Agency and the Amusement Park


There have been many discussions about Player Agency and the differences between Sandbox and Railroading. I believe these topics are intertwined and it is nearly impossible to discuss one without the other. My blog has been silent for far too long and its high time I get my fingers going again.

I’m going to throw my hat into the ring, and this is what I’m going to talk about. I’m breaking this conversation out into four parts. Unfortunately, I exceeded my goal of five thousand words. At least I didn’t go over ten thousand.

Part 1 covers Player Agency. I break down Agency to three distinct aspects: Tactical, Situational, and Narrative.

Part 2 talks about the differences between Sandbox and Railroad.

Part 3 counters those that claim railroading when it really isn’t.

Part 4 proposes a hybrid between Railroad and Sandbox, which I call the Amusement Park.

Without further ado…

Part 1: Three Types of Player Agency

Player Agency is what makes D&D and other role-playing games (RPGs) what they are. It’s what separates them from board games and computer games, even those computer games that claim to be RPGs.

I’ll start with a simple definition: Player Agency is the degree of choice a player has within the game’s setting. Imagine you’re standing in a room with three doors. One door is to your left, one door is in the middle, and the third door is to your right. You know, through some observation, some information about these doors. The left door is red. The one on the right is blue, and the one in the middle is yellow. Based on this information, you are given a choice. What do you do?

If you’re playing a computer or board game, you only have 3 choices. Well, you could always quit, but that option is always available. Or you could cheat.

Player Agency presents you with a multitude of choices. Your obvious choices are to open and go through one of the doors, but if you think about it, your choices are not so limited. Perhaps you want to sit in the middle of the room and wait for a while and see if something might happen. Maybe someone might come through one of the doors. You could go to each of the doors and feel it or listen for noises coming from the other side, and hopefully glean more information about what is beyond it. Maybe you can examine the room you’re in, because I haven’t given you much information and there’s something you’ve overlooked. What if you just stand there and yell “Help!”? How about this: you take an axe out of your pack and start hacking at the wall.


The idea of Player Agency in a role-playing game means that as a player, you are not limited in your options. Having options makes the game fun. But more than making decisions, you must also deal with the consequences. You, in cooperation with the other players and the Game Master, are building a narrative story. You’re not reading a novel or watching a TV show or movie. You’re not constrained by the limiting rules of a board game, or the lines of code written by a team of programmers. The experience is interactive and unlimited.

I believe there are three distinct types of Agency. Each of these types are very important, and they are important to different types of players. Some people value one type more than others. The challenge we face is balancing these three types of Agency so that everyone is having a fun time and we leave the session feeling satisfied and engaged. As a Game Master, I must provide balance so the players get what they want. Players must be cognizant of the other players’ needs and desires and be patient during those moments when you’re not getting what you specifically want.

What are the three types of Agency?

  1. Tactical
  2. Situational
  3. Narrative

I’ll break down each one and give an example. I’ll talk a bit about the types of players who gravitate toward that type of Agency, too. I’m not a psychologist. I’ve just been a Game Master for a very long time. I’ve also done a bit of research and read what others have said on the subject. You can read what they said, as well. In fact, I encourage it. It’s a smorgasbord. Take what you like and ignore the rest. Play and run the game that works for you and your players. If you’re having a fun time and your players show up to your games, you’re obviously doing something right. If not, spend some time, reflect, do some soul searching and retrospective; figure out what you’re doing wrong and fix it. Apologize to your friends or make some new ones.

Tactical Agency

Tactical Agency is about making choices in battle or other tactical situations. When the game is broken down into pure mechanics and we’re relying on random rolls to determine the outcome of events, these are the choices a player has at this level: Which enemy do I attack? What weapon or spell do I use? Can I maneuver to get behind the enemy so that I can get flanking, or perhaps help an ally get flanking later? There are a lot of choices here, and for some, many of these choices are obvious.

Some players think this is the best part of the role-playing game. They love rolling dice and living by the results. They love working the numbers, rolling and counting damage dice, removing dead enemies from the board, and seeing progress. They accept when one of their own falls and they accept the results when their character takes damage. They build their character around its combat abilities and they live for the thrill. For some of these players, the stuff between combat encounters is boring. You might call these players murder hobos. You might be right. Not always. In a lot of ways, tactical combat is a board game. There are clearly-defined rules of what each participant can and cannot do. Can you stretch the rules? Sure. We’ll cover some of that in Part 3.

Tactical Agency can be lost or stolen. A Game Master should be careful and watch for certain things, because if it happens, players can lose their Agency and won’t have fun. The greatest theft of Agency is the Alpha Player. One dominant player dictates to other players what they should be doing. “You. Go here and attack that creature. You. Cast fireball over here.” The weaker player may be rolling the dice and moving their miniature on the battle map, but they aren’t really participating. They aren’t making their own decisions. As the Game Master, you must recognize and stop this immediately. Tell the Alpha Player to only talk on their initiative turn and encourage each player to make their own decisions on their turns. Give each player, especially the new ones, time to learn and grow. The second Agency stealer is you, the Game Master, in how you set up the tactical situation. It’s not easy, but I try to make every encounter unique. It’s not just the choice of creatures in the battle, but the environment and situational condition where the battle takes place. In my setting, there’s one “dungeon crawl” where the environment doesn’t change from battle-to-battle. The last time I ran through it, the players were bored with it by the time they found the exit. I’m making some adjustments before the next group gets there.

Situational Agency

Situational Agency is about making choices outside of combat, usually in social encounters. These players look for ways to avoid combat and use their character’s non-combat skills. Puzzles and mysteries are great, but if the player can’t solve them, their expectation is that a die roll on an appropriate skill will.

It’s not that they will completely avoid combat. If backed into a corner, they will fight. They will also fight rather than surrender, even if the odds are overwhelming. At the same time, a group of these players will plan and plot for three hours of real time to execute a plan that takes one round of game time. I’ve seen it done. Trust me. And if they do, and it works, they should be rewarded for it. This is Situational Agency at its best. I watched a group of five 9th level characters defeat a Rakshasa in one round.

Players who love getting “into character” are your Situational Agency players. They love interactions. They love talking and acting out. They want to play out the conversation with the blacksmith and negotiate the price for their new masterwork longsword. They want to skulk the dark alleys of the city, find the thieves’ guild hideout, and find out what’s really going on. As Game Master, these players are a lot of fun, as they will force you to stretch your imagination, get you to think outside of the box, and, quite often, get you to share secrets and clues that you hadn’t planned on revealing. But there are drawbacks. These are the same players who will search the dead bodies after a fight and get upset with you when you report openly what they find. You must write a note or text them. Likewise, they will split away from the party taking your time for one-on-one asides, leaving the rest of the group to sit and play Angry Birds (or whatever the popular phone game is these days. I can’t keep up.) There’s no Agency when you’re waiting for one player and the Game Master to return to the table.

Something else to be careful about: these players are incredibly easy to manipulate. Dangle the right kind of carrot and their character will follow it. Nearly every time. Sometimes they will go looking for the carrot. I have a player in one of my groups that every time they are in a city, he wants to send his familiar to go looking for stuff that’s easy to steal. His character is multiclass rogue/warlock, so of course, it’s totally “in character.” Following a carrot is not Agency. You may need to use this gimmick to get them back on focus, but don’t do it more than once. Players will catch on to these tricks.

Narrative Agency

Narrative Agency is the big one. It’s the deal-maker or the deal-breaker for a lot of players. This is the one that bridges the gap between the other elephant in the room: railroading. Before I talk about railroads and sandboxes, I need to define this. Narrative Agency is when the players have a control over the story. Matt Colville provides a very eloquent description in his video, which a provide a link at the end of this blog. I highly recommend it.

Here’s a map with some pictures, names, and some geographic features. The Game Master has a few notions of what is here or there, but nothing specific. The players huddle together like a squad of American Football players and suggest an idea. We want to start over here, travel over there, do some things, collect a bunch of treasure, defeat some bad guys, and have a great time doing it. The Game Master pulls out a few books that are loaded with random encounter tables and proceeds to run the game. The players have complete Agency. They chose the story, the plot, the narrative, and are essentially the authors of the story. The Game Master is a facilitator. I describe this as an extreme case. I’m not saying it doesn’t happen, because it does. I’ve run this type of adventure. It can be a lot of fun. For the player. It can also be very frustrating. For the player. The reason will be made clear in Part 2 of this article.

Before we get to that, I want to provide an example where there’s almost no Narrative Agency. It will allow us to segue better into Part 2 as well. At the beginning of the first session, instead of pulling out a big map, the Game Master puts a published book on the table. There’s a beautifully-painted cover with a title written in a fancy custom font that is probably hard to read except that it’s written in 64-point size. “We’re running this module. I hope none of you have ever heard of this, or if you have, you don’t know anything about it. Let’s make your characters.” One player says he started playing this module two years ago in another group, but only got into three sessions but the group broke up when the Game Master had to move away. “No spoilers, then. I’m trusting you to not metagame.” The player immediately tells the other players, “We need to be sure we have at least three characters with healing abilities because the monsters in this module tend to target characters with healing abilities.”

The Game Master politely ignores the comment and proceeds to read from the first page of the book, “You are all starting in the town of blah-blah-blah, in a tavern called blah-blah-blah, when a man dressed in purple comes in looking for adventurers. You just happen to be sitting in there looking for work, and you have nothing better to do. You can come up with a back story if you want, but it really doesn’t matter to the plot, though if you do, and you play your character accordingly, there are some rewards along the way, like experience and magic items.” He holds up the book and on the back page there’s a half-page table with the table entitled, Rewards for Good Role Playing. He reads the blurb, “When a player exhibits exceptional role playing, consult this table and roll d100. For example,” He rolls the dice. “57. You find a dagger that glows when it comes within 10-feet of a secret passage.”

This is the other extreme where there is virtually no Narrative Agency. The players are playing a published adventure module where nearly every detail is already pre-determined. A player can’t say to the Game Master, “I want to be the prince of the neighboring kingdom, where my father is on his deathbed, and I’m in this region because part of the requirement before becoming king is to prove my worth to my people by defeating a great evil.” There’s nothing in the adventure module that covers this. Another player can’t say, “My character is actually related to the big, bad evil guy. I’m his long-lost daughter and if I can prove it to him, he might choose to become good instead.” Great story ideas. Great narrative. Yet, when playing published modules, I rarely hear people complain that they don’t have Agency. They’re happy with their available choices: you can go through the left door, the center door, or the right door. You can hack at the wall all you want, but nothing happens.

Part 2: Railroading or Sandboxing, or Is There Something in Between?

If you so much as mention railroading online in a forum or discussion group, you will be tied to the busiest high-speed commuter tracks available. Railroading equates to saying “no” to players. The players want to do something and the Game Master says “no.” Oh oh no no no no no no no. Oh mama mia, mama mia, mama mia let me go. Beelzebub has a devil put aside for me? For me? For meeeeeeeeee? (Thank you, Freddie.)

But you leave me no choice!

There are rules. There’s physics. There’s logic. Sometimes the right answer is “no.” When the right answer is no, it’s not railroading. There are other times that may seem like railroading when it’s not. I’ll cover it. Let’s be clear on what railroading is, first.

Railroading is specifically taking away player’s Narrative Agency. Railroading is not taking away Tactical or Situational Agency. If someone is taking those away improperly, it’s just bad gaming. I think it’s worth talking about, but it’s not railroading.

I ended Part 1 talking about published modules. Even unpublished modules can do this. Think about one-shots and convention games. The Game Master shows up, meets 4-6 players that may or may not be new to the game, hands out pre-made character sheets with pre-written backstories and proceeds to run a very short adventure lasting roughly 3-6 hours of real-world time. Often, the game system is new for some or all the players. The scenario is straightforward, stereotypical, and lacks depth and subtlety. It’s usually a lot of fun, and it’s a great way to spend a few hours. No one complains about railroading. The Game Master came prepared to run a very specific adventure, and the players don’t try to do anything beyond the constraints. It’s kind of like riding a roller coaster at an amusement park. It’s short and it’s exciting. Maybe you buy the pictures they took during the ride, and maybe you throw up in the barrel. Once you sat down, they strapped you in and they snapped the bar in place, you were in it for the ride. There was no getting off. There was no turning left or right, taking the high hill or the low hill. One track, one set of rails, forty-five seconds of pure adrenaline rush.


Don’t complain. You saw the advertisements on TV. You had your tickets to the park for six months. Your vacation was scheduled. You’ve been looking forward to this since they built the ride. Weeeeeeeeeee!!!!!!!!!

Just because the module the Game Master is using isn’t published doesn’t make it wrong. The Game Master spent a lot of time planning, designing, and writing this adventure. It’s not that railroading is good or bad, it’s about your expectation. Even if the Game Master’s setting is homebrew, it doesn’t mean their world is sandbox.

Let’s talk about sandbox. What is sandbox?

Sandboxes are fun. When I was young, my dad bought four 1x10x12s and several bags of sand. With another board cut into corners, he set them in the back yard, outside the area where the dog’s chain reached and the underground septic system, nailed the boards together, and dumped the sand into them after churning up the grass and dirt, and told me and my brother that we could play there. We had a sandbox. The boards lasted one season. We played in the area for six or seven years.

I can’t remember everything we did there. We had Tonka cars and trucks that we made roads for. We broke branches off the nearby bushes to make trees. We had little plastic soldiers and we reenacted the Battle of Normandy, though we got in trouble when we threw rocks and clods of dirt at each other. Plastic astronauts turned the backyard into Mars. Leftover bricks from remodeling the house became building materials. If a paving stone cracked, we claimed it. In middle school, my brother and I made 16mm and super-8 movies. Fireworks were illegal in our state at the time, but a friend procured some. We enjoyed setting cherry bombs off under the soldiers and smokers under the tanks and filming it for the movies.

The stories we told were endless. Every weekend it was something new. Something different. Whether we had friends over, it didn’t matter. At the end of the day, mom would strip us down to our underwear in the garage and hose us down before coming in for dinner.

I remember mowing the lawn during high school finding plastic soldiers or astronauts stuck in the dirt.

In role-playing it’s no different; a sandbox is where the players get to decide where they want to go and what they want to do. The Game Master’s job is to put things in the way and run the encounters. Though it’s not as simple as it sounds. The setting I’m currently running started out as a sandbox, but I discovered very quickly that it doesn’t work. Not with the players I had. I’ll explain in Part 4.

I want to break down the sandbox setting into more detail. What’s good about it and what isn’t. Ideally, the sandbox setting is the ultimate playground for Narrative Agency. The players are in control of their characters’ destiny. Here’s the map. Here’s your starting point. Where do you want to go? Here’s something cool on the map. Let’s write our story. Our heroes start with humble beginnings, we face trials, fight great beasts, overcome dangers, solve puzzles and riddles, and eventually achieve the goal we established for ourselves. As players, we escape into a world where we live out our fantasy dreams as wizards and paladins, elves and dwarves, heroes and heroines. We were paupers who became kings. What could possibly be wrong with that?

A lot of things. Really. A lot of things. Number one: a burned-out Game Master. Only one thing will burn out a Game Master faster than a pure sandbox setting: running nothing but published modules. What? (I hear the needle scraping the old vinyl record.) Running someone else’s stuff get’s boring. At least, it is to me. There’s no creativity in it. Therefore, I don’t run Adventure League or Pathfinder Society. I want to run my own stuff. Likewise, I won’t run pure sandbox, because pure sandbox will burn me out. Why? Let me count thy ways.


Since there is no set storyline, and since the players can change their mind on a whim, there is no way to predict what they will do. While adaptability and the ability to improvise are important for Game Masters, doing it one hundred percent of the time is taxing. It is difficult to ensure that every encounter is unique, original, fresh, inventive, and fun. This is especially challenging for me because I like to be prepared. I’m just OCD enough that I don’t like to use dice for creatures. While I don’t use miniatures, either, I have printed standees with images of the monsters and I have index-card-sized cards that have all the information I need to run the battle on them. I can’t do that when every encounter is randomly generated.


Likewise, as a storyteller, I like my stories to have a certain amount of flow and continuity. Novels, movies, plays, etc., follow a predictable pattern. There are acts or chapters. Within each are scenes. There are highs and lows where pressure ebbs and flows. This isn’t something I came up with. This is how literature works. Every good story follows this pattern to some degree. The same must be true for an adventure. Why is Raiders of the Lost Arc a great movie, but the sequel, Temple of Doom, is considered mediocre? Because Temple of Doom doesn’t follow the pattern. It doesn’t give the audience enough break between action scenes. Instead, you get shock. Eyeball soup and monkey brain stew. Each encounter should logically flow to the next. Each encounter should have a logical reason for being where it is. Random monsters don’t stand behind doors guarding a pile of treasure waiting for adventurers.

I’m Not a Robot

There are hundreds or more random encounter generators available on the market or for free in the marketplace for virtually every game system. These generators are broken down by environment, character/monster level, combat/non-combat, or several other factors any Game Master could ever want. Some are online, some are printable. Some are in books from official publishers and others are homebrew-sourced. I can even randomly generate a map. What do you need me for?

But, But, But

We’ve all heard the story of how the Game Master showed up to the session prepared for an adventure. The players had other ideas. Instead of going into the dungeon to claim the MacGuffin, they decided to leave it where is and run off into the wilderness, find the witch’s tower and ask her for some of her world-famous homemade apple dumplings. The Game Master quietly sets the stack of papers on the floor, sighs, and pulls out the dungeon master book and starts rolling random encounter tables. This can happen to you. A friend told me that the way to win RPGs is to get the Game Master to say, “F. U.” This is one way to do it. Do it more than once, and you’re likely to not have a Game Master much longer.

The point is that Game Masters put a lot of work into building a setting. I can take a blank piece of printer paper and hand-draw a decent world map with all the features in an hour. Mountains, swamps, forests, rivers, cities, roads, trade routes, and so on. I can think up a few tropes and draw them on the map, like dragons and elves, dwarves and ancient ruins. Setting up the foundation for a sandbox is easy. I can add some framework around these tropes and maybe create some semblance of a story or history. I can then use a random generator to give the cities and geographic features names. I’ll add lines to indicate regions and national borders. I’ll make some decisions on where natural resources are likely to be and this will help me figure out where conflicts are likely to arise. Slowly, the setting begins to take form. Stories arise. History grows. Continuity.

Enter the players. No plan of battle ever survives contact with the enemy. I don’t know who first said that, and I’m too lazy to look it up. Tell me in the comments. It’s not that the players are the enemy, but they are the foil. They are also the collaborators. They are also the reason we, the Game Masters, do this. There’s no way, unless your name is J.R.R. Tolkien, you can create a fully-encompassing world history for your fantasy (or whatever) setting that covers every region for every year for every race for every religion. There will be gaps, and you want there to be gaps. Huge gaps. Wide gaps. These gaps are for your players to fill in with their characters’ backstories. This is the start of how to properly implement a homebrew sandbox setting, with a non-railroaded plot. We’ll finish this conversation in Part 4.

Part 3: Just Because I Said No Doesn’t Mean It’s Railroading

But before I do that, I skipped a few important things that are important to cover. At the beginning of Part 2, I mentioned that there are times it’s okay for the Game Master to say “no” to players when it’s not railroading. Something about Rules, Physics, and Logic.

There’s one other aspect I want to cover first, before I get to that. A few months ago I wrote and published a FREE supplement on DMsGuild called Casual Conversations. Here’s the link:

I’m thinking about making a follow up to this where the party’s reputation modifies the result.

I went to FaceBook and solicited ideas. Those that gave me ideas are credited in the document. The description and request were straightforward: You’re in a tavern or town square. You’re listening to the locals and what they are talking about. What are they talking about? Roll randomly on this table. As the Game Master, you want to give the players an idea of what’s going on in this town without throwing them a bone or giving them a side quest. Not everyone in town needs something done. Most people are just talking about local issues or their own personal problems. Not stuff the player characters can or should get involved with or even care about. Several people responded to the thread accusing me of railroading. I was shocked. How does this constitute railroading? Not offering random quests, missions, tasks, or distractions, when the PCs already have enough on their plates, isn’t railroading. It’s saving me and the rest of the group a lot of time and effort. “Yeah, my son is going to ask the Walker girl out for a date. She’s kind of cute. I hope she says yes.” Or “What did you do to get that grease stain out of your apron?” Just because there’s no implied option or quest doesn’t mean I’m railroading the players.

Not saying “Yes,” isn’t the same as saying “No.” They already have plenty of options. I don’t need to give them more. Likewise, if I say, “Well, the guy at the next table is complaining that there are rats in his basement, and he’s willing to pay someone a thousand gold pieces to come and take care of them,” what do you think the PCs are going to do? While in the basement killing the rats, they find a cave that leads under the castle where they find the body of the missing prince. On the body is the clue they need to find evil witch’s tower in the forest (you know, the one with the apple dumplings). Since they couldn’t figure out the other clues, the Game Master had to throw out this bone just to get the players back on track. Remember what I said earlier about using carrots to guide the players? Which sounds more like railroading to you?

It’s not wrong to say no. When you should, that is.


It doesn’t matter what system your using, the game has rules. Some rule systems are tighter and crunchier than others. Some are abstract and narrative in nature. Some have a rule for everything, and if they don’t, there are cross-references and contradictions to make even the most detail-oriented rule lawyers squirm with glee. Nearly every rulebook starts with one that says about the same thing: the game is yours and your rule is law. You, the Game Master, are the final arbiter of all rule disputes. Not the book, not the Internet, not official publications, and not who yells the loudest at the table. Given that, you don’t want your players leaving the table in a huff. The rules are there for two reasons: framework and consistency.

Rules provide a framework within which the game is played. What type of mechanics are used, what dice, what attributes, skill lists, genre, and all the details. When you and your players come to the table each session, there is the expectation that certain things will happen a certain way. If I hit a monster with a sword or shoot the spy with my pistol, I will do a certain amount of damage. If the Game Master must make a ruling on the fly because the rules are insufficient, the advice is to be sure to apply the ruling consistently.

If a player wants to do something and the rules clearly state that they can’t, the answer is no. If the player wants to do something and the rules are unclear, you must make a judgment call. If physics and logic don’t apply, you must weigh factors like game balance, character knowledge and ability, meta-game considerations, and so on. I’ll give some examples in a bit.


Most RPGs are played in fantasy or science fiction genres where the laws of physics don’t necessarily apply. To my knowledge, we don’t know how to teleport or travel faster than the speed of light. Yet, we do this without a second thought in our games. Unless you’re playing in a cartoon world, general laws of physics should still apply. I’m not asking you to get a degree in thermal dynamics or nuclear physics. I’m saying that the simple things still apply. Gravity, inertia, line of sight, two solid objects can’t occupy the same space, and so on.


This is the catch all. Some things simply don’t make sense. A combat round is a certain amount of time, and even if it’s allowable in rules as written (RAW), it isn’t logical. The classic example is the so-called “Peasant Railgun.” In one edition of Dungeons and Dragons, giving an item from one character to another during a combat turn was considered a “Free Action,” which meant, it took no time. The theory was that if you lined up 10,560 peasants and gave the first one in the line an object, who then passed that object to the next peasant in line, and so on. Because of the rules, as written, they could pass that object to the end of the 10-mile-long line. At the end of the combat turn, 6 seconds, that object is moving 6,000 miles per hour. If I’ve done my math right, that’s almost 8 times the speed of sound. This isn’t logical. In a word: no.

Here are some examples for us to consider. Think about how you’d rule in a game you were running. It’s okay if you disagree with me. I accept that we can have a difference of opinion.

Example 1: Running Through the Run Through

My first level barbarian is fighting with a 6-foot spear, and, hey look, the goblins are standing in a line up ahead. I charge at them as fast as I can. I want to try to impale all ten of them at once and make a giant goblin shish-ka-bob.

Answer: No.

Cute. You’re first level. You only get one attack per round and remember that each space on the board is 5 feet. While it may look like the goblins are standing in line, they probably aren’t. Also, they’re wearing armor, they have thick skin, and it’s going to take a lot more force than you can muster to completely impale the first one, let alone run him through with enough force to even damage the second one. If you’re lucky enough to roll a critical hit and do max damage, I might let you roll to hit against the second one, and if you roll a critical hit again, the half excess damage from the first can be applied. That’s as close as you’re going to get.

Example 2: The Immortal Alchemist

My third level paladin, who doesn’t have knowledge in alchemy or nature, wants to make a potion that gives me immortality. Maybe I can’t do it myself, but can I find an alchemist in town that will? Some games allow crafting and provide detailed rules for it. We’re used to video games like Elder Scrolls where you go out and collect ingredients, then stand at an alchemy station and mix these ingredients together somewhat randomly and instantly have dozens of fantastic potions that every general store in every city is willing buy. Many RPG systems provide crafting rules but devote a lot of it to “downtime” activities. There are associated skills and tools. What they often lack, however, is a detail listing of necessary components or ingredients. We might get information about some common crafted items, but it’s not comprehensive.

Answer: Not really, but maybe.

This is a case where the player is unwittingly creating their own quest. Even if the character has skills or proficiencies, at third level, there’s no way they’ll know how to make such a powerful potion. This requires the knowledge and training of a master alchemist. If one is available, he’ll say go bring me the ingredients I need, and I’ll see what I can do, and even with the ingredients, and even if it fails, it will cost you a lot of gold just for the time it takes. The ingredients shouldn’t be easy to collect – like fresh blood from someone that is immortal, quicksilver (mercury), and a few other things. The chance of successful creation should be low. And the time it takes should be long. Furthermore, the efficacy of the potion shouldn’t be true immortality. It just stops aging for a year or two. Result: a huge investment for little return. If the player still wants to do it, okay. I’m not taking away their Agency. Oh, and make an insight check. For all you know, the potion works just fine, but the alchemist kept it all for himself!

Example 3: Swing with Me

It’s the classic scene. The battle is in the great hall of the castle. My fourth level swashbuckling rogue jumps onto the dining room table, pockets a handful of silverware, leaps up, grabs the chandelier, swings the length of the table, does a double back flip in the air and lands behind the villain who’s battling the dwarven fighter, giving both advantage (because I use the flanking rule.) Allowed?

Answer: Of course. Rule of cool, all the way.

There’s nothing in the rulebook about swinging on chandeliers, but it does cover acrobatic maneuvers and using Cunning Action (in D&D 5e, that is.) You use your bonus action and provided you make your Dexterity (Acrobatics) DC 15 check, you’re good. If you fail, the villain gets a reaction attack against you. If you fail by more than 5, your hand slipped when you grabbed the chandelier on some candle wax and you landed on the table and take 1d6 damage from broken glass as a bottle of brandy shatters beneath you. Now, if it was the dwarf fighter trying to do it, he still gets the acrobatics check, but the villain gets the reaction attack either way, because fighters don’t have Cunning Action.

Example 4: Sonic Bolt

I want to create a sorcerer that focuses on sonic damage instead of fire damage. There aren’t very many spells in the book that do sonic damage, so can I just rename the fire-damage spells and say they do sonic damage instead?

Answer: Absolutely not.

This is pure balance. Yes, there are a lot of spells that do fire damage, especially in D&D. Fire is very popular. Creating a fire-based caster is common. However, when you flip through the Monster Manual, you’ll quickly notice how many monsters are resistant and immune to fire. Sure, there are a couple that are vulnerable to fire as well. Conversely, how many monsters are immune to sonic damage? Not very many. Since there are already spells that do sonic damage, go with those. Don’t rewrite the book, please.

Example 5: Healing Cantrip

Along the same lines, how come there are no healing cantrips in D&D 5e? Can’t we just make one?

Answer: No.

If there were healing cantrips, the only reason for rests would be for casters to regain spell slots. D&D 5e’s mechanics are built around action economy and the decisions it forces on players. Among those decisions are when to use healing spells.

Example 6: Fanatical Flying

If I cast fly on the paladin wearing plate armor, can he carry the cleric wearing chain, and the rogue wearing studded leather?

Answer: It’s a stretch, but I’d allow it.

However, I’ll consider you encumbered, and your movement is halved. But my fly speed is… Nope. Half movement. You’re concentrating more on not dropping your friends than moving.

Example 7: Dropping Rogues

Okay, I’m over the bad guy. I drop the rogue on him. How much damage?

Answer: That’s easy.

You make a ranged attack, add your Dex modifier. If you hit, the target takes 1d4 damage (improvised weapon). Whether you hit or not, the rogue takes 1d6 for every 10 feet you dropped him. There’re rules for that. Note that unless you have the Improvised Weapon feat, you don’t get proficiency on your to-hit roll or damage. Who has proficiency weapon: Rogue? Okay. I get it. Rogue, you may use your reaction to make an attack roll on the target. You can also try a Dexterity (acrobatics) check to avoid or reduce the damage from the fall. However, if you reduce the damage you take from the fall, the target’s damage is also reduced.

I hope you get the point. I’m using the rules, I’m using logic, and I’m using a sense of physics to make my decisions when adjudicating what my players are trying to do. I’m okay with it when they want to stretch their imaginations. I want to encourage it. It’s fun. It’s “in character.” It makes the game what it is. Now I’m going to present another example that happened in a game. I won’t name the player that did this, but I want them to understand why it’s important that I’m calling it out here. (I was not the Game Master when this happened.)

Example #8: Fluffy

I want my character to have a white kitten as a pet, named Fluffy.

Answer: Uhm…. Why?

What happened: It was allowed, but late in the first session, a goblin wizard captured the kitten and teleported away. For the remainder of the adventure, the character asked every NPC if they’d seen the goblin wizard or the kitten, which they had not. Neither the wizard nor the cat were heard from ever again.

What should have happened: It shouldn’t have been allowed. Why not? The adventure was adapted from a published solitaire book and the Game Master wasn’t prepared for anything out of the box like this. What originally sounded like a cute character personality trait turned into a major distraction not just for him, but for the other players. The pet served no purpose mechanically. If it was a familiar or had some useful function, it would have been different, and it wouldn’t have been an issue. The point with this example is this: you can have Agency, but don’t abuse it or turn it into a distraction.

There are players who confuse Situational Agency and Narrative Agency. This is the case with the Immortal Alchemist example. Being able to choose your destiny doesn’t mean you should be permitted to stab random NPCs without consequences. Conversely, saying you want to hunt the mystical Golden Stag of the Lalszu Forest doesn’t necessarily define your destiny.

Part 4: The Amusement Park, Installing Railroad Tracks in Your Sandbox

I talked in Part 1 about three levels of Player Agency and how they relate. I talked in Part 2 about the differences between sandboxes and railroads. In Part 3, I made clear what isn’t railroading, even though some players think it is. Now I get to mash it together into a single, cohesive, comprehensive campaign setting that the players, at least most players, want to participate in, and me, the Game Master, wants to run. I call it the Amusement Park.

Neuith is neither a railroad or a sandbox. It is an Amusement Park. It’s a very cool amusement park with rides, attractions, shows, exhibits, a waterpark, a hedge maze, and numerous other features. I didn’t figure out this analogy until earlier this week. Matt Colville helped when he referred to going through a dungeon as a roller-coaster ride. Neuith didn’t start out this way. It evolved. It started out with a single adventure. The players met in a tavern. A dude walks in and says, “I’m looking for some adventures and I need some help.” Somewhere along the way, I made a terrible mistake. Big. Huge. I had no idea how big it was.

What happened? I’m glad you asked. If someone can learn from my mistakes, then I can say that I have succeeded in life. For what it’s worth. An NPC joined the party who was a bounty hunter. He carried a magic pin that allowed him to track his target. Once per day, he invoked the item by saying a person’s name as he dropped it. When it landed, it pointed in the direction of that person. If the person was dead, it would point straight down. After the NPC revealed his true nature to the party, they had no choice but to kill him and claim his stuff. Mistake #1: it didn’t require attunement. Mistake #2: while it only worked once per day, every character in the party could use it, once per day. Which meant it had five uses per day (five characters in the party).

At the end of the first mission, one of the rewards, other than a pile of gold, was a “dragon card.” Great idea. Three-Dragon Ante deck. Within each card is a trapped dragon via the Imprisonment spell. Cool, huh? Not only does each card confer a special, unique ability, but eventually, the dragons could be released from the cards. At this point, I hadn’t thought that far ahead. I’m still thinking pure sandbox. The characters started the next mission and met a character who had another dragon card, which I drew randomly from the deck. Later, they fought some monsters and found another card, which I drew randomly from the deck. I thought I had a great idea. I had no clue.

One of the players had a brilliant idea. “Is there a library?” Yes. “Are there any history books in the library?” Uhm, yes. I wasn’t sure where this line of questioning was headed. “Are there any books about dragons in the library, like one that lists their names?” Oh crap. They dropped the pin and said “Bahamut.” I said, I’m going to have to get back to you on that next week. Suddenly, sandbox just became something else. The party had three dragon cards. I had to figure out where the other 67 were, and since the group met weekly, I had 7 days. In a sense, this is a case where Situational Agency dictated Narrative Agency.

In 7 days, Neuith transformed from a sandbox to an amusement park. I started out with a blank slate full of possibilities. In less than a week, it became a world with over sixty encounters complete with locations, characters, monsters, and stories, all tied together with a common theme. For each dragon card I had to determine where it was, who had it and how they got it. I had to figure out what its property was. I also went on to determine how the dragons were trapped in the cards, who trapped them and why, and how they can be released. This led to identifying what each individual dragon’s name was and its personality, and how it would react if it were freed.

This process sparked the idea for the novel I’m working on. The revision, which I still need to get posted, is up to 19 chapters.

Presently, there are 2 groups playing in my setting – Group 3 and Group 4. Groups 1 and 2 are finished. When Group 3 is finished, I’ll start Group 5. I’ll make a few tweaks and fixes. Group 4 is one session away from completing Chapter 4. What follows is a metaphorical, spoiler-free description of Neuith. I’m describing it in terms of an amusement park so that you can draw your own conclusion on how it differs both from a pure sandbox and avoids railroading, yet shares aspects of both.

Chapter 1

Let’s go on vacation! You went online and bought tickets for you and your family. It’s a long drive that will take two days to get there. You packed the car, but just as you’re getting ready to drive away, the street is blocked because the neighbor’s house is on fire. You wait a couple hours. The neighbors are okay. It was just a minor kitchen fire; some smoke damage and they’ll get a new stove. You give them the keys to your house and they can stay there while you’re gone.

You get on the freeway and the first day out everything is great. You’re making good time. You stop for lunch at a small diner and meet a nice old lady. She tells you a sad story about how her husband and son died, and would you please deliver a message to her other son in a town on the way. Why she can’t use email or send a letter, you don’t know, but you’re happy to oblige.

That evening you stop at a motel. While the kids are swimming in the pool, your daughter runs in crying because there’s a bully calling her names.

Day 2 is uneventful. Late afternoon, you get to the town the old lady told you about. Her son’s service station is right there by the freeway and you give him the message. He thanks you profusely. He fills your tank, checks your oil and tops off your window washer fluid for free. Says if you need anything when you’re in town, just let him know. You check in at the motel and so far, everything is going great.

After driving two hours, you have a flat tire. You debate whether to go back to the town or go up the road to the next town. While you’re trying to decide what to do, the lady’s son just happens to be driving by in his truck and changes your tire for you. No charge. He gives you a coupon for a discount at the tire store in the next town up the road.

It’s midmorning on the third day when you finally arrive at the park. You see the signs and you turn off the freeway. Amusement park traffic turn left next exit. The traffic is terrible and you get stuck in the wrong lane. Your spouse is yelling at you and the kids are screaming in the back seat. They can see the rides and the lights in the distance. You’re wondering if this was all worth the hassle. Going to the beach would have been much easier. You finally get into the parking lot and it’s a hot mess. Teenaged parking attendants are pointing you to go left then right, directing you to a parking spot that’s at least a mile from the gate. Fortunately, there’s a shuttlebus to take you from your car to the gate or hotel. You’re not sure if you want to check in to the hotel first. Maybe your spouse can take the kids through the line and you can meet up with them later. The line. There’s a least a thousand people in the queue. You have your tickets in your hand. Your spouse will take care of the hotel room and catch up to you in line. Hand over one ticket just in case.

And finally, you’re through the gate. You’re in.

Chapter 2

Before you can enter the park, they make you sit through a movie about the park. It’s quite the production, done full Hollywood-style with voices, animation, computer graphics and all the characters. The voice-over narrator is a famous actor. You’re introduced to all the characters and given a virtual tour of the park. You learn that some rides aren’t for the little ones. You must be at least 42 inches tall to go on the Kraken, the Leviathan, or the Dragon Turtle, for example.

From there, you’re let out into the main lobby where there’s a gift shop, a restaurant, an ice cream shop, a candy shop, and a photo booth. The kids immediately want ice cream and candy, and want their pictures taken with the griffons. It’s a bit early, but it’s a good time to stop for lunch anyway. Service is slow, but the food is okay. Nothing like a soda and sugar rush to get the afternoon started.

Chapter 3

Your daughter realized she left her stuffed dolly in the car. Wait here. You get your wrist stamped and head into the parking lot. You can either wait for the shuttle or walk. The next shuttle isn’t for another 30 minutes, so you walk. You get to the car, but the doll isn’t there. You call your spouse and you realize the doll is in the hotel room. You flag down the shuttle and go there. You don’t have the keycard to your room and it takes them 30 minutes to verify your Id with the manager to get you one. While waiting, the kids rode the carousel three times. They learned that each horse has its own name.

Chapter 4

You’re going on our first real ride. It’s the called the Dragon House. You ride a little ship out to an island on a lake and follow a path. As you go through the twists and turns, you’re told a story about dragons and there’s a freaky twist at the very end. You learn that there are dragons hidden throughout the entire park, and there’s a special prize for the guest who finds the most each day. You install an app on your cell phone with the list of the dragons and you must scan the QR Code when you find one. If you find all 70, you get a coupon for free one-day admission next year. “I’ll bet no one finds all 70.” Well, the first three are already marked, at least.

Chapters 5-??

From here, you decide where to go next. There are at least 50 rides, shows, and attractions throughout the park. You choose what adventures and what story goes into these chapters. As you progress through the park, you’re building not only your own story, but you are collaborating with my story. I’m not forcing you into my story; we’re building our story together.

Last Chapter

When the sun has set, and you’ve done everything you want to do, there’s a huge fireworks display. Sit back and relax and enjoy the program. Listen to the music, see the great explosions and the sparkling lights. But wait. It turns out that you are the guest that found the most dragon cards! Maybe not, but something else about you has garnered the attention of some important people. You’re brought up on center stage and…

Remember when I mentioned that some rides you can’t go on until you’re tall enough? In game terms it means you must be at least a certain level. Going against the Kraken at Level 6 is going to get you killed. Level 10? Even fight. Level 14? You can handle it. You will know where the Kraken lives, so you get to choose when to face it. Same for the other high-level monsters. This is part of the sandbox aspect of Neuith.

But there is a railroad aspect as well. The first 4 chapters seem very railroad-like. It’s intentional. As the Game Master and architect, I need you to understand that this setting is deep and involved. There are a lot of stories and plots and intrigues and characters and issues. Not everything can be solved with a skill roll. You will be required to make some hard, tough choices. Railroad with Narrative Agency. These 4 chapters lay the foundation to allow the Narrative Agency in the rest of the campaign.

Wait, there’s more! I feel like a salesman on one of those late-night commercials. If you buy right now, you get two more and these other things thrown in for free, all for just one low price plus shipping and handling. First, there’s a simple rule: once you’ve taken a ride, you can’t take the ride again. Second, the amusement park is under inspection. An inspector is going through each ride one at a time. When she visits a ride, the ride is shut down, making it unavailable for guests to use. This means that if there’s a dragon at that ride, it can’t be found by you.

At some point, all the rides will be either shut down by the inspector or visited by the guests (the players). When this happens, the first of the end-game events begins. This isn’t the end of the campaign. Think of it like this: The adventurers must first meet up with the inspector. Deal with her. Then they must meet with the corrupt city officials who hired the inspector. Deal with them. Then, finally, they must deal with the greedy real estate developer who wants to tear down the entire park and turn it into a horse racing track and casino.

Why the “inspector”? Again, this isn’t a pure sandbox. It’s one thing for you to wander around the park choosing rides and attractions at your leisure. What if someone else was doing the same thing? Specifically, one of the setting villains? As you progress, you gain power, as you should. They do too, though their rules are slightly different. Technically, they start out stronger than you, so you don’t go hunting them and killing them too early in the story, but eventually, you will grow to equal or stronger than them. Why? This gives you a sense of urgency and purpose. Yes, you have your own destiny to follow. You want to fulfill your dreams and desires. You get to do that while you deal with the great evil forces that are bent on world domination.

The metaphor isn’t perfect and is a bit misleading. Intentionally.


Can I have my cake and eat it, too? I believe I can. I can build some railroad tracks in my sandbox and turn it into an amusement park. I can appease the players that want to write extensive backstories and craft their own epic stories within the framework of a setting that also has a lot of history and depth. I can give players the Agency they want and desire while still telling my story. It is a pivotal time in Neuith. A group of brave adventurers have answered the call to help the heir of a city reclaim his role as its lord, only to discover they are getting pulled into a greater epic tale involving powerful witches, fierce dragons, deadly assassins, intrigue and betrayal, and one of the greatest mysteries ever to befall the historians of Neuith, the Paradox of Desire. Does your story fit into this epic saga? Yes. Yes it does.

Matt Colville has an entire series of videos about running RPG games. This video compares sandboxes with railroads using a very familiar setting to make his point. It’s 30 minutes well spent.   (,




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s