiency, weapon dice, armor, and so on.
I also liked going with all dice and no modifier to add more chaos and uncertainty to a fight. Rolling a fistful of d10s when a character gets tagged by several bugbears is scary and threatening in a good way. Over the long haul, variance favors the DM. Let's lean into that.
I'll get into encounter building later this week, but this creature is worth 400 XP. That's exactly enough XP for a hard encounter with one character versus one monster. There's also a simple system for encounter building if you stick to using monsters of the character's level.
You might also notice that I redefined darkvision. As you'll see in the my stealth rules, I allow creatures to hide in dim light. This approach allows monsters to operate in the dark without also shutting down stealth.
Later this week I'll write about monsters meant to fight more than one PC and the special rules they need to preserve their action economy and handle several foes at once.
Comments
Douglas Terbush
Help to describe and give the monster character (present in Red Hand of Doom for reference): -Short physical description -A simple battle cry or insult New field to add: -Affiliation Think of all those Drow affiliated monsters and all those new players who have no clue who they associate with. It took me 45 minutes to find the Bebilith that is associated with the drow. Having that tag AND being able to list out associated monsters greatly speed up encounter building for the DM. It's something WotC screwed up on. Go to D&D Beyond and see if there's a drow tag, nope not there. You should be able to go to a filter, go to the Affiliation tag, set it to Drow, and then get a list of monsters commonly associated with Drow.
Michael Sixel
You know, I like this question. I'd prefer it be really clear, like Stealthy: d20 +8 for all stealth checks.
Andy Shockney
This is excellent! I’d encourage you to look beyond personality traits to consider motivations. You mention that these Bugbears are bullies, but that is lost on this write up. It should be clear in supporting narrative and speak to tactics that can be used to influence the creatures outside of combat.
Tim Kreider
Love the simplicity. For a Stealth check is the total modifier +4 + 4 + proficiency, or just +8? Either way I think I’d prefer seeing the final mod written rather than needing to calculate each time I use it.
