How to plan knowing that things won’t go to plan

No plan survives first contact with the enemy.

The above quote is from prussian military thinker Helmuth Von Moltke and has since become a common phrase used by military strategists and other decision makers. The point of it is that no matter how carefully or thoroughly you plan for a battle, that battle plan will inevitably have to be adapted on the fly in the face of the tactics of the enemy.

Moltke did not mean that you shouldn’t make plans, far from it, more that you had to plan with the awareness that you were working with incomplete knowledge and that you would have to adapt your plans in the heat of battle. He meant that if you stuck rigidly to a careful battle plan and ignored what your enemy was doing, you would probably lose and your players probably wouldn’t have much fun.

Wait, what?

Oh yeah, this is a DnD blog. I think it’s pretty obvious how this applies to being a DM. You can’t predict with any certainty what your players are going to do; they have agency to act as they think best. They also have imperfect knowledge about what is going to happen in the future, so will not always make the choice you think they should. They are also playing a Role Playing Game, so can reasonably expect to be able to influence the course of the game. If not, then they might as well read a book or play a linear video game.

I think once we as DMs embrace this philosophy (not the part about the players being enemies, obviously), it can fundamentally change our approach to how we plan and how we think our game should go. As I mentioned in my previous post, this change happened for me after the first campaign I ever DMed, and the campaign I am currently DMing is unrecognisably different from the previous one.

I want to preface what I’m about to say by saying that I think my first ever campaign went very well. The players and I had a lot of fun. There were many epic moments, the characters, both PCs and NPCs were really strong and the storyline was really interesting. But here’s the problem. It was my storyline. I think I could take that storyline and play through the entire campaign again with a different set of characters and it wouldn’t be all that different. Even if the players made different choices, the overall flow of the narrative would be the same.

My overall plan for that campaign did survive first contact with the players, and I think the game was poorer for it. While the Von Moltke quote is more a caution to generals not to stick rigidly to plans and to expect the unexpected, I think when applied to DnD, it’s more an encouragement to embrace the unpredictable nature of playing a game that is fundamentally about collaborative storytelling.

How do we plan with that in mind? As I said, I have totally changed how I DM. My previous blog about how I use mechanics and concepts from Monster of the Week factors in here, particularly the bit about planning what the enemies will do, rather than focusing on the players, and the game being a conversation, but there are three additional changes that I have made which influence how I plan and run DnD. Firstly, plan problems, not solutions. Secondly, don’t plan too far in advance, and thirdly, don’t feel like you always have to have an answer.

Plan problems, not solutions

Puzzles are and always have been a core part of DnD. We expect as players that there will be puzzles for us to solve in dungeons, even when the presence of such puzzles doesn’t really make a whole lot of sense. But here’s the problem with puzzles: generally speaking there is only one solution to them, so solving them either becomes a case of collecting the clues the DM has scattered around the room and putting them together, or guessing what is in the DM’s head. This can be fun, however if done poorly it can stop a session in its tracks, and in order to do them well, they can often take ages to put together.

I prefer to create problems for the players. Stick a raging river in their path, sit back and watch what they do to get across. There’s no planned way for players to cross. If they ask if there is a tree or rock nearby from which to tie a rope (other solutions are available), then think about where the river is and decide there and then. The solution to the problem is up to the players, your job as a DM is to put the problems in front of them, then interpret through the rules how their solution resolves.

Obviously, you can try to anticipate what players will do. Think about where a rope could be tied, think about the spells they have access to, think about how different skills or abilities could be brought to bear, but if you spend time planning in detail how this river can be forded, then I guarantee players either won’t think of it, or feel like there was only ever one solution.

I’ve done puzzles. I still do puzzles. I think puzzles are fun and generally speaking players really enjoy them, but puzzles are also really hard, and if all you ever do is give your players puzzles, you will run out of ideas and the players will feel like they are trying to beat you and your puzzles, rather than create a story with you.

Problems give the players agency, and have a much better ratio of planning time to playing time.

Don’t plan too far in advance

A few sessions into my first campaign I had basically mapped out in broad outline every single session of the campaign. No, seriously. I didn’t have the details all worked out, but I had the outline of the whole campaign written down. I knew what the story was going to be, it was just a case of the players playing through it.

Don’t do this.

For a start, it’s a lot of work. It also takes away player agency. At one point I presented players with a choice of joining the evil cult they had infiltrated or having a boss fight. The choice was really just a dramatic ultimatum given by the leader of the cult before the boss fight. They decided to join the cult.

This is why you don’t plan too far in advance. I had to change some plans, but ultimately, I was able to work it so that this choice didn’t really affect the overall story. In other words, the choices made by the players didn’t mean anything. The players didn’t have agency. I think they felt it too. They felt bad for not doing what I thought they would, so quickly betrayed the cult leader and did the boss fight I wanted them to do (what was epic by the way!).

If my planning horizon had been less far ahead, I could have had the whole campaign pivot on that exact moment. I could have had the players be part of this cult for a number of sessions, deciding whether and when to betray the leader, and how they could use the power of this cult to their own ends. And how many bad things they would be willing to do in order to not blow their cover.

By contrast, for the current campaign I am running, I have the upcoming session planned, and I have some vague ideas about where I want the story to go, but most of those have been generated during sessions or shortly after them, based on things players have said and done. In fact, the whole campaign is centred around a series of magical items which only exist because the rogue said she was going to steal an amulet off her former boss.

This is what we mean by collaborative storytelling. I try not to generate all the ideas and all the plans on my own. I let the players generate the ideas and build on them. They don’t necessarily know this yet, but I plan on making this explicit fairly soon. The other items they need to find are going to be scattered around the world, but I have no idea where. At the table we will discuss different places and ideas that we want to explore, and I will plan based on that discussion.

Don’t feel like you always have to have an answer

This is similar to what I just discussed. There’s a school of thought when it comes to DMing that you have to have this deep, elaborate lore built up about your world. You have to do what Tolkien did and create this huge, fully-realised world, with a long history and complex political structure. You have to have a map with every area on it detailed meticulously. Sometimes, it is even suggested that you have to have this all prepared before players even get to the table.

This is total nonsense.

I don’t think anyone actually does this. I think it is an aspiration that many people have because they want to do world-building like a fantasy author. The problem is that Tolkien spent literally decades creating Middle Earth. He published his books while in the middle of that process. He built the world as he went along.

He was also a single man writing books on his own. This is not the same as being a Dungeon Master. For a start, most DMs are not professional writers whose job it is to build worlds and write books. We have day jobs, and sessions to plan, and other people to help us build the world.

I think the above school of thought stems from a fear of a player asking a seemingly innocuous question, like “where do the elves live?” and not being able to answer. We’re the DM, we should know, right?

But ultimately, the answer to ‘where do the elves live?’ probably doesn’t matter. If it mattered to the session you were currently playing, chances are you would have thought of it. If you’ve not considered it when planning the session you’re playing, then it can’t be directly relevant to the session, so you don’t need to know where the elves live.

That the player has asked the question is not a problem, it’s an opportunity. An opportunity to ask the player where they think the elves live. This gives them an opportunity to worldbuild with you. You can turn it into a conversation about what elves are like in this world. They can then influence your planning for when elves turn up later in the game, or even for your roleplaying of elves in this session.

I recently had a player send me the backstory for her character. She said she would happily change any of it if it went against my plans and my response was that so long as it didn’t disagree with what had been said in any sessions to date, I would change my plans. This game is a collaboration between all the players at the table. The DM doesn’t have to have all the answers. It makes for a much more engaging and interesting game if you work together to build the world.

No plan survives first contact with the players. This is because players have their own mind and agency to do as they want. You cannot possibly hope to predict what your players will do in any given moment, because you are not them. This is not a problem. This is a virtue of TTRPGs. They allow a group of friends to come together and tell a story together in a structured way. Your job as the DM is to facilitate and direct this storytelling. Your plans will not and should not survive the session. That’s ok, you’re not fighting a battle with your players, you’re telling a story with them.