Serving the Story - Narrative and Game Play in D&D

I’m going to take a dive into an issue that will likely be contentious, but since many of us are at home with time to kill, why not?

For many, D&D is a “storytelling” game, they feel that the process of playing D&D is a process of telling a story. For others, D&D is a process of playing a game, the story comes out when you talk about it later.

I think D&D can be enjoyed either way you think about this, there is no “right or wrong” to how you choose to approach gaming. Whether you as the DM think of yourself as a storyteller or just a referee you can have a fun game.

One thing I have noted though, is the tendency to see people making suggestions like these:

“Don’t fudge the dice, unless it serves the story”

“Make sure when you are running a game that everyone gets the chance to shine”

Now, before I get into discussing these, I want to make it clear that there are certain instances where I agree with these ideas. For example, let’s say that you are a new DM and the party sets out for its very first adventure, five minutes in you roll the dice and a trap kills one of the PCs. Now everyone has to wait around while you roll up a new PC for that player, and they may just have been put off playing the game forever, dying so quickly in such a mundane way. 

Or, for example, you might have a player in your game who is dominating everyone else at the table, speaking over them, directing their characters, and generally making the game less enjoyable for everyone. 

In cases like these there is probably a need for the DM to intervene in some way, this is a game, and sometimes it goes sideways, so it isn’t a bad idea to do some course correcting.

However, I’m of the opinion that this sort of stuff should be minimized as much as possible. 

Serving the Story

When the dice turn up something that you don’t like, as you feel it doesn’t “serve the story”, ask yourself, what story isn’t it serving? This seems like a simple question, but I find that DM’s have a difficult time answering it in any concrete way. 

It can’t be the story as the players think of it, as when you roll the dice to see what happens to the players they aren’t the ones deciding to change the results, you are. Now, some games explicitly give the players the opportunity to change dice rolls (for example, Stars Without Number allows experts to re roll failed skill checks, or warriors to re-roll failed attacks), but D&D is not one of those games.

So when a dice roll “doesn’t serve the story” the DM makes this call. 

There are a few reasons why this can be problematic. 

Predictability 

People are notoriously bad at spotting their own patterns of behavior, it’s often blindingly obvious to others, but not to you. This is what good poker players rely on. So one concern with being a DM who directs things to “serve the story”, is that you might (and quite often do) fall into predictable patterns of storytelling. Indeed, I have seen many people online suggest that every session of D&D should have a basic narrative pattern, e.g. exciting inciting incident, pause in the action, discovery and investigation, near death or threat, reversal of fortune, big exciting event, that sort of thing. 

I used to have patterns I wasn’t aware of, it wasn’t intentional or planned, but I would tend to make sure that initial encounters were dangerous but not lethal, and ensure that there was a big fight at some point, often with the main opponent “just” getting away to draw the players into the next session.

It worked pretty well, until my players noticed it, and started to get cheeky. Since they realized that I wouldn’t kill them off in the initial encounter, they started doing crazy stuff to see how far they could push it. Then they just got bored.

Nothing drops you out of the immersion sweet spot faster than seeing the Great and Powerful Oz shifting levers behind the curtain. A predictable game is a dull game, and one that players will lose interest in pretty quickly.

I have been DMing for 3+ decades, and I run three concurrent campaigns at the moment, all of which are ticking along nicely with enthusiastic players. I manage to avoid predictability because I DON’T have a story structure in mind, I let the player’s actions drive the game, not my ideas as to where the story should go.

Responsibility

One aspect of this that I almost never see discussed is how “serving the story” puts a ton of responsibility on the shoulders of the DM. The DM has to have some sense of what the story IS, in order to determine if the action or result fits that story. The DM is also responsible for making the story move forward. If the players make a decision that doesn’t serve the story they won’t know it, as they don’t know what the story is supposed to be, so it's 100% the DM’s responsibility to track and correct to ensure the story is being served.

Not only that, but if the story takes a bad turn, if something happens that is upsetting or frustrating to the players, it is the DM’s responsibility, as they are the ones charged with making sure the game serves the story. On this model, if you don’t like the story being told at the table it’s the DM’s fault, as the DM’s job is to ensure the game is being true to the story.

That’s a lot to carry around on your shoulders. Worse still, on this model if you do like the story being told it’s also the DM’s doing, as they will be working behind the scenes to ensure that the story is being served. In a very important way this robs the players of their victory. How can anyone at the table playing a game where the DM alters results to “serve the story” know if they achieved their goals because of good decisions and clever play, or if they won the day because the DM fudged the dice to “serve the story”?

The DM has a lot of power in D&D, and once you make the decision that “the story” is to be served, since the players can’t change things behind the screen, the ultimate responsibility for making the game enjoyable falls on the DM. I don’t think this is healthy or fair.

Railroading versus Sandbox

There are two options available to a DM who is letting the story decide what should and shouldn’t happen in the game, one extreme is “railroading”, where the DM has a very strong idea of what should be happening and when, and then forcibly manipulates the game into making those things happen. Some degree of railroading is fine, but heavy handed railroading is not, it can make the game feel constrained and it utterly removes player agency. 

The other option is “sandboxing”, which essentially means you have no predetermined idea as to what is going to happen, and you improvise on the spot. However, if you have no idea what is supposed to happen, then how can you direct the game to serve the story? Sandboxing DMs who make changes to serve the story often do so in bizarre and inconsistent ways that make for a choppy, arbitrary and tonally chaotic game. 

Either extreme, when combined with a desire to “serve the story”, can make for unpleasant gaming.

Collaboration

This is one of the dimensions of story serving games I find the most baffling. For all people’s talk of D&D as a collaborative storytelling enterprise, the DM deciding when something “serves the story” is a crystal clear example of non-collaborative storytelling. Yes, the players contribute by deciding on their player’s actions, but the DM has the power to change anything in the game, and nullify the impacts of those actions completely. 

Changing aspects of the game to serve the story makes D&D a non-collaborative storytelling enterprise, it takes the players out of the equation, removes their agency, in order to ensure that a greater narrative is served. 

And this can even happen in games where the DM “reads the table” to decide if the events serve the story or not. One thing that people often forget about narrative is that the reader brings her own particular perspective to the process of interpretation and meaning. What the story means to you is often not what it means to others. 

So a well meaning DM who makes alterations to ensure that the game “serves the story” may do so in ways that work for a certain number of players, even the majority, but not for all. I’ve seen this at the table before, the DM fiddles with things to ensure a certain outcome she feels serves the story but one or more of the players don’t like how it has turned out. 

Collaborative Gaming without the Storyteller

The best way to ensure that your D&D game is truly collaborative, whether you view this as collaborative storytelling or just collaborative play with emergent storytelling, is to let the events unfold as they may. When you do this, the players are collaborating by directing their PCs to perform certain actions, and the DM is collaborating by creating a game world and having it react to player actions. 

Indeed, that unique hybrid of player decisions and DM reactions to those decisions is a powerful combination, it feels less like a story you are being told, and more like a story you are participating in. It maintains a sense of agency and accomplishment, as you know the highs and the lows are yours, not a result of the DM taking pity on you, or playing favorites, but instead the result of the combination of your decisions, the game world and the dice.

Two examples to close this off. 

I had a very telling experience early on in my gaming career. In college I was running a game for my friends and there was a paladin in the party. Paladin’s have strict conditions on their behaviour that make them challenging to role play. One of my players was doing a great job at it, sacrificing himself and many of the rewards that came his way to behave as a paladin should. 

At the culmination of this campaign the party confronted an arch-mage that had been working against them for the past year. He was responsible for slaying the paladin’s war horse, one that the player had loved, and he also blinded the paladin in one eye. The party wanted a piece of this guy like crazy. And I decided that it would make for the most satisfying storyline for the paladin to be the one to finish the guy off. He had sacrificed the most and played well, and I thought he deserved it. It would serve the story for the paladin to close the campaign this way.

So when they finally faced off against this guy things did not go as I had planned. The paladin made a lot of terrible rolls, and some bad decisions. I fudged the dice pretty hard to ensure that he wasn’t killed during the fight. 

The party wizard, however, was rolling hot, and made some excellent tactical calls. He managed to set himself up to cast a powerful spell from a scroll, in 1e wizards can cast higher level spells than they can normally cast from scrolls, and he pulled out a scroll with disintegrate on it. I rolled to see if the spell failed behind my screen and it was successful, but it felt wrong to me to have the wizard take out the arch-mage, after all the paladin had been through, it didn’t serve the story for the wizard to be the one to take him out, so I fudged a failure of the spell, and the paladin did the deed.

It was an exciting session, and I felt it was a good ending to the campaign, I thought it was a good story, and I had ensured that the story ended on this high note. I was also sure that the players had no idea what I had done.

However, a few days later we are all talking about the end of the campaign and I mentioned how cool it was that the paladin had finished the guy off, after all he had killed his war horse and blinded him, so the ending was dramatic and fitting. Several of them disagreed.

“Well, we all knew he was going to finish him off, the paladin should have died halfway through that fight”

I was floored, I had no idea what I was doing was so transparent, but there it was. They felt like I had engineered the whole thing, which I essentially had, and it wasn’t anywhere near as satisfying.

Fast forward to 6 years ago when I was running a game for my son and his friends. They were going through the Expedition to the Barrier Peaks, and there is a section between the decks of a space ship where there is a doppelganger and an intellect devourer working in tandem.

The party split up and the wizard was waylaid by the doppelganger. It was a bad surprise roll on his part, the doppelganger surprised him, and the intellect devourer slayed him and took over his body, killing him instantly. 

The temptation on my part to change that roll was intense. The player had no chance to react to this, he was surprised and lost. It was like getting one-shotted by a trap. Did it “serve the story” for a PC to be killed that quickly without any real chance at response? This was a high level PC as well, and one that the player loved.

I let it stand, and the now possessed PC rejoined the party, and waited for an opportunity to strike at the group. He took it soon after, and there was an epic fight where the party was baffled as to why the PC was attacking everyone (I had him run the PC even though it was taken over to ensure that the party wouldn't know what was going on). The party thief managed to maneuver behind the intellect-devourer-possessed magic-user and backstabbed him, slaying him again, and the intellect devourer skittered off into the bush. 

The party tracked it down to gain revenge on it for making them kill their fellow PC, and they cast a reincarnation spell on him from a scroll and he came back as a wemic. He played that wemic magic user for two more years, and had a blast doing so. They TO THIS DAY talk about that fight, and TO THIS DAY my son’s friend will tell you how amazing that PC was.

If I had done what “the story demanded” and kept the PC alive then none of that would have happened. Over the years I have discovered that whatever I have “lost” not directing the game to “serve the story” I have regained in terms of a far more interesting story that I can tell about the game after the fact. Sometimes the “story” is best served by letting the game happen, seeing where it goes, and talking about it afterwards with your players and others. That way the story serves you, not the other way around, and it evolves in unpredictable, exciting and immersive ways.