Chapter 05
Common pitfalls
Most bad homebrew monsters miss the mark in one way or another. This chapter goes over the most common mistakes, explains why they happen, and gives the fix, including the single best piece of reframing in the community literature: the Lazy GM's "lightning rod".
The HP sponge
The mistake: the fight was over too fast last time, so this creature gets triple hit points and nothing else. The party wins with certainty, just slowly. Nobody makes a decision after round two; they're mining hit points.
The fix: if you want a longer fight, you need to add threat, not just durability. Raising the danger level of a round (a recharge spike, a second phase at half HP, reinforcements) makes the extra length carry tension. And keep HP inside the benchmark range for the CR; the top of the range is generous. If the party is simply out-leveling your monsters, raise the CR rather than inflating one stat.
The save-or-suck stun-lock
The mistake: a creature that paralyzes, stuns, or banishes on a failed save, every round, at a high DC. One player fails one roll and stops playing the game, sometimes for the whole fight.
The fix: control effects are the strongest currency in the game; spend them gently. Prefer conditions that degrade (restrained, poisoned, frightened) over ones that delete turns; give repeat saves at end of turn; and if the creature can lock someone down, have it switch targets rather than re-stunning the same character round after round. As the Lazy GM's guidance puts it, control used too often or too effectively stops feeling like challenge and starts feeling like punishment.
The overloaded statblock
The mistake: eight traits, six actions, three reactions. The creature is so interesting that the person running it forgets half its abilities and the fight takes twenty minutes a round.
The fix: a fight lasts three to five rounds; about three actions is enough. Cut anything that doesn't express the concept or create a decision. If you can't bear to cut, split: two lean monsters that each do one thing well beat one bloated one, and play better, too.
The unsupported solo boss
The mistake: a single impressive creature versus five characters. It gets one turn to their five, fails one save against banishment or a stunning strike, and the climactic battle ends as an anecdote.
The fix: action economy is covered in Fundamentals and the legendary toolkit in Offense & defense: legendary actions and resistances, lair actions, and above all allies. Even two or three minions transform a solo fight. If the boss must truly stand alone, budget it like a legendary creature and accept that its CR should sit well above the party's level.
Shutting players down (and the lightning-rod alternative)
The mistake: the subtler cousin of all of the above. The party's wizard trivializes encounters with fireball, the monk stuns everything, so the new monster is immune to fire, immune to stun, and spread-proof. It's a rational response, and it quietly deletes the players' favorite buttons.
The fix: invert it. The Lazy GM's Resource Document calls this building lightning rods: monsters designed to absorb the party's best tricks, spectacularly.
- Facing area-effect casters? Send hordes of low-CR creatures that bunch up and beg to be fireballed. Being tactically "wrong" is being cinematically right.
- Facing banishment, polymorph, or a stunning-strike monk? Add expendable lieutenants, like a bruiser or defender with bad mental saves standing next to the boss. The spell lands, it matters, and the boss still stands (that's what legendary resistance is protecting).
- Facing burst damage snipers? Add fragile damage-dealers (low AC, low HP, scary output) that reward being focused down with one great shot.
Then telegraph them: make it readable that the flameskulls are the thing to fireball and the fire giant bodyguard is the thing to banish. Watch what your players love doing, ask them at level-up what new tricks they've picked, and build monsters for those tricks, not against them. Immunities make players smaller; lightning rods make them heroes, and the encounter is harder to survive anyway because they spent their best resources exactly where you wanted.
This guide draws on openly licensed community works. See Sources and attribution.