Combat Design: Beyond the Gate of Win or Die

For a while now I’ve felt like my combat encounter design has grown stagnant. I’ve been running my very first long-term entirely homebrew campaign for about a year. Now that my players are entering the higher tiers, it’s a struggle to create something that poses a serious threat. If my players don’t immediately crush something in a matter of turns, combat slogs on, round after round, as I wait for them to whittle down the HP of the enemies I thought were going to kill them, or at the very least give them a run for their money. Combat design might be the crunchiest part of D&D, but mine had turned into one hell of a numbers game.

I like numbers a lot, but I wondered if that was really the solution to my problem. I started to think more critically about combat and the purpose it serves in my 5e games and in D&D as a whole.

Win or Die

I’ve read, played in, and run plenty of combat encounters. Across it all, I’ve noticed many of them follow this model:

1. The party is exploring an area.
2. The party comes across a group of enemies that prevent them from exploring.
3. The party rolls initiative.
4. The party brings the enemies’ hit points down to zero or otherwise removes them from combat.
5. The party continues on their way.

I’ve started referring to this type of encounter as “win or die.” The goal is typically to get the enemies’ hit points to drop to zero. Some players may opt to use Charisma to Persuade or Intimidate enemies into leaving combat, but the goal is the same: get rid of all hostile monsters before they get rid of you.

The problem with win-or-die encounters is that there is too small of a sweet spot between one that is mechanically too easy and one that is so difficult it becomes unfair. And because the result of a win-or-die encounter is player character death, things start to get hairy. Some groups shy away from death entirely, which guts the stakes from this type of encounter. If your options are “win or die” and you know you can’t die…

I’m not saying win-or-die encounters are bad or wrong. I use them all the time. But using them too much has started to make combat stale. The other night I figured out what I could do to freshen up combat without making any truly “crunchy” adjustments. It was all thanks to my players and the way they royally screwed up.

What Happened

My party of level 11 adventurers arrived in the White Kingdom in time for an annual feast. This year, the White Queen and her advisor The Holy Man planned to use the feast for their own destructive intentions. They wanted to sacrifice the guests to open up a connection between the Material Plane and the underworld, wherein they could search for the queen’s dead daughter and drag her out.

While the ritual succeeded, it did not go as planned. Instead of a princess, many shades and spirits emerged to harass the feastgoers still standing. The Holy Man had taken on several levels of exhaustion and was forced to his knees, unable to get up and correct his mistake. We rolled initiative, and thus began Phase One of the boss battle.

I told my players that the shades (represented by the purple tokens) were rushing to attack the incapacitated guests. Though my players worked hard, three managed to approach, eat, and absorb party guests, transforming into stronger creatures in the process.

In Phase Two, the three transformed shades became two shadow assassins and one death knight. With their newfound power, they set their sights on exacting their revenge upon the queen.

The queen, for her part, did not run away. She had staked the lives of her entire court on the chance to see her daughter again, and she clung hopelessly to the notion that one of the shades was the answer to her prayers.

All of a sudden, the goal of combat wasn’t “get these baddies before they get you.” It was “get these baddies before they can kill a woman who has less hit points and more national importance than you.”

These shades barely went after the players, except when they perceived those players to be actively stopping them from attacking the queen. Sure, there was some collateral damage due to the death knight’s casting of destructive wave. Yes, several of them went down, but they were not the knight’s target. The queen was.

And she died. While most of the players were busy attempting to hit the death knight from afar, one of the shadow assassins snuck into melee range and dealt a killing blow to the one thing holding the White Kingdom together.

For a moment I considered pulling this punch and allowing her to live. Then I remembered everything I wrote at the beginning of this post about stakes and meaningless encounters. So I kept it, and she died. And for reasons you can DM me about if you’re truly interested, she can be resurrected with nothing short of a wish spell. My players do not have access to that at level 11.

For the first time in months, something had gone so catastrophically not according to plan that I was once again reminded of the true purpose of combat in D&D.

Combat is there to advance the story.

That might seem like an obvious statement, but it’s actually a forest. And I’ve been stuck staring at the trees this whole time:

“Combat is there to help my players establish their character in ways that role play and exploration simply cannot.

“Combat is there to help establish that my world is dangerous, and talking isn’t always going to be an option.”

“Combat makes up a third if not more of the language by which this story is told, because we’re playing D&D, and D&D is the grandchild of a wargame

Combat can advance the story by fulfilling the above, but it doesn’t have to. It can do something else entirely. Combat can be a question of win-or-die, a binary question of pass/fail, or it can be an intersection that branches in more ways than one. Combat can be a choice with a bunch of tiny steps.

This encounter was exciting because it gave my players the chance to change the story in a way I wasn’t expecting. I see that plenty during role play and exploration, but once players roll initiative, things change. They focus exclusively on keeping their own hit points high while keeping the enemies’ hit points low.

That’s because win-or-die encounters often stand in the way of the story as opposed to advancing it. You’ve got to fight these goblins in order to search this dungeon room. You have to defend yourself from this random encounter in order to progress to the next location. You have to reduce a villain’s hit points to some level before surrender. If you fail, the gate closes, and you (or likely a freshly-rolled party) must try again.

From Gate to Intersection

So how can we make these gates into branching intersections? We can tweak the encounter to change the goal of the enemies, which will hopefully change the goals of the players. I’ve started asking myself these questions:

• What could the enemy want that isn’t the incapacitation or capture of the player characters?
• What, besides a direct threat of violence, will cause the player characters to draw their weapons?
• What choices have I baked inside this encounter for the party to take besides “target the enemy,” and what are the impacts of those choices?

One of my favorite published encounters (or set of encounters) that helps illustrate this is Chapter 4 of Waterdeep: Dragon Heist. In the first half of this chapter, the players are led on a string of encounters that turns obtaining and keeping hold of the Stone of Golorr into a contact ball sport. As the DM guides the players through chases, hunts, and alleyway combat encounters, the stone can pass hands frequently. Nobody’s looking to kill anybody else. They just want the stone. And the result of this encounter determines whether your party can move on to access the vault with the stone’s help, or if they’re stuck planning to enter a villainous lair to recover the stone one last time.

The first time I ran Dragon Heist, I ran it in the fall. While Jarlaxle was the villain, I wanted to make Waterdeep feel more lived-in. If the stone were that important, I wanted everyone to come after it. I wanted it to feel like a fantasy Rat Race. The final encounter in the autumn chain involves chasing two drow to a Dock Ward neighborhood called Mistshore. The drow stall the party, waiting for their getaway submarine to come and whisk them—and the stone—away.

I turned this encounter into a battle royale. As Jarlaxle brought the submarine to Mistshore’s run-down docks, members from the three other factions appeared to tear the stone from his greedy hands. The party had to choose: who would they work with, and who would they actively attempt to hinder in this free-for-all? At the end of it, the group worked with the drow and just narrowly failed to prevent Manshoon’s goons from getting away with the stone.

The encounter and its fallout felt fresh. The party had failed, but it was a pivotal moment for the story. A new villain. A new objective. And it felt fresh for me, the DM. I had no idea Manshoon would be walking away with the stone that night, and I couldn’t wait to figure out what happened next.

If you break free of the “win-or-die” encounter, then your players can fail in ways that aren’t “die.” The story becomes more malleable and more real. Failure in combat can feel less like an end and more like the dramatic and memorable moment we’re always looking for out of a D&D session.

And during the cool-down the other night after the queen died, my players told me they loved it. In their words, “every story needs an Empire Strikes Back moment.”