I like handouts.

I like them a lot.

If you’ve seen the Alexandrian remixes for Eternal Lies or Dragon Heist, then this won’t come as a surprise to you. Those campaigns are fairly representative of what my games look like: There will be dozens or even hundreds of handouts. Photographs, letters, lore books, artifacts — anything I can get into the players hands enhances the experience and becomes a tangible touchstone for what’s happening in the game.

In running a single adventure — like Left Hand of Mythos — keeping track of the props is pretty straightforward. Over the course of a campaign, though? Things can get more complicated.

It’s not at all unusual for my players to pull a sheet of paper out of their notes and say something like, “Hey, this letter from Lady Scarlet to Thornai that we got thirty-seven sessions ago — can we pull that from the evidence bag and dust it for fingerprints?”

Now, somewhere in my notes is likely the information I need to answer that question (i.e., who handed this letter before the PCs snagged it). But where, exactly, is it?

Well, almost certainly keyed to whatever location they found the letter in.

… and where is that?

Damned if I know. It was, after all, thirty-seven sessions ago.

What I’ve learned to do is put tracker IDs on my paper handouts. That way I know where to look up my notes about them later on.

I use alphanumeric codes for my scenario notes. For example, scenarios in my In the Shadow of the Spire campaign include:

• BW03B Alchestrin’s Tomb
• BW06 Chapel of St. Thessina
• CC07 Porphyry House of Horrors
• NOD2 The Secret Meeting

And my current Night’s Black Agents campaign has scenarios like:

• CS03 Paymaster
• CS06 Dragovir Monastery
• PP01 Arkady Shevlenko
• SJ02 Serbian Mafia

These are just generally useful for keeping stuff organized, but are particularly useful for tracking props. In pencil, I lightly write the alphanumeric code somewhere on the prop. No matter how much later the prop surfaces, I can just reference the code and know exactly what notes I need to reference.

If you don’t want to use scenario codes, another option might be maintaining a master handouts index, listing every prop you’ve prepped and the adventure it comes from. The tip here isn’t the specific method of the tracking; the tip is that having some way of knowing the provenance of your props becomes important as you begin running rich, complicated, long-lasting campaigns.

If you’re using node-based design, does that mean you’re prepping a plot?

No.

We’re talking about “plot” in the sense of Don’t Prep Plots:

Don’t prep plots, prep situations.

Plot, in this case, means the sequence of events that happens in a story. Prepping a plot in an RPG means you’re predetermining what the PCs will do: A will happen, then B will happen, then C will happen. If by “plot” you mean something else — a villain’s scheme, a ground plan, etc. — then the answer might veer closer to “maybe,” but it’s also outside the scope of this discussion.

If you’re familiar with node-based design, then it’s likely you’re scratching your head right now: Obviously node-based design isn’t about prepping a plot, so why is this even a question?

But it’s actually a question I get asked several times a year. This is sometimes because people are using some other definition of “plot,” but, based on conversations that I’ve had, it’s frequently that they’re so deeply entrenched in plot-based prep (including railroading) that they have difficulty comprehending any other paradigm. Even when they look at alternatives, they subconsciously think to themselves, basically, “Well, obviously I would use this to prep a plot.” And then sometimes they go farther and say, “Why is this guy lying when he says he doesn’t prep plots?” Strangely, this even seems to happen with people who are virulently opposed to prepping plots.

This isn’t limited to node-based design, either. I’ve seen the same attitude applied to everything from clocks to hexcrawls to faction turns. No matter what the structure is, the GMs trapped in this way of thinking start by trying to guess what their players will do and/or figuring what they want to force their players to do, and only then do they try to figure out how the structure can help them do that.

This, of course, is really unfortunate. It’s a massive blindspot. And I’ve seen this enough — and been asked this enough — that I think it’s worth taking the time to take a closer look at these misapprehensions.

THE MANDATED MYSTERY

The first argument I often see is that:

1. Node-based scenario design is used to design mysteries (e.g., figuring out who’s selling red opium).
2. A mystery scenario means that the GM is dictating the scenario concept to the players. (“Thou shalt figure out who’s selling red opium.”)
3. This is a plot.
4. Therefore, using node-based scenario design means that you’re prepping a plot, not a situation.

To start, let’s accept as a given that node-based scenario design means that you’re designing a mystery. (I’d actually quibble with that a bit, but it’s not important here.)

Next, I think we need to define what a sandbox campaign is: This is a campaign where the players can either choose or define what the next scenario is going to be. The second premise being asserted here is that if the GM is the one assigning scenario concepts (as they might in an episodic campaign where the PCs are cops being assigned cases to solve, for example), then this is not a sandbox campaign.

This is, of course, true. I actually describe this as the “lightest form of railroading” in Part 3 of The Railroading Manifesto. (Which is not to say that there’s anything inherently wrong with this sort of campaign structure. Quite the opposite. There all kinds of diegetic and non-diegetic reasons for running an episodic campaign and, as I point out in the manifesto, lots of people wouldn’t consider it to be railroading at all.)

But is this, in fact, a plot? Is the GM prepping a predetermined sequence of events? Well, if you squint hard enough and are sufficiently liberal with your definition of “sequence of events,” you can make a case for this being true. (For reasons pretty similar to why I refer to it as “the lightest form of railroading.”) Personally, I think what you’re actually discussing here is campaign structure rather than the scenario structure, and I think putting this much weight on the nature and presentation of the scenario hook is more deceptive than revealing when it comes to the overall design of the scenario, but there’s certainly a semantics debate to be had.

Ultimately, though, none of that is really relevant, because nothing about mystery scenarios or node-based design requires the GM to dictate scenario concepts to the PCs.

If you’ve never experienced a player saying, “I want to do X,” and then the GM designs the scenario that results from them wanting to do X, then you might find this confusing. But I do this all the time.

The type of scenario you design, of course, will depend on what X is and how the PCs are planning to do it: If they want to steal the Ruby of Omarrat, then I’m probably prepping a heist. If they want to travel from Neverwinter to Waterdeep, then I’m prepping a travel route. If they want to attend Burning Man, I might prep a festival. And if they want to figure out where their rival gang is sourcing red opium from, I’d probably use node-based scenario design.

So even if we believed that scenario hooks are plots, once we realize that mysteries do not require GM-mandated scenario hooks, we can easily see how this entire line of argument collapses.

CLUES = PLOT

This brings us to the second common argument, which is that by designing clues and placing them in a scene you are prepping a plot. For example, by saying that Rachel works for Bobby (and, therefore, the PCs can discover this connection and follow it to Bobby), you are predetermining events.

This is, again, certainly something that you CAN do: The breadcrumb trail of clues, each of which can only be found in one specific way and used in one specific way.

I suspect, though, that most people reading this are already sensing that something doesn’t quite feel right here. How is stating “Rachel works for Bobby” a plot? Is that not clearly a situation — a description of the world state?

Imagine that I created a room in the game world and I said, “This room has a door and two windows.”

And then Bob said, “THAT’S A PLOT! You are predetermining that the PCs will enter the room through the door or the windows!”

I’m very hopeful that you can understand that Bob’s not making any sense here.

First, the players could easily enter the room in other ways: They could chop a hole in the wall. They could teleport in.

Or they might choose NOT to enter that room. Either because they simply choose to go somewhere else, or because they figure out some way of accomplishing their goals in that room without entering it. (They could scry on the room. Hire someone to search the room for them. Burn the house down and force the threat inside the room to come running outside.)

The leads in node-based scenarios work just like the doors and windows of that room, and stuff like permissive clue-finding is analogous to chopping holes in the wall. Node-based design is a way of thinking about how different parts of the game world are connected to each other — Rachel works for Bobby; Mathieu has a treasure map revealing the location of Shandrala; the street dealers get their red opium from a house on Oak Street — and prepping scenarios in which the PCs use information (i.e., leads) to navigate the game world.

You can hypothetically use node-based scenario design to force a plot, the same way that you could build a room with adamantine walls and endless GM fiat to force the players to solve the riddle that will unlock the door. But that’s something you’re choosing to do.

CONCLUSION

I’m mostly writing this essay because, when I get these questions in the future, I want to be able to just point people here. But I’m also hoping that it might help some people break out of a paradigm that’s limiting them as both GMs and players.

If you see a GM create a tree and your first thought is, “There’s no other explanation for this than that the GM is going to force me to climb that tree,” it’s important to understand that this is a warped perception. Even if that’s been your experience with one GM, you should know that there are other GMs running their games in very different ways.

And if you’re a GM who either (a) can’t create a tree unless you’re planning to force your players to climb it or (b) are paralyzed at the thought of creating a tree because you’re afraid it means you’re railroading your players into climbing it, then I truly believe your games will be better if you can jettison that way of thinking and, instead, embrace the simple maxim:

Don’t prep plots, prep situations.

Review Originally Published in Games Unplugged (August 2000)
Republished at RPGNet – May 22nd, 2001

Robin D. Laws is the esteemed designer of Feng Shui and Hero Wars, among sundry other games of high quality. Hogshead Publishing’s New Style line of games has included games such as Baron Munchausen and Puppetland, which have met with great critical acclaim. What happens when the two of them come together?

Pantheon. Five roleplaying games, under a single cover, of a curiously different sort.

The five games in question are Grave and Watery, Boardroom Blitz, The Big Hole, Destroy All Buildings, and Pantheon itself – each of which is based on the Narrative Cage Match (NCM) system.

What’s the NCM like? Think of it as a splicing of Once Upon a Time and Baron Munchausen, with a dash of Amber and Puppetland thrown into the mix. Like many of the other New Style games, the NCM is a storytelling game in the truest sense of the word – a system which doesn’t just talk about using traditional systems in order to create a story, but a set of rules which actually serves to focus the game session on the joint creation of such.

Basically it works like this: Each NCM game takes the form of a storytelling scenario – complete with plot seeds, goals, and characters. The degree of detail given varies depending on the particular game. For example, Boardroom Blitz has a Set-Up (detailing the fight to inherit the fortune of Dash MacMillan) and a Cast of Characters (from which the players can select their characters and gain insight into the supporting cast). The Big Hold, on the other hand, gives you a Set-Up and an Opening Scene (where the action starts), but doesn’t detail a specific Cast of Characters (leaving character creation up to the players).

Now here’s where it takes a turn off the beaten path: There is no GM in Pantheon. Instead gameplay begins when the first player submits a sentence. Play then proceeds to the second player, who submits another sentence, and so forth. This basic device is then complicated by a challenge system in which a combination of bidding counters and dice rolling will allow one player to rewrite the sentence submitted by another player. Eventually the story comes to an end (either because all the characters except one are dead, or because only one player has any bidding counters left) – at which point players score points based on the actions their characters accomplished (or failed to accomplish) during the course of the story. The winner, of course, is the player who has scored the most points.

Conceptually this is a really powerful system – not only can an endless variety of scenarios be plugged into it, but almost any given scenario can be played either humorously or seriously depending on which direction the players decide to take it. It is also a very different type of roleplaying game, which may leave open the question in the minds of some whether it is a roleplaying game or not.

The answer to that is an emphatic yes. On the one hand the game is clearly designed so that you assume and play a specific role. The methods by which that role is presented are very different from those used in a “traditional” RPG, but that merely means that a different set of creative skills are being used (with all the resultant changes in the types of stories you can tell). On the other hand, this is clearly a game – complete with goal-oriented awards. The fact that Pantheon is a different breed is a definite strength, not some sort of hidden weakness.

Unfortunately, the system does have its share of flaws in practice. Games with small groups can easily be ruined by an obnoxious player – primarily because the rules can easily be stretched to absurdity without actually breaking (run-on sentences, for example). The challenge system provides some recourse for this, but in a small group it becomes very easy for a single player to end up with more bidding chips than everyone else combined – essentially making it a cakewalk for them to force their distorted gameplay into continuity. This is particularly true since the mechanics of the bidding system make it inevitable for a consistently obnoxious player to amass more chips than everyone else (since the only person who sacrifices their chips are those who win challenges, if a person is consistently obnoxious – and therefore other people are challenging him to keep him in line – he is eventually going to have more chips than the other players).

Larger groups, on the other hand, tend to be more stable – but at the cost of some flexibility in character interactions (if there are always four or five sentences between you and another player, it becomes difficult for your two characters to meaningfully interact when all of the PCs are together). I also felt that the rules should have specifically addressed dialogue. Specifically: Just how constricted is the dialogue of our characters by the “one sentence” rule? And if it is constricted, then doesn’t that end up distorting character presentation?

Although these seem, at first glance, to be glaring problems, in practice they ended up being fairly minor concerns. The complications of large group interactions, for example, were overcome with a little practice and cooperation. The ability for a single player to ruin a small group game, on the other hand, is more troubling – but when push comes to shove, this isn’t really a game you want to be playing with those type of people, anyway. On the other hand, if a little more forethought had gone into the design of the rules (for example, by taking run-on sentences and dialogue into account) this would be a less pressing issue.

At the end of the day, though, there can be only one conclusion: Hogshead and Robin D. Laws have struck gold again. Pantheon is a solid kick in the pants of the traditional RPG form, and is pure fun through and through. Whether you play it with your tongue in your cheek or in pursuit of high pathos, this one’s definitely worth taking the time to check out.

Style: 5
Substance: 5

Designer: Robin D. Laws
Publisher: Hogshead Publishing, Ltd.
Price: $5.95
Page Count: 24
ISBN: 1-899749-25-X

The New Style games from Hogshead Publishing, although mostly forgotten today, are some of the most important narrative tabletop games every published. James Wallis, the founder of Hogshead, was a visionary and he deserves a lot more credit that he gets for laying the groundwork that the Forge and the indie RPG movement would start building on a few years later.

Pantheon is devilishly difficult to get your hands on today. Which is unfortunate, it lay the groundwork for a lot of Robin D. Laws’ later work with storytelling games, including the DramaSystem. Some time after writing this review, I had the chance to play in a session moderated by Laws at Gen Con, and that was really special for me as a young fan and creator.

See the note on my 1999 review of Baron Munchausen for how my thoughts on roleplaying games, storytelling games, and narrative tabletop games were being challenged here, eventually evolving into a much more robust understanding of the medium(s). You might also enjoy checking out my near-contemporary article “Hog Wild – The New Style of Hogshead Publishing.”

For an explanation of where these reviews came from and why you can no longer find them at RPGNet, click here.

When I’m running a campaign for a dedicated table, I try to make sure that the pacing is effective and engaging for the players, but I don’t worry too much about whether any specific scenario takes one, two, or a half dozen sessions to complete. (Among other things, it’s not at all unusual for there to be two or three different scenarios active in any given session.) I’ll still try to give each session a good conclusion, of course, but the campaign is not going to live or die depending on whether the PCs catch the ethereal troll serial killer in one session or three sessions: Wherever we happen to be in the action, we can wrap things up for the night and then pick up where we left off when the next session begins.

But sometimes you’re on a time limit.

A good example of this is a one-shot. If I’m running an adventure at a convention, for example, I can’t say, “To be continued!” There is no next week! I’ll likely never have a chance to play with these players again, let alone wrap up our story!

Fortunately, there are several techniques we can use when we need to hit a deadline.

ONE-SHOTS

Let’s start by taking a closer look at what it takes to run a true one-shot. How do we make sure everything wraps up in one session?

Prep an appropriate amount of adventure content.

The easiest way to blow your time limit is to prep too much adventure material. Ain’t nobody getting down to Level 5 of your dungeon in a four-hour session.

The appropriate amount of content will vary by system and circumstance, but my general rule of thumb for a four-hour session is generally five or six “meaty” pieces of content. Examples of this include a 5-Node Mystery or a 5+5 Dungeon.

Use a session timer.

I almost always keep a timer behind my GM screen that’s counting down to the end of the session. It’s very useful for maintaining pacing in any session, but it’s invaluable when you’re on a time limit.

I use a dedicated timer — rather than my phone or the like — because I want it to be visible at all times, so that I’m constantly aware of the temporal pacing of the adventure.

Running long? Start with aggressive framing.

Roughly two hours into a four-hour session, I generally want the PCs to either be in or heading towards the third meaty piece of content. If they’re still in the first or second node, it’s time to push down the accelerator with aggressive framing.

Aggressive framing means seizing suggested courses of action — don’t let players dither in their decision-making and don’t waste time on transitions. Instead, as soon as they say they want to do something, push hard and cut straight to them in media res doing it.

Then, on the back end of the scene, don’t be shy about sharp, definitive endings. Be quick to say, “This scene is done. What next?” in whatever forms works best.

Tip: There’s a converse of this at the one-hour mark, where — if they’ve already reached the third chunk of the scenario — I’ll relax the pace and maybe bring in a proactive element or reincorporate an NPC or other feature from earlier in the scenario. But, in my experience, this is rarely a problem.

The final hour.

If we hit the point where there’s only one hour left in the session and we’re still running behind, that’s where I pull the emergency lever. (You want to do this here because there’s still time to do it gracefully. If you wait until they’re only fifteen minutes left, it’s too late to adjust.)

Start by looking at where the PCs are and where they’re headed next. (Or, alternatively, where they could be headed next.) Following that line, what’s the vector that gets them to the conclusion of the scenario?

Aggressively prune everything else away. For example, do they head off to investigate a node that isn’t going to yield a lead pointing them to the end? Resolve it rapidly with one or two skill checks and then move on.

Basically, at this point, everything that isn’t essential should be treated as a dead end (even if it technically isn’t and you’d normally play it out in full). Check out Dead Ends in RPGs for more details on how to handle this.

Note: This is one of the reasons why it’s so useful to prep scenarios rather than plots. Actively playing a scenario means there are LOTS of potentially satisfying conclusions that can emerge from play, making it far more likely that the PCs will be near a potential conclusion when the time comes.

If, at the one-hour mark, the PCs are so far from a conclusion that it’s clear they’re never going to make it there no matter how aggressively you pace and prune, then you need to start taking more dire actions. (This includes all sorts of retcons and other stuff that I would never do in a full campaign.)

Option #1: Edit the vectors.

For example, whatever the next node that PCs go to is now miraculously stocked with clues that all point to the concluding node! Yay! Alternatively, a proactive node — similarly festooned with Clues Pointing Straight to the Finale — shows up.

In a dungeon? They search a room and find a secret staircase leading down to Level 5! Or the next hallway they go do down suddenly leads to the Fane of Nyarlathotep. Or the intervening rooms are still there, but it turns out all the monsters in those rooms were actually called away to another part of the dungeon. Maybe they’re responding to reports that the PCs were attacking the West Gate? Wow. It’s ironic that in trying to find the PCs they actually left the route to the Fane unprotected!

Option #2: Move the conclusion.

Instead of opening a path for the PCs to reach the ending, you can instead move the ending to wherever the PCs currently are. This often takes the form of either:

• The Big Bad finds the PCs and attacks them. (For example, if the scenario is hunting a werewolf, the werewolf pops up and attacks the PCs.)
• Someone tells the PCs where to go. (The werewolf attacks the chalet and an NPC calls the PCs to beg them for help before it’s too late!)
• It turns out the Big Bad is at whatever location the PC go to next. (Maybe the scenario was designed for the PCs to confront the Evil CEO in his office atop the Moebius Tower, but I guess it turns out the Evil CEO is conducting a surprise inspection at the warehouse where all the illegal drugs are being kept!)

When looking at your options here, it’s generally more effective to have the ending triggered by something that the PCs do. (In other words, if they choose to go to the warehouse and then the Evil CEO is there, then it feels like they found the Evil CEO! Good work! If the Evil CEO just walks through the door where they’re having lunch, it can very easily feel like nothing they did actually mattered.)

There are exceptions to this, but they tend to still be based in the PCs’ agency. (If you really need the bad guy to just teleport to their location and trigger the final fight, it helps a lot if they shout stuff like, “You’ve interfered with me for the last time!” and “You meddling fools! You thought you could blow up my chalet and kill my pet werewolf and there wouldn’t be any consequences?!”)

Keep in mind that you can also mix-and-match your options here. A good combo is triggering a proactive count that you can lard with a bunch of leads pointing to a location (or multiple locations!) where you can plausibly relocate the bad guy.

Tip: You may have noticed that having a proactive node designed into your scenario is incredibly useful for problem solving here. It’s always a good idea to include one. If you forgot to include one, Raymond Chandler’s “a guy with a gun kicks down the door” is always a good fallback.

LIMITED SESSIONS

Now let’s expand our horizons a bit and look at how we can handle a campaign with a limited number of sessions — either because we launched the campaign that way or because real life has imposed itself in some way. It turns out that a lot of the same techniques apply, just twisted slightly to account for the larger scale.

First, though, it can be useful to see if there’s an alternative solution to cutting the campaign short. For example, if you have a player who’s moving away, you might be able to arrange a satisfying send-off for that player and their character while the rest of the group keeps playing. (Check out Saying Goodbyte to a Player for a deeper dive into how to handle this.)

Alternatively, is there a way to increase the number of sessions you can play before the end? When I first ran Eternal Lies, one of the players needed to move to Atlanta to pursue her career as a stuntwoman, but we didn’t want her to miss out on the end of the campaign, so we ran ten sessions in fifteen days to wrap things up.

If options like those don’t work, then you’ll need to figure out how to wrap things up in the time that you do have. Start, of course, by figuring out how many sessions you have left. I recommend immediately assuming that at least one or two of those sessions won’t happen: Either something will come up and actually cause those sessions to get canceled — in which case you’ve preemptively solved the problem! — or they’ll provide some breathing room in case anything goes wrong. It’s much better to wrap things up early (and maybe run an epilogue session or something) than to run out of time!

Now, remember our guideline about five or six meaty chunks of content per four-hour session? Just multiply that by your sessions and you’ll know what your “adventure budget” is.

If you’ve been prepping your campaign as you go along, you just need to identify where your potential conclusions are and then vector appropriately through the amount of adventure content you have to work with.

If, on the other hand, you have an existing structure of some sort — a published adventure, a set of linked node-based scenarios, etc. — that exceeds your adventure budget, then you’ll need to figure out how to cut things down!

It turns out, this largely works the same way it does for individual adventures, you just have more flexibility and the luxury of prep time to think about how you want to handle it. For example, in an individual adventure your might say, “I don’t have time to run this full dungeon, so let’s remove Levels 3 through 5. The stairs on Level 2 go straight to Level 6 now.”

You can apply the same technique to, say, node-based campaigns: You can redesign the clues from Adventure 2 to point to Adventure 6 instead of Adventures 2 through 5.

Alternatively, if you have a Big Bad, you can have them turn up in almost any scenario.

Also look for places where adventures can be dramatically trimmed down instead of cut entirely: Maybe the Tomb of Raknar-Thalla was originally supposed to be a large dungeon with dozens of rooms and multiple levels. You can have the same clues pointing to the tomb, but instead design it as a 5+5 dungeon that can be wrapped up in a single session instead of several.

OPEN TABLES & UNFINISHED SCENARIOS

Another place where a GM can often run into a time limit is an open table: Here you want to wrap up a scenario by the end of the session because there’ll likely be a completely different set of players at the next session and you can’t leave things dangling or stuck on a cliffhanger.

As discussed in the Open Table Manifesto, one option is to sidestep the issue entirely by immediately scheduling a bespoke sequel session with the same players: Now you likely can leave things unresolved and wrap everything up next time!

If that’s not an option (for whatever reason), then you can, of course, always use the same techniques you’d use for any other one-shot and get things wrapped up by the end of the current session.

When it comes to an open table, though, it can be useful to ask yourself another question: Do you NEED to finish this scenario?

Sometimes you do: Investigating half a murder mystery and never getting the solution isn’t satisfying. If the PCs are in the middle of trying to escape a haunted ghost ship, then it’s probably important to know whether or not they get out!

But in other cases you clearly don’t: If the PCs are investigating a megadungeon, for example, they can easily have a satisfying experience, accomplish many things, and then leave without “finishing” the dungeon. (That’s because scenarios like this are holographic — any part of the adventure contains the full experience of the adventure.)

Other scenarios will exist in a gray area between these extremes. In my experience, there are two keys to figuring out whether you can leave a scenario unfinished or not.

First, can you still give the current players a meaningful conclusion? For example, maybe they’ve been tracking the illegal drug trade on LX-510. They weren’t able to track the drugs back to the black market chemlab they’re being sourced from, but with an hour left in the session they are on track to take down the gang responsible for the local trade. Framed right, that’s a solid conclusion.

Second, what are the consequences for leaving the scenario unfinished? For example, with the gang taken out of commission, how does the rest of the criminal network respond? What new scenario hooks might be generated from that? What leads do the current characters have that they could follow up in future sessions (and will that follow-up result in a full adventure experience for them)?

WE DIDN’T MAKE IT!

Whether it’s an individual adventure or a full campaign, sometimes — despite your best efforts — you just won’t be able to succeed in wrapping things up. The best thing to do, when it becomes clear you won’t be able to reach a satisfying conclusion, is to let the players know what’s happening and then work with them to find some sort of closure.

For an individual session, this likely means that about ten minutes before the end, you say something like, “I don’t think we’re going to get to the end of this scenario today.” You can then switch to highly abstract time and wrap things up in broad strokes: “Okay, so when you got to the warehouse, you find paperwork implicating Rebecca Li in Helen’s murder.” You can — and should! — still ask the players what their characters are doing. It’s just those declarations and their resolutions will be handled in much broader strokes. You can even call for dice rolls from the players, but each one is likely to resolve entire scenes, not individual actions.

For a campaign, if you’re going into a final session where there’s no chance of bringing things to a conclusion, you need to accept that reality. Try to wrap up whatever the players were doing at the end of the last session in an hour or less, and then, similarly, transition to an interactive “summing up.” Some of the techniques discussed in Epilogues & Skipping Time may be useful here.

The other technique here is to look at each remaining adventure in your campaign structure — whether that’s levels of a megadungeon, nodes in your Night’s Black Agents Conspyramid, or semesters at your magical academy — and resolve each one with a single round robin of action checks around the table. (I have a technique for resolving side jobs in my Mothership open table that can be useful for this sort of thing. I’ll try to share it here in the near future. You might also think in terms of each adventure being a clock, skill challenge, or complex skill check.)

It’s not ideal. But it’s better than nothing, and sometimes that’s the best we can do.

You probably know this technique.

Problem: Combat is taking too long. The players have analysis paralysis or they’re not paying attention.

Solution: Add a timer, requiring each player to wrap up their turn before the timer runs out.

It seems simple, but I’ve seen a shocking number of GMs screw this up. It turns out there’s a fairly large amount of finesse required to pull this off to best effect, so let’s take a deep dive.

MAKING THE TIMER WORK

A lot of what follows boils down to my personal opinion, but it’s also based on a lot of practical experience – not only mine as a GM and a player, but also my discussions with other GMs and players who have used combat timers in their games. There are some pretty common pitfalls, and often those who have gone astray won’t even realize they’ve fallen into a pit. Conversely, there are several best practices that will maximize your results when using combat timers.

A combat timer should not be standard operating procedure.

The first mistake is thinking you should always be using a combat timer. In reality, it’s an emergency measure that you deploy when something has gone wrong and you need to fix it. If combat at your table is moving at a good clip and with satisfying pacing, then you don’t need a combat timer. In fact, you definitely shouldn’t be using one. Even under the best circumstances, the combat timer introduces additional complexity – it’s one more thing that you, as the GM, need to keep track of. If you don’t need it, then focus your attention somewhere else.

More than that, combat timers are usually a temporary measure. In more than three decades of GMing, I’ve only had to implement combat timers a handful of times. In every case, we were able to drop the timer a little later with the game significantly improved as a result. Once people get a feel from what a properly paced combat feels like, they don’t want to go back.

Make sure you’re not the problem.

On that note, before introducing a combat timer, make sure you’re not the problem. If combat is bogging down because you’re the slow one, then that’s only going to exacerbate the problem. At a minimum, in my opinion, if you’re introducing a combat timer for the players, then you should also abide by the combat timer. Also spend some time practicing multitasking and other techniques for speeding up the group resolutions for  your NPCs.

Use a generous timer.

“If I’ve got six players and each of them takes 5 minutes, then every round takes at least half an hour. If the average combat lasts five round, then every fight is burning up two and a half hours!”

That math checks out, but it can lead to a mistake: “I want my fights to take up no more than half an hour. There are typically twelve combatants in every fight, so if we assume the fight will last five rounds, then every turn needs to be no more than 30 seconds. But we’ll want a margin of error, so let’s set the timer for 15 seconds.”

At first glance, this makes sense. But in practice it’s WAY too aggressive. I recommend nothing shorter than 90 seconds, and even two or three minutes might be the right fit for your group.

What happens in practice is that people don’t wait out the timer: The slight time pressure keeps them focused and, usually, decisions are made in 30 seconds or less even though the timer is longer. Occasionally they’ll take more time because the situation radically changed just before their turn or they misunderstood something and now they need to look up a rule or reconsider their options, and that’s okay.

Missing the timer delays your turn. You don’t lose it.

Time’s up? You’re obviously frozen in a moment of indecision. I’m going to resolve the turn of the next character in the initiative order and then we’ll come back and see if you’ve figured out what you’re doing.

The goal of the combat timer is NOT to punish the player. It’s to create a sense of purpose and focus through time pressure.

It’s a hard cutoff.

Making sure the timer isn’t too punitive also removes the temptation to ignore it, which is one of the biggest mistakes you can make. If things have gotten bad enough at your table that you need to implement combat timers as a corrective measure, then you have to stick to it!

Once you start letting the combat timer slide, the time pressure evaporates and it all falls apart. Now the combat timer is just meaningless busywork and you’re wasting everyone’s time with it. (This is also why it’s important to use a generous timer! So that you don’t have to make exceptions!)

Time to decision, not resolution.

On the other hand, generally speaking, the timer is for the player to declare their action. If the timer runs out while they’re resolving their action, that’s okay.

This is partly a matter of practicality (putting half-resolved actions on hold can create all kinds of weird mechanical issues), but it’s also because the primary thing I’m trying to tighten up with combat timers is the decision-making process. That’s almost always what’s kill the pacing.

If you’re also having problems with players taking too long to actually resolve their actions (e.g., the guy who shakes their dice for ninety years before finally rolling the damn things), you might want to address that by expanding what the timer covers. But I’ve found it’s usually more about helping the players refine their resolution techniques in other ways.

Note: The exception to this, for me, are systems where characters get multiple actions per turn. In the second edition of Pathfinder, for example, it’s not unusual for a turn to consist of decision-resolution-decision-resolution-decision-resolution. In those cases, you may want to have the combat timer apply to the character’s full turn. (Although, obviously, you’ll likely want to get a little more generous with the timer to account for this.) If the timer runs out, finish resolving the current action (if any), and then put the rest of the character’s turn on hold.

Two hourglass timers.

A few features you want your combat timers to have:

• The timer should be visible to the players. It puts pressure on them.
• As soon as a turn ends, the timer for the next turn should immediately
• You don’t want to have futz with your phone or some complicated interface.
• In my opinion, you don’t want something that beeps. Paradoxically, you want the timer to apply constant pressure, but when things are flowing well you also want it to seamlessly fade into the background.

The solution I’ve found works best is to have two hourglass timers: When a turn ends, flip the unused timer over and place it on the table in front of the players. Simultaneously grab the previous timer and place it behind your screen. Then repeat, switching back and forth between the timers.

This way you don’t have to wait for the timer from the previous turn to run out before starting the new one (which is a distraction and can create delays). If the players are resolving their turns so fast that the timer from the previous turn still hasn’t run out when it’s time to cycle back to it: Congratulations, you’re winning. You can flip that timer out whenever it finally runs out, which might not be until the start of the next turn.

Help the players.

Make a habit of putting the next player on deck, particularly when they’re losing focus. This will give them extra time to think about what they want to do.

Keep an eye on the timer and give them a ten second warning (or “sand’s running out!”) when appropriate.

Again: The purpose of the timer is not for the players to suffer the consequences of a timer running out.

The goal of the entire table — including you as the GM! — is for the timers to never run out. You should all be working together to accomplish that. So help ‘em out.