Order In The Sandbox
Today, I get to focus on one of my favorite subjects, campaign structure! There are lots of choices out there, but today I’m going to focus on one that I don’t use all that often, the improv sandbox.
This article was prompted by something I saw on the screenwriter’s blog that I linked to an article or two back, The Story Guy Newsletter by Scott McConnell (Click Here to read his posts online and subscribe if you find them worth your time).
The subject this time was how to untangle a story that has become too convoluted. That’s not usually a problem in an RPG because we don’t have to tie up every plot thread in the one adventure; we can leave something dangling to be picked up in a later adventure.
It’s at the campaign level that things can become messier over time, especially if you do a lot of improv in a sandboxed campaign – something I was reminded of by a Reddit thread.
The two thoughts connected, and I quickly devised the organization methodology that I’m going to focus on in this article, but the more I looked at it, the more I realized that it’s utility went beyond just organizing your way out of a plot quagmire; if you employed it from the get-go, it would prevent you ever getting into that sort of tangle in the first place.
Long-time readers will recognize the system as a simplified version of the structure that I use most frequently (see, for example, Back To Basics: Campaign Structures), though that involves a lot more pre-planning.
1 The Sandbox
The concept of the RPG Sandbox is simply a restriction on the scope of the campaign to a small subset of the entire game world; content beyond the sandbox may be hinted at, but the campaign overall is self-contained within the scope of the sandbox.
The chief benefit of sandboxed campaigns is limiting the scope of required prep. Other than rare and isolated exceptions, no campaign adventures take place outside the sandbox, so no development of them is necessary.
1.1 Improv in The Sandbox
A significant number of GMs who sandbox their campaigns also like to improv the adventure content as much as possible, responding directly to player agency – at most, they do some prep detailing the locations and events that the players indicated they were heading into in the previous day’s play.
The GM may not have a choice. The shorter the interval between game sessions, the stronger is this inclination, simply because there is greater restriction on the amount of prep time available. Playing on two or more consecutive days – what I used to call an “RPG Marathon” – is the most extreme example. Playing once a week is more typical. Playing once a fortnight allows greater scope for prep; playing once a month or less frequently allows a lot of scope for prep.
It’s more normal for the prep requirements to dictate how often you can play, but this isn’t always possible.
At one point, I was living many hundreds of miles from my gaming group. As readers can imagine, this was quite a handicap to regular play. Once every 3 months or so, I was able to save enough money to travel to the group. But every week, I would prep enough game material for a full day’s play, so that I had 12 game sessions’ worth when I reached the city.
My players, kept informed of when I would next be available, would use RDOs and a day or two of annual leave to add to the playing schedule. I also made maximum possible use of Public Holidays. With luck, 5-6 days of game play would be possible, with 2-3 game sessions per day (3-4 hrs in length); only a short break was taken between one adventure and the next.
In effect, we were still playing to a weekly schedule; it was simply compacted in time to a couple of days every three months. To make this practicable, there were often several simultaneous plot threads, so that each could be the spotlight for an adventure or two without too much cross-connection between them. Even so, of necessity, the later adventures in a marathon were a lot more loosely defined than the early ones.
This is an example of game play dictating the time available for prep, with attendant consequences. Amongst other techniques, later adventures would be contained within ‘temporary sandboxes’, deliberately confining their scope to a sufficient extent that prep could be completed in 30-60 minutes, which I often did while everyone (including me) ate. But even so, there was a lot more improv later in the marathon than later.
It wasn’t the ideal situation, but it worked.
1.2 Plots entangling
You can only sustain improv over multiple game sessions before your loose plot threads start to become entangled. This is often a consequence of forgetting the significance of past improv’d events – you may have had a clear idea of where things were going at the time, but that clear idea has vanished like dew on a summer’s morning.
No problem, you just improv a new significance to the past events, creating the day’s adventure, advancing that plot thread more-or-less randomly, and perhaps connecting it to some other plot thread that it was never intended to intersect. That’s enough to get you through that day’s play, but by the time you’ve thrown in a plot twist or two along the way, plots are beginning to grow entangled, and often hampered by mutual contradictions if you look too closely.
So you improv an explanation for the contradictions, adding a new layer of confusion to the whole ball of wax, and cementing the plot tangles into place. Two relatively straightforward plotlines can become so convoluted and complicated that they bring the entire campaign to its knees.
Remember, too, that the GM has the advantage of having had a clear idea of where things were going, at least at one point; if the resulting morass is too complicated for him to figure out, what hope do the players, bereft of that advantage, have?
So that’s the worst-case outcome that we are trying to prevent if not evade completely.
2 The Plot Map
The solution that I have devised uses three tools. The first of these, I have named “The Plot Map”. It’s a simple page of square-grid graph paper – preferably 1/4-inch or 5mm separation of squares. Especially large or detailed campaigns may need multiple pages.
2.1 The Vertical Axis
Along the left-hand side, using as little space as possible but aiming for maximum clarity, list every important NPC and organization within the campaign that the PCs have interacted with (provided they still exist). Some larger organizations may be listed as a generic entry or as individual branches, that’s up to you to decide once you understand how the whole package works.
Where possible, it’s best to list these in sequential order of appearance within the campaign, or as close to it as you can get.
The list should include any NPC that has only appeared once, if the GM thinks there’s a reasonable likelihood that they might turn up again in a future adventure. In particular, anyone who has resources that the PCs can draw on, or who is an active enemy of the PCs (or vice-versa) should be listed. NPCs who don’t meet those criteria aren’t generally as important.
2.2 The Horizontal Axis
Vertical divisions along the horizontal axis are adventures. A dot gets placed on the map each time that NPC appears in an adventure, and an index number (I’ll get to those shortly) gets written next to each dot.
2.3 Connecting The Dots
The dots are then connected by a horizontal line.
Where there has been a revelation of a relationship between two entities, or a new relationship established in the course of an adventure, a colored curved line may be used to connect the two – that’s up to the GM. For example, there may be one line for the Chancellor Of The Exchequer and another for a Snake-worshiping Cult; if the PCs have just discovered that the Chancellor is a member of the cult, it’s reasonable to connect the two. The same is true if the Chancellor was (instead) persuaded by the PCs in a previous adventure that the Snake-Cult was a significant threat to the Kingdom, or if he simply became opposed to them for some reason.
If you choose to forego the vertical connections, you can fit the entire campaign prior to the last adventure on the first vertical division after the names; if not, you may need more just to keep the connections clear.
Depending on the neatness of your handwriting, you may be able to space adventures one division apart, but it would be more normal to need two.
Other possible refinements include color-coding for the role played by the NPC / Group in the adventure, but this can rapidly become complicated if the PCs had one perception of that role and the NPC another. I generally wouldn’t do so, myself – it’s one complication too many.
You’ll get a clearer idea of the structure from an illustration later, when I discuss how it is used.
3 The Plot Concordance
The second element of the solution is a full-page-width table in a spreadsheet or word processor. I don’t know of any decent one that isn’t a plain-text editor only that can’t handle a table being placed on a page.
In the past, I’ve used Word, StarOffice, and currently use LibreOffice for this purpose. There are others – I don’t know for a fact that they support this function, but it’s such a ubiquitous part of word processing that I would expect this to be near-universal.
3.1 The Horizontal Axis
Across the top, you list first the whole group, then each of the PCs, and thirdly any significant pairings of PCs, one to a column.
Trusting that the meaning of the first two of these is clear, let’s talk about what I mean by a “significant pairing”. If two PCs have a mutual enemy, or some other relationship that means they would share the spotlight in some adventures related to the connection, that’s a ‘significant pairing’. There might even be three-way connections, depending on the campaign and the characters.
For example, in one campaign, three of the PCs decided to form a conspiracy aimed at overthrowing a local Bishop who was blatantly corrupt and despicable. They deliberately left the 4th PC – a cleric of the same faith as their target – out of the conspiracy.
3.2 The Vertical Axis
Subsequent cells in the row will contain plot-thread titles, plot index numbers, and an empty cell for spacing, respectively.
While that’s the minimum structure, I’d actually go further. I wouldn’t let any cell grow more than two lines line before shifting to a new cell, so that the spacing remains fairly tight.
I’d also start a new cell when another plot thread ‘links up’ with this one. So the number of cells with content would vary from one ‘plot heading’ to the next.
Again, you’ll get a clearer idea of what I’m talking about when you see an example being used later in the article.
4 The Master Plot Index
That’s twice that I’ve referred to “index numbers”, so readers may be wondering where those numbers come from. The answer is the Master Plot Index, which is a series of simple text lists.
4.1 Plot Developments
The first line of any such text list is the name given to that plot thread – this should match the one listed in the Plot Concordance. Each entry in the resulting list consists of two items:
1. An index number (starts at 1 and ends at whatever);
2. a minimalist summary of a plot development, which lists any significantly-affected / involved PCs and NPCs. Ideally, you would get this down to a single line of text, more commonly, you’ll need two or even three. It’s most important to list unresolved plot developments, but the more comprehensive you can be, the better.
4.2 Plot Endpoints
Beneath these summaries, there’s a separate entry that is distinguished from the others somehow. This contains the intended resolution of the plotline, again in severely-minimalist form.
I would probably use a plain-text editor for this, because not all word processors permit multiple documents to be open and viewable at once, and some of those that do can take a significant amount of time to switch from one page to another. I use Kate for the purpose because it does let me view multiple documents at once (useful for some purposes) and is a lot faster than recent versions of Wordpad, which used to be my go-to. But even Kate can be more word-processor than you need for this purpose.
What you end up with is an ongoing list of the significant campaign events, grouped by plot thread, and a map of which plots advance in each adventure, connected to each other by the concordance and the index numbers.
5 An Existing Campaign
Things are a lot simpler if you start using these tools from the very beginning of a campaign, but the original intent was to use them to wrestle some control over an entangled existing campaign, and that will take more work. So that’s the place to start. There are three primary processes to apply, most with multiple simple steps. I’ll try to make it as easy as possible.
5.1 Populate the Master Plot Index
5.1.1 Unsorted
The first “Plot Thread” that gets listed is called “Unsorted”. Leave blank space for entries below it.
5.1.2 Plot Thread Titles
Think about the plotlines that you have underway. Give each one an appropriate name. Leave black space for entries under each. try to keep them as sequential as possible, so the first adventure, then the second, and so on. Note that an individual adventure may touch on multiple plot threads, especially early in the campaign when you’re setting ‘campaign pieces’ in place.
5.1.3 Populate ‘Unsorted’
List every plot development or key fact that you need to know to run the campaign in bullet-point form as described earlier, up to the beginning of the most recent game session. Be as exhaustive as possible. Don’t worry about numbering them yet. Continue until you can’t think of any more. Don’t be concerned if you know you’re forgetting some.
5.1.4 Migrate ‘Unsorted” into Plot Threads
Starting with the top item, cut it from the “unsorted’ list and paste it under the appropriate plot thread heading. Repeat until either they are all done, or one of 5.1.5 or 5.1.6 needs to happen for progress to continue.
5.1.5 Generating Additional Plot Threads
This won’t always be necessary, but it often is, especially if there are plot threads that have been carried through to a conclusion. If you need to, simply insert a new heading at the appropriate spot.
5.1.6 Adding additional ‘Unsorted’ items
Don’t be surprised to find your recall being stimulated as your organize things. Add new items as they come to mind.
5.1.7 When ‘Unsorted’ is empty, rationalize the plots
Time to get creative. In your own mind – no need to put anything in writing if you don’t want to – make sense of each of the plotline as an isolated story outline – beginning, middle, and end. What doesn’t make sense, and how do you have to change things so that this problem is resolved?
Try hard not to make plotlines too similar. If two plotlines are essentially the same, look for a way to (eventually) link the two so that they eventually become one plot thread. This creates a future event, Merges with [plot thread title], and a matching entry under the heading [plot thread title].
List any events/revelations that have not yet happened/come to light below the existing items in each plotline. You will need some way to distinguish these from the ones from the past. This can be as simple as putting a % or > sign in front of the text. Or maybe putting the (eventual) list entry numbers in brackets.
Note down any plot twists that you want to use in the plotline – again, avoid using the same plot twist in multiple entries. It’s better to forego a plot twist than to program the same one for multiple plotlines – unless you can come up with a single plot twist event that will affect multiple plotlines, of course!
If you want to continue with the ad-hoc approach, it’s best to be as general as possible.
5.1.8 Enter Plot Resolutions
Now that you have decided how each plotline is to end, at least in theory, add this information in the space reserved for resolving the plot. This outcome is not locked in stone; you can change anything or everything as plots proceed to keep them interesting to the players. This is a general guideline of what you (perhaps subconsciously) intended all along.
5.1.9 Number the list entries
The last step is to start at the top of the first plot thread and number each of the entries. Don’t restart the numbers when you cross down to a new plot thread.
One possible refinement is to number anything from the initial campaign background / briefing as it was given to the players “0”. This helps distinguish implied promises deriving from the initial briefing so that you can make sure the campaign (eventually) ticks all those boxes. Anything with a non-zero number is therefore something that happened or came out in play.
You may not have realized it, but you’ve just unentangled your plotlines! The rest of the process dos two things: helps to keep your plots untangled in the future, and turns the work into an awesome planning tool that will actually enhance your improv plotlines.
5.2 Populate the Plot Concordance
The Plot Concordance is mostly just a reorganization of the information in the Master Plot Index. As such, this is a relatively simple process.
5.2.1 Allocate and Enter Plot Threads
Starting with the first plotline on your list, determine which PC(s) have the most direct interest in this plotline, the most engagement (it might be the whole group).
Find the matching names in the column headings, and list the heading there. Append the PCs names to the name given the plotline in the Master Index.
Repeat for all plot groups.
5.2.2 Muse over Empty columns
Do you have any empty columns?
An empty ‘all-hands’ or ‘whole-group’ column (however you’ve labeled it) means that the group of PCs will lack cohesion and may drift apart as the campaign proceeds. That’s obviously not good, but it’s better than having a half-baked idea in this particular space. So don’t do anything about it now, just note it down mentally as something to fix when opportunity and ideas strike.
An empty ‘single PC’ column means that character, and their player, are not getting their fair share of the spotlight. This problem is arguably both more immediate and more important than an empty whole-group column. Again, not something to fix right now, but something to intend doing something about ASAP.
An empty ‘PC group’ means that the PCs in question have something in common that no adventure has yet drawn attention to. This is not a critical problem, provided that every individual is getting their fair share of the spotlight. It’s entries in these columns that are more important than the lack of them, because they add to the cumulative spotlight-share that each character is getting.
Don’t be surprised to discover that one particular PC is all over your Plot Concordance like a rash – it often happens that one particular character becomes the linchpin or focal point of a campaign, simply because they are more engaged with gameplay than others. So long as they are reliable attendees, this is not necessarily a problem. Instability in their attendance record signposts a potential campaign problem that you need to think about, though.
5.2.3 Extract Past Plot Element Numbers and post them
Because we already have a sequential list breaking down the plot, consisting of numbered items, we don’t need another one – the list of item numbers is enough. Put them on a single line, separated by commas and spaces, until there are so many that they stop fitting, then continue on another line.
5.2.4 Extract Future Plot Element Numbers and post them
We may have untangled the plotlines in your head and on paper, but it won’t happen in the game until those necessary plot developments that you wrote down actually take place or get discovered (depending on their nature). Since they also have numbers, and some sort of a flag to indicate that they are future events in the plotline, it’s easy to list them (with the flag) on the next line. When they actually happen in-game, all you need do is remove them from one list and append them to the other.
And that’s all there is to it.
5.3 Populate the Plot Map
It’s likely that there will be more empty holes in your Plot Map than anywhere else. That’s because an NPC can be a featured character in more than one plotline at the same time, simply by virtue of having some sort of relationship with the PCs.
That’s fine – just do the best you can.
5.3.1 Initial Dots
This is actually blindingly easy in terms of the initial population – because every NPC / Organization on your list has presumably already shown up in the campaign at some point, they all get a dot in the first column or columns if you needed more than one.
5.3.2 Numbering The Map
This, on the other hand, could be incredibly messy if someone has shown up a lot. You only have a limited amount of space to use – enough for one, or at most, two index numbers.
There are two ways to solve this – one that’s easier, one that’s more useful in the long run. My preference and recommendation is to choose the second, but that’s up to you.
The easier one is to list the first number in each plotline listed in the Concordance. This makes it easier to locate the matching concordance entry, which then gives you everything you need to get plotline information from the index.
The more useful one is to list the most recent in-play number. This makes it a little harder to find the matching concordance number, but because the numbering starts more or less sequentially within each plotline, and will become more strongly and accurately so over time, the lower the number, the longer it has been since the NPC / Organization last appeared in the campaign.
So make your choice and annotate the logical map accordingly.
5.4 Bonus Technique: Annotate a Geopolitical Map
If you have a hardcopy of your game map – the type of map that shows terrain, landmarks, population centers, and so on – you can annotate the locations where events occurred with the corresponding event number. This is useful because every time the PCs choose to go to or through a specific location, you can identify events that occurred there in the past, and use that information to help you come up with new events within a given plotline.
Equally, if you decide to add a development in a given plotline, the map shows you where those developments should logically take place. Either way, the information helps you define and fill out the improvised plot development, while keeping the geography of events consistent.
6 Ongoing Campaign Tracking
As the campaign continues to unfold from this starting point, after each game session, you have to update the three tools with the latest plot developments. This might be taking an item from the future developments list or it might be independent of those. It doesn’t have to be pre-planned, it may have been entirely spontaneous on your part, or a decision by the players that resulted in the event. Either way, the process is the same:
- Add the event to the plotline in the Master Index;
- Number it;
- Add the number to the plotline in the Concordance;
- Add a tracking point to each NPC who had a significant role in the day’s play on the Map
- Add the number to that tracking point and draw a line connecting the character’s appearance tracking point to their previous entry.
That’s as painless a record-keeping process as I could make it.
Readers may have noticed that I earlier specified that the most recent game session should be left out of the initial compilation. That’s so that you can get a head start with the Campaign Tracking by making the most recent session the first one indexed, as the start of an ongoing process.
This is an example of the map once this first round of information is entered. I find that, having done one, the process becomes pretty self-explanatory as soon as you look at the document. Because I had to shrink it to fit the column space at Campaign Mastery, I’ve included a closeup of the first few entries.
Notice that this person is even more organized than most, distinguishing between NPCs and Villains in their character list. They’ve also had a go at indicating relationships with the vertical bars, but after three of them, they’ve given up on that.
This is the matching Concordance. I’ve simply invented some characters and plotlines out of whole cloth for illustrative purposes, I have no idea what any of them are!…
And this is the Master index that matches. Because I don’t know the content of the supposed “plotlines”, I’ve just greeked some random text to illustrate the results of the process.
7 Ongoing Inspiration, directed plotting
I promised earlier that the results would not only help keep your plotlines untangled, they would make improv plotlines better / easier. This section is all about how you use these three tools in that way.
7.1 The Plot Map
We start with the plot map. Here’s a simulated example of one that’s been in use for a while. In fact, it’s the earlier map with 5 additional game sessions added.
The day before play is the time to start putting ideas together in your head for what might happen the next day, if you haven’t done so already. You don’t have to make any decisions yet, this is simply trawling for ideas to have on standby.
Let me start by drawing your attention to the string of hollow circles and lightly filled circles on the right hand side. The hollow circles identify characters who have yet to feature since the map was started. The filled ones show characters that haven’t featured for a while. None of which means that you have to use them in your next game session; this is just a reminder of their existence and a prompt to spark your imagination.
7.2 The Master Plot Index
Next, look at the Master Plot Index and remind yourself of both the resolution that each plotline is supposed to be progressing towards, and any clarification events that have been listed to occur (eventually).
Again, you don’t have to use any of this in your next session; it’s helpful just having them in mind when you engage your creativity.
7.3 The Plot Concordance
The concordance supplies a third source of information. Look at the last number in each panel; the highest one shows the plotline that was most recently progressed, the next the second most recent, and so on. Look for the lowest of those numbers – in the example, that’s a “7” and it’s part of the plotline named “Nagatele” (which presumably is everything the GM in question needs to know). This tells you what plotline you want to consider prioritizing – but skip it until you have a reasonably good idea.
The whole point of these tools is to spark and focus your creativity into the spaces that have been unchanged for the longest period. Even if the only practical impact is that you try to think of ideas that plug one or more of these holes before you go looking for other plot ideas, that in itself is a massive increase in campaign cohesion.
7.4 Wandering Twists
So the dice are indicating a wandering monster or other random encounter. A quick glance at the three tools can suggest a way to advance one of the plots (or just have it be relevant to the encounter) – suddenly, that random encounter has Significance. The tool to use for the purpose is the Concordance, because it lets you see a high-level view of the whole campaign.
And if no ideas come to you, you can always run it as a typical random encounter. The only difference is that you have given yourself the opportunity to make the encounter more important.
8 Structured Unstructure
These tools are not about restricting or preventing you from your chosen plotting method of improv deriving from what the players decide to focus on in the current game session. You can pre-plan or improvise exactly as much as you always do. That’s what makes this methodology different to the other plot structures and schemas that have been offered on Campaign Mastery in the past.
9 Afterwords
I had a couple of thoughts that I didn’t find a way to incorporate into the main text of the article. It’s possible that these aren’t anything new to you, but the only way to find that out is to throw them out into the public and see what happens – so I have tacked on this final segment of the article specifically to house them.
9.1 Playing In The Sandbox
Players are sometimes like children, they love to push beyond any boundaries that have been imposed. The sandbox itself is just such a boundary. The desire to find out what lies beyond the marked lines on the map is a fairly universal drive, so this is not all that surprising.
It can be extremely convenient to emplace something on the fringes of your sandbox to turn the curious back. This could be a natural barrier, or a hostile race-in-residence, or any number of other options; the nature doesn’t matter, it’s subordinated to the game purpose of the inclusion.
The best time to emplace these ‘cushions’ is when you’re first setting up your sandbox, but the only time when it’s too late is when the players are already exploring the sandbox fringes – and sometimes, it’s not even too late then; you can use the ‘Nested Sandboxes’ idea, below, to emplace the barrier just beyond the sandbox.
Of course, the usual mechanism for ‘enforcing’ the Sandbox is simply never setting anything interesting, in terms of adventure, beyond the barrier. Sandboxes are at their most artful when they are completely invisible to those subject to them.
9.2 Sandbox Nesting
Another way of thinking about a sandbox is for the boundaries to be reminiscent of a snake’s skin – every now and then, the old one gets shed, enabling the new one to grow beyond the limits to which the old one constrained the wearer.
In other words, the campaign isn’t just set in one sandbox, it is contained within a series of them, each nested within the next, with the transition from one to another completely under the GM’s control.
I used this technique in the original Fumanor campaign and its sequels, expanding the PCs horizons east, south, east, east again, south, north, and then – in three stages – west. There was never anything overt keeping the players inside those sandboxes; it was just a matter of there being nothing of immediate interest outside the current sandbox.
Ultimately, Sandboxes are about focus. So long as the settings for adventures are kept within the Sandbox and there’s nothing accessible beyond, they work. Plot elements can originate from beyond the edge, signaling that there is more beyond, but their impact is always felt within the confines that the GM has set, a restriction designed to give him as much time as necessary to develop what lies beyond in playable depth. It’s like writing on the edges of the map, “Here there be Dragons”!
Discover more from Campaign Mastery
Subscribe to get the latest posts sent to your email.
June 29th, 2024 at 8:57 am
[…] Campaign Mastery writes up Order in the Sandbox. […]
November 15th, 2024 at 3:07 pm
This article on “Order in the Sandbox” does a fantastic job exploring how structure can bring out creativity, even in seemingly chaotic places! I loved the perspective that adding just a bit of order to a free space, like a sandbox, can lead to even more inspired and collaborative play. It makes me wonder, do you think there’s a point where too much structure might actually limit creativity?
Sarah J. Goin recently posted..Car Parking Eski Sürüm APK Android için [Son 8]
November 18th, 2024 at 1:27 am
Absolutely, Sarah. With a structured process of any kind, following it by rote numbs the mind and stifles creativity. If there is just enough structure that it serves your needs and those of the plot you are working on, that permits sufficient flexibility to perceive and implement alternative pathways through the narrative and a variety of outcomes; too much can be too confining, spending all your time and thought on the process itself and not on the desired result of the process.