Running the Blood War in D&D 2014

Introduction

Most tables treat Devils and Demons as the same monster with different horns: both are fiends, both are evil, both show up with a pitchfork and a stat block. Mordenkainen's Tome of Foes spends its first chapter arguing the opposite - Devils and Demons are opposing armies fighting an endless war precisely because they represent incompatible visions of evil, and a third faction, the Yugoloths, profits from both sides never actually winning. This guide from the D&D 2014 Dungeon Master Guides hub distills that material into one behavioral signature per faction and a set of table-ready tactics for each, in the same format as the Monster Guide's section on running Volo's Guide to Monsters' iconic monster factions.

The war itself has a simple shape worth knowing before running any of it: devils hold Avernus, the uppermost and most contested layer of the Nine Hells, as their defensive front line against a demonic tide that never stops trying to break through, and the fighting has ground on so long and so evenly that neither side has managed a lasting victory. That stalemate is the whole point of the conflict rather than a bug in it - Asmodeus openly justifies the entire infernal soul trade as fuel for an army holding back an even worse chaos, and the endless grind of Avernus is why both sides are constantly hungry for resources, recruits, and mercenaries a party might have to offer.

This guide is not a stat-block reference. The D&D 2014 Monster Guide already covers reading a stat block, and the D&D 2014 DM Quick Reference already covers encounter-difficulty math. What follows is the missing piece: what actually makes a Devil, a Demon, and a Yugoloth feel like three different kinds of threat once a party is in the room with one.

Devils: Corruption Through Contracts

Every Devil in the Nine Hells answers, eventually, to Asmodeus - not because he commands universal loyalty, but because the Nine Hells runs on an immense body of law that even its ruler can't openly break. The layers below him are governed by archdevils who each hold real, separate power: Zariel drives the war effort from Avernus with reckless battlefield aggression, Dispater trades in secrets and paranoia from Dis, Mammon measures every soul in gold from Minauros, Fierna and Belial rule Phlegethos as manipulative rival-partners, Levistus offers escape to the desperate from inside the iceberg that imprisons him, Glasya runs Malbolge's prisons and an organized-crime syndicate in the same breath, Baalzebul recruits the desperate from Maladomini under a curse that forbids him from lying to another devil, and Mephistopheles draws arcane talent to Cania. None of them fully trusts the others, and all of them are constantly maneuvering for advantage - but every one of them still obeys the same rules, because the rules are the only thing keeping the whole hierarchy from tearing itself apart.

Asmodeus's own defining story shows exactly how far this legalism goes: when the celestial host of Mount Celestia formally accused him of crimes during the Blood War, he didn't flee or fight - he petitioned for a trial, drafted the contract governing its terms himself, and argued his case for weeks before an impartial judge, framing every soul he'd claimed as a lawful transaction and every mortal victim as complicit in their own downfall. The presiding judge never ruled him innocent, but never punished him either, which is itself the point: Asmodeus doesn't need to win an argument, he only needs the process to end in a stalemate that leaves the law - and his place at the top of it - intact. That same instinct explains how devils recruit talent from outside the Hells entirely. Zariel, now the general driving the war effort from Avernus, was once a celestial who grew so obsessed with the fighting she was ordered only to observe that she abandoned her post to join it; Asmodeus didn't conquer her, he simply gave a desperate soul with a very specific need exactly the outlet the rest of the cosmos had denied her, then immediately made her useful.

That structure is the actual engine behind every Devil encounter: a Devil's real weapon isn't its claws or its fire, it's the contract a mortal already signed, and the fight is what happens after the terms stop being convenient for one side. A Devil rarely opens with violence because violence risks breaking a bargain, and breaking a bargain is the one thing every Devil, from a lowly lemure up to Asmodeus himself, is built to avoid. Instead, Devils recruit through cults tailored to what a specific archdevil offers: Dispater's agents chase secrets and court renegade spellcasters hunting forbidden knowledge; Fierna's cultists want to be loved, feared, or respected and get a gift for bending others' emotions in return; Mammon's followers get irresistible wealth and a hunger to spread greed like a disease; Glasya's petitioners want power gained through a legal loophole rather than an outright crime; Levistus offers a single guaranteed escape to anyone who is desperate enough to take it. Each cult reads differently at the table because each one is selling something different, and the price is always a soul, paid on terms the mortal agreed to and often didn't fully understand.

The archdevils' domains reinforce the same bureaucratic character outside of combat entirely. Dispater runs the Nine Hells' arms industry from a paranoid iron palace, sending orders branded onto the backs of imps that self-destruct if opened by the wrong hands, because he trusts almost nothing he can't personally verify. Mammon runs the soul trade's actual accounting: every claimed soul passes through his capital to be logged, valued, and distributed according to whatever contracts and laws currently apply, with any surplus quietly pocketed for his own treasury. That bureaucratic machinery - ledgers, couriers, audits, quotas - is a useful default texture for a Devil-controlled location a party investigates: not a smoking ruin, but a functioning institution with paperwork a clever party can exploit as readily as any devil can.

That legalism extends into how Devils fight once combat is unavoidable. Devils are organized into disciplined military legions - Baalzebul's shield legions and dreg legions appear in this material as concrete examples of how the infernal war machine is structured, with the least valuable souls funneled into legions whose entire purpose is to die slowly enough to blunt a demonic advance. A Devil unit fights with that hierarchy in mind: it holds a line, it follows a chain of command, and it treats retreat or insubordination as a legal offense rather than a tactical option. Even a Devil's ambition operates inside the rules - Baalzebul's attempted coup against Asmodeus wasn't punished as treason so much as converted into a permanent legal penalty (he's forced into a hideous slug-like shape for a year every time he lies to another devil, a debt so old it took him millennia to work off), and Glasya's counterfeit-currency scheme against her own father succeeded specifically because she found a real gap in infernal law rather than breaking it outright. Devils scheme constantly against each other, but always through the law, never around it.

That internal rivalry gives a DM ready-made leverage even inside a single Devil encounter. Levistus, frozen inside an iceberg as punishment, still runs a functioning cult by proxy, while the archdevil he displaced, Geryon, wages a running low-grade war against him from outside using mercenaries and skirmishers because he can't touch Levistus's prison directly. Fierna and Belial rule the same layer as declared rivals who nonetheless close ranks the instant an outside threat becomes real. A Devil-themed encounter doesn't need to be a single loyal unit - it can be two factions of the same hierarchy testing each other while a party gets caught in the middle, which reads as far more political than a monolithic "the devils attack."

Running a Devil encounter well means resisting the urge to open with an attack roll. Introduce the archdevil's cult through the deal it's already offering, or the deal an NPC the party trusts has already made - a noble who wants a rival's fortune without murdering for it, a spellcaster desperate for a secret only Dispater's agents can supply, a criminal who wants Levistus's one guaranteed escape badly enough not to read the fine print. When the fight does start, run it as disciplined, hierarchical, and utterly literal: a Devil that loses an advantage doesn't panic, it falls back on the contract's actual wording, because the contract - not the sword - is what it actually trusts to win.

Table-ready signature: introduce a Devil through the terms of a contract someone has already signed, and when combat starts, run it with legion-style discipline that never breaks a literal term of a deal even while exploiting every loophole in it.

Demons: Chaos Without Limit

Where a Devil's power comes from law, a Demon's comes from the absence of it. Every demon believes, without exception, that it is the rightful center of the cosmos and that the universe will eventually bend to match its personal idea of perfection - a belief that scales from a mindless manes clawing at whatever's nearest up to a demon lord like Demogorgon, Orcus, or Graz'zt reshaping an entire layer of the Abyss around its own obsession. Demons have no natural affinity for their own kind and cooperate only when a more powerful demon forces them to, which is why demonic forces are dangerous in raw numbers and aggression rather than coordination. A vrock and a dretch fighting side by side aren't allies pursuing a shared plan - they're two individually self-centered monsters who happen to be pointed in the same direction for as long as something bigger is making them.

This chaos is also why demons resist the kind of tidy categorization Devils invite. Sages sort demons into broad types for convenience, but the source material is explicit that individual demons routinely defy those categories - a vrock might crawl out of the Abyss with an extra eye or vestigial wings it has no business having, and a chasme might manifest an ability nobody expects. That's a standing invitation for a DM to reskin freely: a demon's stat block is a starting point, not a fixed identity, in a way a Devil's rigid hierarchy never quite allows. Even the demon lords aren't immune to crossing the line that's supposed to separate the two factions entirely - a persistent rumor holds that Graz'zt, now one of the Abyss's most powerful demon lords, was once a devil in the Nine Hells before he abandoned Asmodeus's hierarchy for the Abyss's chaotic freedom. Most who study the planes hope that defection was a singular event; if a DM wants a Blood War campaign that hits hard, having it happen again is exactly the kind of tipping point worth building a whole arc around.

That same self-centered chaos is what makes an unchecked demonic presence spread like an infection rather than stay put like a lair. The source material lays out concrete stages: a demon that rampages through an area for a few days leaves no lasting mark, but demons left undisturbed for weeks or months begin to twist the local environment - plants mutate, water turns foul, weather runs to violent extremes. Left longer, a portal to the Abyss tears open in the corrupted ground, and if that portal survives long enough for more demons to find and use it, the region's transformation accelerates past the point where killing the demons already present is enough to fix it. In the worst case, a demon lord itself steps through and takes command of everything nearby, and by the time that lord is finally driven out, nothing of the original place is left. That escalation ladder is a useful campaign structure on its own: a demonic incursion doesn't have to be solved in one fight, it can be a countdown a party either catches early or lets run long enough to change the stakes of dealing with it.

Demon cults reflect the same self-interest rather than genuine devotion. Followers of Baphomet see themselves as superior predators and everyone else as prey to be hunted through labyrinths built for exactly that purpose - cultists drug their targets, strip them of gear, and stalk them through a maze the cult built specifically to exhaust every scrap of hope before the kill. Demogorgon's cultists are compelled almost involuntarily - simply viewing a true copy of his symbol is enough to twist a mortal into a lurking agent of paranoia and violence, no persuasion required. Fraz-Urb'luu's followers are mostly people he's lied to rather than true believers, drawn in through whatever flattering illusion gets them to lower their guard, playing a saintly figure to a desperate paladin or a wise sage to a credulous wizard. Orcus offers grieving mortals the return of a lost loved one as undeath, and demands the same fate for anyone who accepts and then has second thoughts. Juiblex's cultists tend the oozes he leaves in his wake and feed captives to them as a preview of the fate awaiting everyone who doesn't submit. None of these demon lords actually wants devoted worshippers in the way a god does - they want instruments, and they discard them the moment those instruments stop being useful, which is a sharp contrast to how carefully a Devil protects a signed contract.

The Abyss itself resists any attempt to map it cleanly, which is worth carrying into how a DM frames a demonic threat rather than treating it as a fixed dungeon. Its layers shift, merge, and vanish as demon lords fight over them, and portals between them flicker unreliably rather than connecting the same two points twice. A demon lord that survives centuries can even hide a piece of its own essence in a demonic amulet to cheat death, which means the "final" defeat of a recurring demon threat can hinge on finding and destroying an amulet hidden somewhere entirely separate from the demon's own lair - a concrete, reusable plot hook for a demon lord a party thought they'd already beaten.

Run a Demon encounter as overwhelming, escalating, and personally selfish rather than tactically clever. A demonic war band doesn't hold a defensive line or retreat on a schedule the way a Devil legion does - it keeps attacking past the point a disciplined force would break off, because individual demons rarely calculate risk the way a lawful creature does. When multiple demon types share a fight, let them squabble or compete for the kill rather than execute a coordinated plan, unless a demon lord or a powerful demon is physically present to force cohesion - that presence should read as escalation, the moment a chaotic mob becomes a directed army.

Table-ready signature: run demons as an environment-corrupting infection that escalates the longer it's ignored, and as individually self-interested attackers that keep pressing an assault past the point a disciplined force would retreat.

Yugoloths: Opportunism for Hire

Devils and demons both believe they're fighting for something - law, or its absence. Yugoloths believe in neither. They are professional mercenaries who work Blood War contracts for whichever side is paying at the moment, undertake any task no matter how despicable, and famously look for every loophole a contract's wording leaves open. That neutrality isn't a lack of conviction so much as the entire point: a Yugoloth's loyalty belongs to the deal in front of it, never to devils, demons, or even other Yugoloths, and running them as a third flavor of demon throws away the one thing that makes them interesting.

That mercenary specialization shows up in how differently each Yugoloth type does its job. A canoloth takes contracts to guard something and does precisely what it was told - no more, no less - whether that means merely restraining an intruder or tearing them apart, entirely dependent on the exact wording of the job. A dhergoloth is muscle for hire, effective at simple destruction but so dim that any contract more complicated than "kill everything here" risks being botched or forgotten outright, which makes it a liability for a client who isn't careful with instructions. A hydroloth specializes in amphibious raids and stealing memories rather than souls, delivering stolen thoughts to whichever employer wants them. A merrenoloth captains vessels crossing the River Styx and other dangerous waters, and typically writes itself out of any obligation to fight, because its actual contract is to protect the ship, not the passengers. An oinoloth is the last-resort option when a siege has dragged on too long - it spreads plague and mass death for a price, and can be paid separately to lift the very sickness it caused, at a cost that leaves the survivors permanently worse off. And a yagnoloth is the one every other Yugoloth ultimately answers to on a given job: it negotiates and writes the contracts, commands lesser Yugoloths once terms are set, and always makes sure the fine print leaves it an escape hatch if the job turns bad - all while reporting up to still more powerful Yugoloths above it.

None of these creatures pretend to be anything other than what they are, which is its own useful contrast with the other two factions. A Devil disguises a bad bargain in flattering language, and a Demon doesn't bother disguising anything at all - but a Yugoloth will plainly state the scope of a job, because a client who understands exactly what they're paying for is a client who can't later claim the contract was broken. A canoloth's employer knows precisely whether they've hired a guard or an executioner; a merrenoloth's passengers know it has no contractual duty to fight for them. That bluntness reads as a kind of professional honesty a party can actually work with - a Yugoloth won't lie about what it will do, only about what it privately intends the moment the letter of the deal stops being convenient.

That layered specialization is what makes Yugoloths function as a genuine third faction rather than abyssal or infernal scenery. They're built to be hired, which means a party can hire them too - a yagnoloth negotiator is exactly as useful sitting across a table making a deal as it is standing across a battlefield leading enemy troops, and the difference is entirely about which side currently has the better offer. Even the yagnoloth answers to something above it: this material is explicit that yagnoloths take their own orders from arcanaloths and ultroloths, the Yugoloths' shrewder, more powerful leadership caste, which means a single hired mercenary company is rarely the top of the chain a party is actually negotiating with. A campaign that keeps returning to Yugoloth contractors can escalate the same way a demonic incursion does - the party starts dealing with disposable canoloths and dhergoloths, works its way up to a yagnoloth's contract terms, and eventually has to reckon with whichever arcanaloth or ultroloth actually commissioned the job.

Running a Yugoloth encounter well means treating the contract as the real content of the scene: what was this creature actually hired to do, what's the loophole it's planning to use if the situation turns against it, and what would it take for the party to become the better-paying client instead of the target.

Table-ready signature: a Yugoloth is defined by its current contract, not its allegiance to devils or demons - reveal the specific job it was hired for, the loophole it's protecting itself with, and the price that would flip its loyalty.

Putting It Together at the Table

None of these three signatures require running the full Blood War to be useful. A single archdevil's cult, a single demonic incursion, or a single Yugoloth mercenary company each work as a self-contained antagonist with no reference to the larger conflict at all. The three-way rivalry becomes valuable once a campaign has introduced more than one of these factions and the players start asking how they relate - at that point, the Blood War stops being setting flavor and becomes a lever the party can actually pull.

That's the payoff of running all three factions distinctly instead of as interchangeable "evil": their opposing signatures create leverage for a party smart enough to use it. A group that's uncovered a demonic incursion spreading through a region can feed word of it to a Devil's cult, which has every incentive to intervene before the encroaching chaos threatens infernal territory or disrupts the soul trade - and a Devil that arrives to fight a Demon on the party's behalf is a far more memorable ally than one the party had to fight directly. Conversely, a party cornered by a Devil's contract-driven cult might find that offering the right price to a Yugoloth mercenary company gets them muscle that owes the Devils nothing and answers only to whoever's currently paying. None of this requires elaborate setup - it only requires that the DM already know a Devil fights by contract and discipline, a Demon fights by unchecked aggression and corruption, and a Yugoloth fights for whoever's currently winning the bidding war, because those three answers are what let a party actually plan around the encounter instead of just swinging at it.

The material also gives a DM two useful examples of powerful beings who deliberately stay outside the fight rather than pick a side, and both make good models for a party-adjacent third party rather than a straightforward enemy. Tiamat resides in Avernus, the Blood War's front line, yet keeps her own court and dragons apart from the fighting entirely - neither side wants to provoke her, and she seems content to let both exhaust themselves without her help. Lolth, the Demon Queen of Spiders, rarely joins the war directly; instead her agents spy on both sides and quietly maneuver rival demon lords into leaving the Abyss altogether, expanding her own territory into whatever vacancy they leave behind. Neither figure needs to appear as a combat threat to matter at a table - a party that learns a dragon queen is deliberately staying neutral in a war raging around her, or that a demon lord's spies have been manipulating both sides from the shadows, gets a sense of scale and factional intrigue no single stat block could deliver on its own.

Building a one-off fiend cult from scratch for a party to uncover doesn't require inventing a whole theology from nothing, either. This material's own approach to a generic cult breaks it into a handful of independent questions worth asking about any fiend-aligned group a DM improvises: what does the cult actually want (political control, revenge, a specific artifact, or simply comfort for its leaders at the rank and file's expense), what resource keeps it funded and protected (a respectable business front, a captured noble house, corrupt officers of the law, a direct planar portal), how is it organized (a tight conspiracy, a false-front civic group, an entrenched local tradition nobody questions), and what internal weakness threatens it (infighting, a rival cult, a leader who's really a figurehead for someone else). Answering those four questions for any Devil, Demon, or Yugoloth cult gives it a working identity in minutes, without needing to reach for a named archdevil or demon lord at all.

Hot Take: Running Devils and Demons with the same tactics - both charge in, both hit hard, both die the same way - throws away the one worldbuilding idea Mordenkainen's Tome of Foes actually earns: that the Blood War is a fight between two incompatible philosophies of evil, not two reskins of the same monster. A table that's never seen a Devil offer a contract instead of an attack, or a Yugoloth calmly change sides mid-campaign for a better price, has never actually met these three factions - it's met three colors of "horned enemy" wearing their names.

For the mechanical side of placing and reading these creatures' stat blocks, see the D&D 2014 Monster Guide. For eight more iconic monster factions run the same way, see that guide's section covering Volo's Guide to Monsters' Beholders, Giants, Gnolls, Goblinoids, Hags, Kobolds, Orcs, and Yuan-ti.