Recently, my kids and I finished our long-term campaign in Eberron. The big climax to the campaign was a battle with the mad scientist who had transferred his consciousness to a great glass orb, and whose stat block was similar to a Beholder. But then, I started having misgivings, and was worried about the kids getting killed because the challenge rating was just on the edge of what the kids’ party could hope to defeat.
I downgraded the stats and removed some bodyguards, but when the big battle came, it was clear that I had watered down the battle too much. The kids happened to roll well on saving throws, but also they quickly overwhelmed the boss and killed it.
Later my son (9) told me, “Daddy that was kind of easy.”
I remember years ago when my firstborn also played D&D with me, and we had the final boss battle then with a powerful, custom fiend, but the same problem occurred. The boss couldn’t do as much damage as the rest of the party.
So, it’s not enough to use encounter charts such as those in the Dungeon Master’s Guide and Xanathar’s Guide to Everything, to make a battle compelling you have to either go big or go wide.
Go Big, as the name implies to push the envelope of what the party can safely tolerate in combat. This means using the encounter charts above, but then increasing the difficult just a bit more. The goal as a DM is to foster a compelling challenge, not to kill the party. That does require the need to instill a genuine sense of threat, but also if the party is large, don’t let the boss sit around and just take a bunch of hits.
- Consider using features like Legendary Actions, and Legendary Resistances.
- Consider increasing hit points, or the armor class beyond regular statistics.
- Consider adding some flare to some of attacks or spells used. These may simply be for dramatic flavor, but they do make a powerful impression on players.
- Add some environmental difficulty, such as a force field to block the path, or environmental hazards such as lava.
Go Wide, on the other hand leaves the main boss largely unchanged, but adds more monsters to combat: soldiers, body guards, lair attacks and so on. This solves the problem of the heroic party having plenty of attacks to concentrate against one target, but on the other hand, it makes combat more difficult to manage. However, with so many potential threats, this can foster a sense of peril like above, through different means.
One other tip that might come in handy, which I regret not doing, is to rehearse the combat on your own time. Assuming you have the player character sheets (or a reasonable facsimile), you can simulate the battle in your own time and get a good sense of whether the battle is too easy, too long, or too deadly.
These are just some tips learned the hard way, but hopefully it will help. Good luck!
2 thoughts on “Designing Big Boss Encounters in D&D”
Good stuff. I don’t often see the suggestion to rehearse the combat ahead of time but I think that’s a really good idea. Even if you don’t literally break out the dice, I think there’s a lot to be gained by thinking out how the first few rounds will play out. You’ll usually be quite familiar with the abilities of the PCs. Knowing that they’re finally facing the BBEG, you can be sure that certain heavy hitting building are going to be used early on. That will usually give you a good feel for how you villain will stack up.
Another thing I’d add is that adding some type of timer to the encounter can really ramp up the tension, without necessarily making the PCs more likely to die. With this I’m picturing something like a prisoner in some type of peril on the other side of the room or a precious McGuffin sinking into quicksand. Sure, the PCs might focus their fire on the villain but are they willing to lose as a result?
LikeLiked by 1 person
Hi Talaraska, I wish I had done the rehearsal more often in the past.
The timer idea is a really great one! In the aforementioned encounter, there was a Metroid-style self-destruct sequence after the battle, but if I had somehow moved it up to the boss battle, that might have gone better.
LikeLiked by 1 person