I recently got myself into short forms RPG as a GM and I really love the potential! I tried a short game (3 hours) and it went OK, but I am not quite satisfied yet. I need more practice, that's for sure, but there are parts where I don't know how to improve, and that's where I need your help.

The system I'm using is a personal design by myself and my group, not an established system. It has quick rules, a d20 for actions (with a consistent but not fixed outcome), and a lot of improvisation and flexibility from the GM. World is realist, but not too much: quite like an action movie, so if jumping though a window adds to the scene, only a bit of damage will be done.

A specific problem: a fast paced pursuit (20-25 min)

In one of the scenes of my game the PCs met a enemy NPC with valuable information. The NPC managed to get away and fled into the city.

I wanted this scene to be as great as pursuit scenes from action movies—fast paced, intense, challenging, leaving them short on breath in real life.

My approach: 1 minute hourglass, 1d20 for actions

I give a brief description of the scene to the group, turn the hourglass, let them decide what they do, and when the minute is over I move game time forward (e.g., make the NPC run, or attack them while fleeing).

Outcome: The feeling of being in a hurry worked well at the beginning, but they adapted by giving really short actions ("I run", "I jump", "I attack").

Reaction: As it was boring I decided to let them describe their actions more ("How?")

Outcome #2: More creative answers, less boring, but we totally lost the rhythm and it became a turn-by-turn RPG scene, definitely not what I intended.

I feel the group can manage to be creative while being stressed but as a GM I wasn't able to obtain the scene I wanted. How can I get the best of both worlds: fast paced action scenes and creative actions from the group? Props (a clock, time-tokens?), techniques (words, tone?) or just more practice?

  • 3
    \$\begingroup\$ If you're playing in an established system, I suggest you say what it is, because then people can point out mechanics your system offers that can help you out with doing this, or point out which mechanics to ignore that will be unhelpful. It sounds like you might be using something of your or your group's independent design, though, in which case it might be worth saying so for clarity. \$\endgroup\$ Jun 2 '14 at 14:00
  • \$\begingroup\$ So your system doesn't have anything for a chase? \$\endgroup\$
    – okeefe
    Jun 2 '14 at 14:11
  • \$\begingroup\$ I use a personal design. Quick rules, d20 for actions (consistent but not fixed outcome), lot of improvisation/flexibility from the GM. World is realist, but not too much: quite like an action movie, if jumping though a window add to the scene, only few damages will be done. \$\endgroup\$
    – Clément
    Jun 2 '14 at 14:25
  • \$\begingroup\$ My answer may not be quite so helpful then; you've already got a pretty minimal system! \$\endgroup\$
    – Rob
    Jun 2 '14 at 14:50
  • \$\begingroup\$ It's my fault, my question wasn't complete. Thank you for your answer, I upvoted it and may chose it as the answer because the 4 first dots are totally relevant and useful. \$\endgroup\$
    – Clément
    Jun 2 '14 at 14:57

Lose the battleboard, abstract, shortcut and improvise

TLDR: This answer focuses on downplaying the importance of rules, making them a more abstract guide for capacities in order to speed things up.

Mechanics can really slow down a scene, if you've ever played a battleboard system (like D&D 3.5/4.0) then you know players can spend ages just working out where they're going to step, onto which square, what attack they're going to use with with feat and yawwwwwwn I'm already bored - this is meant to be exciting. Rules and feats and skills restrict what the players think they can do, restrict them.

I ran Rolemaster for many years (a pretty heavy rules system) and I kept the details down to a minimum and described everything I could; if I really had to I drew a rough sketch with a pencil on a piece of throwaway paper but everything else was description, the rolls were frequently abstract checks, but checks to keep things going.

Important: The players need to get on-board with this idea, that the rules are there to assist, but are not the be-all and end all - rules lawyer characters who start quoting pages are going to kill this sort of game flow and won't like this sort of game.

How do you do this?

  • Ignore initiative; go round in order of seating. You make one check; the players win or the monsters win; you play the monsters, then the players go in turn. If you want a bit more leeway then the player at the left or right of you who gets the highest roll goes next.
  • Keep it rolling; time is abstract for this; use verbal cues in the game to indicate the impending problems and time-critical nature and it will spur them on "You see the escaping thief is just one roof away from diving into the river!"
  • "Yes you can"; let players try almost anything improvise what skills, feats and spells can do, be adaptable - that magic missile can blast a rope or blow out a window, that cleave feat can chop through a door quicker, you get the idea - the sooner players realise that they can abstract their abilities the more creative they'll be. If they ask if they can try something tell them they can try! However beware that it doesn't lead into sillyness, you want a serious chase, right?
  • Lead by example; anything the players can do, the NPCs can do too; show them what sort of thing is possible, get a lightning bolt to bounce off a mirror, etc.
  • Keep it short; use character names, when a player finishes their go immediately switch to the next "Sir Bob, what are you doing?" get them to answer quickly; it keeps them in thread, if they take too long, tell them they are dithering and come back to them after you've asked the next player along.
  • Ignore/shortcut the rules that bog stuff down, facing, distance, etc; don't use a board; use "about 20 feet away" if they want to hit, let them, if it's reasonable; but keep it dramatic; give them bonuses for other stuff; if they want a flanking bonus, if there's a couple of people fighting the same guy, give them the bonus, if they've swung over and kicked someone in the head from a chandelier give them a bonus for that.
  • \$\begingroup\$ Question updated: I use a personal design with really light rules. However I find the 5 first dots really useful. I'll just wait a little time before accepting it as the answer. \$\endgroup\$
    – Clément
    Jun 2 '14 at 15:04

The hourglass approach you describe sounds fun and exciting, and it sounds like it was - for the first few rounds. After that, the players started only giving short answers because they felt it was sufficient. When they gave longer answers, it no longer felt like a chase.

The problem you faced, I think, is that after a few rounds the players were used to the situation and already knew what they wanted to do. The best way to counter this is to change the situation up every two or three rounds, so the players have to change their approach. If you keep making the players come up with new ideas, the time constraints become meaningful again.

Think of action movies. The person fleeing does things like ramp over rivers, take dangerous turns, try to shake off the pursuer and many other things. You could make the NPC cover the road in rubble, or turn into oncoming traffic, or make a handbrake turn and disappear down a narrow alley... there are many options but the key thing is that the PCs have to respond to it - if it doesn't lead to a meaningful choice, it's not a great idea.


There are two ways in which it can be fun:

1) Interesting choices 2) Interesting outcomes

Interesting choices is stuff like which path you take, whether it's a better idea to jump over obstacle A or try to break through obstacle B, etc. etc. The problem is for a choice to work, the player has to know what the obstacles are, make a choice, and go with it, and you also have to generate interesting choices continuously. This is not a quick process.

Interesting outcomes is descriptions that result in mayhem and action. Instead of making this a result of specific choices, you can make it a challenge to the players to creatively come up with better, fun solutions.

"I jump over the cart, and swing from the hanging flag pole and land on a messenger's horse" - the player describes this, even if you didn't say there was a flag pole, etc. because it makes an interesting narration - they narrate things into play, and things move quickly and in an exciting fashion. This then becomes a lot like FATE games, or many narration-trading games.

The benefit is this is pretty exciting and quick to do, the drawback is that your players have to exercise a different sort of gaming skill - creative narration, rather than tactical decision making. Not all players can do this, not all players can do this quickly, and not all players are into it.

Because chase scenes are very dense in terms of action-reaction choice to outcome options, they're actually one of the most complex things in all media and one of the hardest to model in roleplaying games.

I'd focus on getting interesting outcomes via open narration, finding a way to reward good narration (bonus to rolls, xp, whatever), keeping it relatively short 5 minutes or so in total, even if that means just 1-2 turns for everyone.


The mechanical solution

I think that the hourglass is a step in the right direction, but it can't work alone as we saw here. In order to solve that, I have a little suggestion that I've used for quite a time in the past, when I GMed D&D games on a regular basis.

Instead of a minute for the group, I gave about 15 seconds for each player to decide what her character is goona do. This gives them roughly the same amount of time to decide what they really want to do while still keeping the time limit close to home. As I've encountered the same problem as yours I decided to give a mechanical benefit to actions that are being described in greater detail.

This came to be my go-to way of GMing chase scenes in D&D. Each player has 10-15 seconds to act. If they describe what they do in a more exquisite way than just saying "I attack" or "I ran after her" I give them a +1 to their action. Have they used the entirety of the 15 seconds for their description they get a +2 to the action. This way, I give them a carrot for describing things, while still managing to keep it short so it will feel like there is a time limit. And there is a stick also, in the way of a time limit which makes them want and need to act fast or they will lose their turn to act.

Solving through GMing techniques

I think that there are three important things to consider that will hopefully help them to describe their actions in greater detail.

The first thing is to lead from example. Show them what they need to do, how one should describe, let them see the way it needs to be done and they will copy you, imitate you, and they will describe in far greater detail.

Make them swim in details. Let them know what lies in the area, so they'll know what they can use. Let them know that there's a barrel to the right, a rooftop to the left and a gun in the store next door. If they'll have the details they'll have the tools from which to build their descriptions.

Let them add details of their own. Many a time they'll really want to do something which will require them to have a thing that you haven't thought about. Let them have it, let them add small details to spark their descriptions. As long as it is logical and/or cool, let them have it.

In addition, I highly recommend reading a much related question from some time ago. I truly believe that it can prove relevant to your problem, even though it is tailored towards FATE.

  • \$\begingroup\$ Helpful answer too (in fact every answer makes a unique contribution). I liked the emphasis on rewards and adding details the group can use. Related question contains valuable information indeed. \$\endgroup\$
    – Clément
    Jun 3 '14 at 14:35

Don't ask how

This is so, so critical to running timed systems! Of course your players give shorter responses under pressure! That's the point!

I have a fair deal of experience running timed games, including one-sided timed games, and this is how you do it right (at least, according to me. Feel free to do differently, but doing it this way will solve your problems, I think):

you describe how the players do what they do. They do not describe that. That is not compatible with a one-sided timed round structure. Instead, you describe what they do. The player says "I run". You say "You lean forwards, pumping your arms and drawing breath as best you can despite the bitter cold and the growing dull pain in your chest. He's fast. You have to duck your head to keep up with him, only glancing up every now and then to keep him in your sights. Fortunately, with the drop, there's only one way to go. You aren't faster, but you are in much better shape, and eventually he is almost within your reach, his will to flee slowly eroding in the face of his mounting exhaustion." The player says "Can I attack him?" You say "you can try. He's still running, you aren't quite sure you'll reach him and it will slow you down to try. You might be able to, though." The player says "I do it anyways". (roll) You say "You leap for his back, and manage to grab a shoe as you plow down face-first into the snow. You immediately shove yourself back up, adrenaline numbing the pain and rawness where your left palm found concrete. You tripped him, and he's not quite as quick to get up as you are. You quickly move to hold him down. Where do you put your hands?"

and so on. In a timed game where the GM is not timed, you narrate the players' actions. This is why, in such games, it is extremely important that you learn the players' characters and take the time to feel out how they normally act before getting into the meat of the game: you need to be able to take a short thing like "Bullet in the guy by the window" and turn it into a thorough description with whatever details you feel like adding with automatic player buy-in. It needs to be good enough, true enough to the player's version of their character that they think 'yes; that is what I meant for my PC to do, or at least, that is how my PC would do what I have said they are doing'. True enough that then, later, when you punish a PC for having used a bad approach, for having engaged a detail that you put there yourself with no player explicitly saying so whatsoever, that that PC's player thinks 'darn! I should have been more careful about suchandsuch' rather than blaming you for the failure.

Timed games can be great fun, in a wide variety of forms, but eloquent discursion and time pressure are not compatible and so those operating under time pressure must not be expected to use that time on uneconomic discourse.


I have a lot of chases in my games and developed my own quick system to balance these two concerns. The trick is to abstract it enough (off the battlemat, so to speak) to be able to throw in complications that the PCs then need to do things to avoid - this is where the interesting bits come from. But since there's still a chase on, if they bog down in a given complication they fall behind, which provides incentive to keep the pace up.

Here's my full chase rules (9 page pdf), which is meant for Pathfinder but really can be used for anything that's d20-ish.

  1. Turns a fixed Move into a d20-based move check (allowing mixing of movement types - on foot, vehicles, swimming, flying, whatever), as different movement rates should be relevant but not deterministic
  2. Simple "range bands" say how far apart two participants are
  3. Provides guidance on when/how you can attack someone and the consequences
  4. Obstacles allow for a variety of techniques/skills to bypass them and promote PC innovation

My PCs have used these to chase people on foot through a city while parkouring fruit carts, to chase a pirate ship down the length of a pier trying to get on, to run from rampaging shoggoths... The monk ends up vaulting over things, the barbarian ends up crashing through things, and the portly cleric ends up puffing in the rear just trying to keep them in sight. It's made our chase scenes as engaging as any other kind of scene.


One of the most successful extended scenes I have ever run was a chase scene through an unfamiliar forest in a D&D 4e campaign. I thought it was successful and had a blast running it, and the players seemed genuinely caught up in the tension of catching or not catching their target.

I just ran it loosely as a skill challenge per the 4e rules, where the idea is that the group needs to have so many successful skill checks (say, seven) before then have so many failed skill checks (say, three-- I don't remember what numbers I actually used.)

The keys to using this for a successful chase scene were, in my opinion:

1) The ultimate success or failure was not, "Do we catch the fleeing whatsit," but, "Do we catch the feeling whatsit before or after it reaches its destination and does bad things?" There was going to be a climactic fight either way, but the location mattered for both role-playing reasons (someone they wanted to protect might be dead) and combat reasons (fleeing whatsit might be stronger if it reached its target.)

And that was all more or less understood.

2) In contrast to some of the more open, player-led skill challenges where they are pondering what skills to use and how to apply, I had prepared a bunch of foresty obstacles-- a sudden wide ravine, the negotiation of a swift stream, signs that they were approaching a bear's lair-- etc, and doled them out in rapid order (I think I had shuffled the in a pile of cards and picked randomly.)

The idea is that each one could be resolved quickly with one or two die rolls, and a success or a failure result could be quickly improvised by me based on how they handled it. Usually a failure meant loss of time or resources, and success meant someone did something awesome to save time. For example:

  • "The ravine looks like a serious problem, but the druid finds a half-dead partly rotted tree, and the warrior manages to fell it with his axe so it forms a bridge!"


  • "You've tied yourselves quickly with ropes while trying to step across the slick stones in the stream... and that's good, because the sorceror misses his step and is almost swept away. It takes time, but you fish him out and continue on your way, wet and bedraggled."

(I polled a bunch of gamer and non-gamer friends with more outdoorsy experience than I have in coming up with the list, by the way.)

3) It was ABSOLUTELY IMPERATIVE that the players be given an IMMEDIATE sense of progress or lack of progress along the way. And endless series of skill checks with no sense of progress is a bad idea in general, but a terrible idea for a suspenseful chase scene. (In this case, the chase target was visible from a distance. Don't ask.)

It was also imperative to not let the players dither too much over the obstacles in their way. You don't want 2 minutes of action and description with 28 minutes of discussion and planning. You want 5 minutes of action and description and 2 minutes of discussion and planning for each obstacle. A half dozen obstacles can go a long way and keep things moving smartly along.

Over all, it worked beautifully. I never fully explained the mechanics to players (at least, not during the scene) but just kept the pressure on. The successes were well received, and the failures made for some comic moments that were remembered through the rest of the game, because they were meaningful without being devastating.

I had always planned to revisit that chase thematically (in plot terms as well as mechanics terms) with an urban chase scene, but it never happened. Still, there is no reason this could not be extended to an urban chase, or to a weeks long chase over open terrain, etc. You just have to be knowledgeable enough (or have access to that knowledge) to come up with appropriately scaled obstacles.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.