Difficulty, Danger, and Death. Have you ever played Dungeons & Dragons, like D&D 5e, and noticed how Monster Encounter Difficulty becomes increasingly unpredictable at high-level? Ever wondered why? Conversely, ever had a low-level Total Party Kill (TPK)? The traits that really make encounters in D&D 5e difficult or dangerous, don’t always boil down to Challenge Rating. What makes for Deadly Encounters? Using this guide, you as a Dungeon Master can better balance encounters, or just adjust difficulty up or down dramatically. And you as a D&D Player can approach the game more tactically. I contrast Apparent Challenge Rating and Actual Challenge Rating, and discuss Challenge Rating Instability. Beyond just Action Economy Parity, inspired by projects like The Monsters Know, I examine how Supplies, Schema, Status, Space, Structure, and Strategy all shape Game Balance.
PART I: SUPPLIES
#1 – Inventory & Off-Label Uses
PART II: SCHEMA
#2 – Ignorance & Overconfidence | #3 – Scrying
PART III: STATUS
#4 – Exhaustion or Energy | #5 – Time Constraints
PART IV: SPACE
#6 – Fog of War | #7 – Terrain Hazards, Obstacles, and Liabilities | #8 – Terrain Boons, Opportunities, and Assets | #9 – Collateral Damage
PART V: STRUCTURE
#10 – Squads | #11 – Hordes, Waves, and Swarms | #12 – Incompatibility In Depth | #13 – Charismatic Leadership & Logistics | #14 – Battle Formation Exposure
PART VI: STRATEGY
#15 – Lethal Pacing, Gambits, and Targeting | #16 – Combat Traps | #17 – Superior Action Economy | #18 – Superior Mobility & Evasiveness | #19 – Unique Monsters | #20 – Hyper-Strategic Monsters
Regarding Encounter Balance, the Dungeons & Dragons 5th Edition Monster Manual states, “An appropriately equipped and well-rested Party of four adventurers should be able to defeat a monster, that has a challenge rating equal to its level, without suffering any deaths.”
5e has a baseline notion of pacing called the Adventuring Day as standard gameplay practice, with one Long Rest and several Short Rests covering roughly 6-8 encounters. Often the majority consist of Random Encounters, developing Party attrition through travel and combat exhaustion. D&D gameplay design assumes that baseline: primarily attrition, but punctuated by periodic Deadly battles against foes like champions and bosses, sometimes with lairs.
5e also models through Encounter Difficulties:
1. Easy: 25% XP – Doesn’t tax PC resources, no serious peril. Victory nearly guaranteed, at the cost of a few hit points.
2. Medium: 50% XP – One or two dicey moments, but PCs should prove victorious without casualties, though it may require some healing resources.
3. Hard: 75% XP – Could go bad. Weaker PCs might go down, slim chance of one or more characters dying.
4. Deadly: 100% XP – Could get lethal for one or more PCs. Survival requires planning and wit, and the Party risks defeat.
Many DMs, such as Sly Flourish, or I, use mainly “Hard” encounters, as it prevents monotony by allowing fewer random, insignificant encounters, while maintaining appropriate pace, risk, and reward.
Based on Sly Flourish’s reading of the DMG,
“For monsters with a CR of 1/4 the character’s level, use 2 monsters per character.
For monsters with a CR of 1/3 the character’s level, use 1 monster per character.
For monsters with a CR of 3/4 the character’s level, use 1 monster per 2 characters.
For monsters with a CR equal to or above the character’s level, use 1 monster per 4 characters.”
Party Size to Mob Size ratios set good baselines for things like numbers of monsters because Action Economy Parity strongly affects difficulty. Using Kobold Fight Club and the Goblinist Tools for example, with one Wight (CR 3), four Level 3 PCs will have a “Medium” Encounter. Probably no one should really risk death at all there. Add one Beggar Ghoul (CR 1/2), that bumps it to “Hard”: someone could possibly die. Add two Beggar Ghouls, and that bumps it up to Deadly: someone could definitely die.
But what about one Wight (CR 3) and one Brown Bear (CR1)? It counts as “Hard”. Notice that 3+1=4, and someone could possibly die, but 3+1/2+1/2=4, and someone could definitely die. It comes down to Action Economy Parity: trying to balance the level of actions the PCs and the monsters can take. With four Level 3 PCs, four Apes (CR 1/2 each) count as “Medium”, five as “Hard”, and seven as “Deadly”, again because of the Action Economy: at least 1 PC will likely get highly outnumbered at seven monsters. Because of this, 5e has added Bosses with “Lair Actions” that can overcome the Action Economy, making for more dramatic and challenging encounters, with fewer monsters.
I use the Encounter Size rules primarily for preventing Deadly encounters when I don’t want them, and for giving an appropriate level of risk and reward to the PCs. I like to know just how big a mob it would take to knock down my PCs, applying that judiciously. I can usually tell when the PCs have it too easy — everyone sees that clearly — but difficulty gets increasingly hard to measure as it grows. I think if you use XP calculations you will find the difficulty variables I mention here useful for balancing encounters too, and you can even factor it into XP calculations.
Personally, I prefer Milestone Leveling to avoid a lot of the extra math. But that doesn’t help you as a DM predict when PCs will die. It offers a good baseline average, but what variables make encounters difficult, dangerous, or deadly encompasses many other things. Such as…
PART I: SUPPLIES
#1 – Inventory & Off-Label Uses
Sometimes bosses get ahold of powerful artifacts. But do the Party’s sentient enemies have access to basic things like Healing Potions? How about utility items like scrolls and wands? Rules-as-written (RAW) in 5e, they basically never do. The stat block for Zariel, the CR26 fallen angel with 26 INT and 27 WIS, who can teleport at will, nevertheless has precisely zero scrolls or wands on her. Why? Because 5e keeps everything simple and streamlined. If you DO give your monsters an inventory, the ones with lairs, high CR, or high INT, may also sometimes find off-label uses for their items. Like, consider all the traps a boss could Rube Goldberg up with some caltrops, ball bearings, oil, nets, and Alchemist’s Fire. If we play monsters like PCs, this quickly gets out of hand.
PART II: SCHEMA
#2 – Ignorance & Overconfidence
If the Party rolled really poorly on certain Skill Checks to gather intel on their opponent, or simply didn’t spend enough time on that, or if you just have a baddie with a strong crypto game, things get difficult. And it gets even worse with the possibility of misinformation, with the Party in a position of overconfidence. This doesn’t have to come accidentally either, a cerebral opponent will intentionally lay decoys and deceit. Likewise, if you as a DM decide that the monsters have poor encounter schema, this can give the Party a huge edge, particularly with false vulnerabilities leading to wasted moves.
#3 – Scrying
Just how much awareness one group has of the other via things like Scrying spells factors into challenge tremendously. At a certain point, overuse of Clairvoyance, Arcane Eye, or Scrying removes a lot of the uncertainty from a character sheet, as well as the terrain Fog of War. More on that later. Past a certain point, this level of slowly accumulating intel can trivialize encounters for one side.
PART III: STATUS
#4 – Exhaustion or Energy
D&D at its core comes down to a game of attrition. When an exhausted Party faces a fully-rested enemy, the situation can become dire. Similarly, worn-down foes can prove easy pickings for the Party; RAW, the monsters always have their full daily abilities, but the Party fights at varying strengths that determine actual difficulty, rather than apparent difficulty.
#5 – Time Constraints
Time constraints have an implicit cost associated with them: “solve this soon or else something real bad happens”. Ticking timebombs. Slowly closing gates. Kidnapped PC or NPC allies, soon-to-be-executed elsewhere. Rescue, escape, and chase modes of play change difficulty by impacting player agency directly, changing what it means to “win” and “lose” an encounter.
PART IV: SPACE
#6 – Fog of War
Fog of War refers to the general uncertainty of combatants interacting with the unknown landscape and opposition, which feeds mystery and suspicion. Primarily this deals with the number, formation, position, and capabilities of one’s opposition. Do they have hidden monsters? Hidden supplies? What options and risks does the terrain offer? It can also entail the possibility of shifting battlefields, or even a backlash from outsiders intervening. All this uncertainty affects player choice and PC use of resources, and thus difficulty.
#7 – Terrain Hazards, Obstacles, and Liabilities
Natural terrain hazards like the ones in the 5e Dungeon Master’s Guide or the ones I imagined in my article on Gritty Gaming factor into movement and formation. (Note: I also recommend the Tribality Collection of Traps & Hazards.) Obstacles slow you down, or impose speed at a cost. And traps hold only downsides, often severe ones. Have you ever played the Divinity: Original Sin game series? Anyone who has can tell you about liabilities like barrels or pools or clouds full of elemental energy strewn about, and the complex terrain interactions that arise from this. In D&D, even simple liabilities like ledges, water, and darkness can prove important as falling, suffocating, or blindness can circumvent encounters entirely. In a similar way, fragile ceilings crashing down or moonshine barrels going boom make a big impact. If one side has better familiarity with a terrain’s various dangers, it can definitely give an upper-hand, allowing them to avoid harm. Or, to lead others into it!
#8 – Terrain Boons, Opportunities, and Assets
To start, I should mention simple terrain boons, like a King-of-the-Hill style fight over a small magical healing pool. Regarding terrain opportunities, Extra Credits’ video on Overwatch and Asymmetric Character Design mentions the many terrain features that game requires to give each role substance and viability. For example, barriers to leap over. Climbable sniping tower overlooks with functional sightlines. Contiguous flanking routes like corridors, and discontiguous flanking routes like archipelagos of spires to jump between. Scaleable walls and battlements, or, concealment and cover generally. Also, various amplifiers, such as a watchtower with an oil reservoir for flaming arrows. Even switches and hidden passages that allow for dynamic terrain shifts. Much of that applies to D&D too, with complex terrain seemingly undervalued overall. We can also consider assets, like a evil overlord lair’s armory, laboratory, storehouse, or stables. Again, can one side put all these to better use? To a significant extent, terrain shapes action, and action determines risk, in D&D too, much more than monster CR may estimate. Just how meaningfully a monster can use a Lair can vary wildly.
#9 – Collateral Damage
Remember the hostage situations I mentioned in Time Constraints? Now add in targetable infrastructure. Add in bystanders. Add in fragile treasure. Yeah, enemies will force players to make sacrifices. Again, these alter what “winning” and “losing” mean beyond simply PC death.
PART V: STRUCTURE
#10 – Squads
Let’s talk about teamwork and tactics. A proper squad has synergistic roles. I have a separate article planned on Combat Role Archetypes, but for now, imagine an enemy group with a variety of strengths and weaknesses covering each of the six Ability Scores. Let’s look at a sample team of just four CR 3 monsters: a Merrenoloth (high DEX & INT, low STR), an Ogre Chain Brute (STR & CON, all else low), a Minotaur (high STR, CON, WIS, low INT & CHA), and a Yuan-Ti Malison (high STR, CHA, nothing low). That covers all the highs, it could make a decent squad. At an even higher level, each squad member could have totally distinct immunities, or overcome different resistance types. RAW, the DM only has to factor in Challenge Ratings when calculating Encounter Difficulty. But this implies little consideration for the emergent threat of cooperative strategy among even weak (but diverse) actors working in concert. Tucker’s Kobolds, anyone? Squads can hit above their weight class, AKA their apparent combined CR. Particularly monsters built for it, like the Star Spawn Seer and Star Spawn Hulk synergy. Finally, many of us have played in a Party with too many redundancies and gaps. It can make things really rough.
#11 – Hordes, Waves, and Swarms
Another point on composition and formation. Few things achieve attrition as easily as hordes, waves, and swarms. Whereas hordes achieve this by scale, waves of enemies achieve this by time, simply fighting in series to sap finite resources like Spell Slots. Too big or too sequential to succumb to a single Fireball spell. Swarms, in contrast, add on immunities and mobility options. In terms of challenge, the first and last segments fought among an enemy using hordes and waves offer very different marginal threat, and swarms further multiply this.
#12 – Incompatibility In Depth
Ok, so squads use variety, and hordes use homogeneity. Let’s look at a special sub-case of the latter. What happens when the massive mob proves immune to the Party’s main damage types? What happens when they all stack a damage type on PC has vulnerability to? Deep trouble. Much worse than normal CR estimations would guess. Imagine a horde of the CR 1/4 Giant Frogs with their on-hit restraint? That could cause a low-level TPK.
#13 – Charismatic Leadership & Logistics
What happens when you have a strong “Face” opponent, one who can effectively manage both leadership and logistics? Specifically, when a Boss can help the Minions and Mooks overcome their limitations? When leaders can effectively project sound strategy to masses of henchmen, like Hobgoblins ordering Bugbears and Goblins, hordes become waves of squads. Challenge increases. Worse, charismatic leaders might even get a Party’s NPC allies to mutiny, decisively undermining CR stability.
#14 – Battle Formation Exposure
Sometimes you can put the tank in the doorway and laugh at your enemies’ frowning faces. Othertimes, the squishy Wizard gets caught flat-footed in front of the ambush and takes all the heavy blows. Challenge Rating indicates averages, but actual susceptibility once the dice roll demonstrates authentic threat. To prove the point, consult any Boss fight that players trivialized by bypassing the champion guards. Formation and flanking can count for a lot, just imagine if an Intellect Devourer, Rust Monster, or Mind Flayer get the drop on a Party.
PART VI: STRATEGY
#15 – Lethal Pacing, Gambits, and Targeting
Do the enemies ration out their finite daily abilities, or blow them all on this encounter with big Nova attacks? Do they have a willingness to use their best moves in this fight? Do they use lethal gambits, ones that target specific tactical roles and target downed opponents? On the one hand, you have enemies that just want to eat prey, or ones that attack randomly or by irrational bias. And on the other, you have foes that meaningfully select squishies or tanks, DPS or utility roles, party faces or problem-solvers. If the CR 5 Beholder Zombie can get a turn, throwing that 45 damage Disintegration Ray gets scary. Lethal intentions build difficult encounters.
#16 – Combat Traps
Returning to the earlier topics of Fog of War, and terrain liabilities, what happens when the monsters are simply stalling, a diversion meant to redirect the PCs into the much more deadly traps? Weak-but-visible enemies combined with strong-but-hidden traps lull players into a false sense of security and lure them into danger. Most players do not expect that during combat, traps will offer the biggest jeopardy. Strategies like these mess with player meta-understandings of the game around perceived CR.
#17 – Superior Action Economy
Of all the aspects, D&D 5e balances the most around this topic. That’s why Legendary Actions and Lair Challenge Ratings work the way they do. Enemies that have extra swiftness, that can summon reinforcements, that can inflict time-sinks against PCs or otherwise employ crowd control, overcome the common 4-on-1 nature of many encounters. And yet, with enough asymmetry of any of these elements, Challenge Rating can quickly deteriorate, particularly with summoning from Druids, Demons, or Devils. We can even find clearly broken RAW content like five CR 1/4 Aarokocra getting to summon a CR 5 Air Elemental. More worrisome than conjuration shenanigans overall in my opinion, though: dungeon alert. D&D gives too few recommendations on enemies sounding alarms, and this can cause Action Economy Parity to vanish instantly when all of a dungeons’ mobs rally together.
#18 – Superior Mobility & Evasiveness
Have you heard of a player trying to stack Tabaxi or Quickling speed bonuses with Monk, Barbarian, Fighter, and Wizard ones, throwing in the Mobile feat, and Longstrider on top of that? Even that can’t hold a candle to flight, phasing (e.g. Blink), or teleportation, in most cases. Mobility and evasiveness can prove very swingy. And monsters that can kite while firing ranged attacks like the Manticore (CR 3) can easily out-maneuver a Party. Shadows’, Wraiths’, and Specters’ Incorporeal Movement does so often. In fact, the lowly CR 1/2 Shadow has 18 resistances and immunities, incorporeality, stealth on top of that, and deals STR damage, which a level-appropriate Party will have issues with. They have probably the wildest swing in difficulty, giving even a Party facing an Invisible Stalker without a sack of flour a run for their money. (Note to self: buy more flour…)
#19 – Unique Monsters
Sometimes a DM has little to compare a monster against. Monsters with very unique perspectives, abilities, or strategies. Some monsters simply prove troublesome for a DM to roleplay sufficiently. Consider, the motivations of entities like a Sibriex. Or the memories of an Aboleth, or Elder Brain, or Oblex. Or the mindset of an Allip, Star Spawn, or Stone Giant Dreamwalker. Incomprehensibility can affect encounter difficulty in hard to predict ways. As an example, if the DM stops making Beholders act randomly and instead gives them their due Intelligence, using frameworks like the ones I plan on writing for roleplaying Intelligence and Wisdom better, difficulty would increase significantly. If the DM approaches it lazily, not so much. Any monster that could use spells like Time Stop, would plan their use carefully.
#20 – Hyper-Strategic Monsters
My later posts on Combat Role Archetypes and Intelligence will cover this more, but, basically, hyper-strategic monsters have a tremendous advantage over others. A hyper-strategic enemy accounts for strong openers and finishers, counters and reversals, mitigation and cutting losses. They may resort to utilitarian self-sacrifice, or they might retreat, rally, and re-engage. They may utilize recursive learning or training, simple shortcuts like manuals or mnemonics or adages, or else possess extraordinary magnitude of mentality or memory, senses or scheming, patience or persuasiveness. The strategies of entities like high-CR Abominations, Celestials, Dragons, and Fiends prove well-justified in avoiding a stable CR. Reiterating some earlier examples, enemies with higher INT or WIS may capture PCs to interrogate or ransom instead of just killing them, or take a hostage when the situation deteriorates. Do you really think Zariel, who has seen the might of both Heaven and Hell, couldn’t devise a plot greater than a simple stat block?
I hope I gave you some food for thought on difficulty, challenge, and balance in D&D 5e. If you liked this article, please share it, and let me know in the comments what you liked or what I missed in this exploration on Encounter Difficulty. I publish new posts each Tuesday, and in the meantime, I post original D&D memes daily over on my site’s Facebook Page and Twitter. Also, to stay updated on all my posts, check the bottom sign-up widget for email notifications. Thanks as always to my Patron on Patreon, who helps keep this project running: Rudy. Thanks for the support, Rudy! Feel free to join him!