How Initiative Makes Running Awful

I think the primary reason that players don’t run from a fight is truly that they don’t even consider the option most of the time. That might just be a psychological attitude that is somehow inherent to people who play D&D: if my character dies, I’ll just make another one, or get resurrected by magic, or something. Death is more of an inconvenience than a finality. Besides, what kind of heroes would we be if we ran away from a fight?

Quite aside from all of that, though, running away from a fight produces a mechanical problem within the rules of D&D, and the problem is the concept of initiative order. After everyone in the battle has rolled their initiative, everyone has their own particular time to act. Aside from certain picky bits like preparing an action for later, or taking a reaction after someone else does something, the combat system is very turn-based, and you always wait until your turn to do anything.

So, if the PC’s want to run away from a fight, the initiative order mechanic makes it necessary for each of them to run away on his or her individual turn. This poses some serious problems, though. What about the characters whose turns come after the decision to run has been made? They end up remaining in the battle, and maybe becoming the only targets within reach of their enemies. And as far as the enemies, how are they supposed to react to fleeing characters? Presumably they have to decide whether to pursue or not, but they have to make that decision based on individual characters running away on their own turns.

All things considered, the turn-based nature of combat in D&D is a good thing. It allows us to handle the fact that we have a lot of entities taking actions which last approximately six seconds, and that some of those entities are going to get to make their choices before the others. What we too often fail to recognize is that, if the battle were actually happening in real life, all of those six-second turns would be happening simultaneously. Everyone would be doing their actions all at once. And realizing that all of the individual turns in a combat round can be considered to be happening all together is what allows us to create a system for combatants to run away from a battle, or surrender to their opponents.

This can be designed to work for both PC’s and their combatant enemies, and the steps are pretty much the same. It’s worth pointing out that allowing the enemies to flee or surrender in this way is really useful for DM’s to portray certain types of creatures realistically. If you have to wait for the goblins’ turn to come around in the initiative order before they can make a run for the door, it’s possible that they might all be killed before they have the chance. If the goblin boss (on a separate initiative count from his followers, of course) gets killed, the rest of the goblins have to wait their turn to throw down their weapons and surrender. So, while I’ll be mostly talking about PC’s running away, remember that this mechanic works for either side of the fight.

How To Do It: Everybody Runs at Once

The upshot of the mechanic for running from a battle is to realize that, in a real-time interpretation of the fight, everyone is doing their actions simultaneously. When you look at it this way, you can get rid of the idea that everyone only gets to run away on their particular turn. Instead, everyone can decide to run all at once. And we can let this happen any time in the combat. Let’s have an example, from the PC’s side of things:

At this point, every member of the party can make a decision as to whether to run or to stay and fight. There are benefits to either option mechanically, but it’s also important to respect that fact that properly playing the role of some PC’s means not running from a fight. So, player agency comes to the front, as it should. Then, we have to figure out what happens next.

Running, Chasing, and Getting Away

In order to make this a workable mechanic, we have to account for certain types of events that occur when a group of combatants decides to run. We have to decide whether their enemies will pursue them, and how dedicated to the pursuit those enemies are. In the event of a determined pursuer, we have to have a way to decide whether individuals or groups manage to escape pursuit. Finally, we have to make sure that we allow PC’s to take actions to alter the situation, instead of holding them to a predetermined course.

As with all of my rules hacks, it also has to be simple, and to rely as much as possible on existing game mechanics. I will always forego some realism in order to keep things from getting too complicated. However, let’s start with some assumptions about fleeing from battle that are based on realistic expectations for a D&D fantasy setting.

What Does a Run-and-Chase Look Like?

When we consider how a D&D party running from battle might play out, there are some assumptions that will either be helpful or just outright necessary. Let’s start with these:

So, we have a picture of a run-and-chase forming: it needs to happen quickly and over a short distance, there needs to be a defining event that determines success or failure, and there need to be several ways for that defining event to occur.

Pursuers and Quarries

We’re also going to have to come up with some basic chase decisions for the DM to make on the part of the PC’s enemies. The PC’s should be able to decide for themselves what they want to do when their enemies run: pursue them on foot, cast a spell to stop them, let them go, or whatever else might occur to the players to try.

However, different types of monsters and other NPC’s are going to have different reactions to the party running away from a battle in progress: in short, there will be different categories of pursuers. In the interest of keeping things simple, there will really be only three categories, and only one of them will require any further mechanics to deal with adequately.

It is important to note right off that the players won’t know for certain what sort of pursuer they are dealing with, although they might be able to come up with a reasonable guess based on the circumstances. So, when we give general behaviors for non-pursuers and ambivalent pursuers, we can keep them very basic, because the players will have to expect the possibility of a determined pursuer.

The behavior of the non-pursuer is easy: when the PC’s run, they get away, and the DM just narrates it as such: “as you flee from her lair, the giant spider clicks her mandibles angrily and returns to caring for her egg sacs.” The retreat has succeeded, and the PC’s end up just a short distance from the site of the battle.

The behavior of the ambivalent pursuer is also easy. When the PC’s run from an ambivalent pursuer, the pursuer follows them. At this point, the players are going to have to make a decision on how to proceed, but it doesn’t really matter what decision they make; they have already escaped, although they might not know it yet. They can use abilities or spells to hinder the pursuer (remember, they don’t know whether this pursuer is ambivalent or determined), they can continue to run away, or they can turn and fight. If they turn and fight, we’re back into initiative order for further combat, and the previous combat continues, but in a new location. If they do anything else, they get away, and the DM narrates it: “the cave bear isn’t hungry enough to continue chasing you, and returns to its cave to wait for easier prey.”

Determined pursuers are where we get into trickier territory, and this is where the concept of “screw-ups” comes to the forefront. We could make this very complicated, but the fact that we’re talking about a short chase and a quick chase means that we can keep it quite simple. We just need to consider what kind of events count as screw-ups, and how we adjudicate them.

Screw-Ups for PC’s and Their Enemies

Essentially, we’ll be dealing with two types of screw-ups: the kind that happen to players, and the kind that happen to their enemies. There will be a bit of overlap, but that’s fine. After all, the rules don’t treat PC’s and NPC’s exactly equally anyway, so we can get away with a little bit of a double standard without causing too much trouble. Because there are fewer of them, and because they are simpler, we’ll cover enemy screw-ups first. Remember, though, that a lot of the items we’re calling enemy screw-ups might apply to PC’s who are pursuing their fleeing enemies. Likewise, the PC screw-ups might apply to enemies being pursued by PC’s. We’ll have a little more discussion on that later, but just keep that in mind for now: we’re mostly concerned with what happens when PC’s run, but enemies can flee as well.

And those are essentially the forms that enemy screw-ups can take. You’ll notice that whether enemies continue a pursuit comes down to two factors: whether they want to continue to chase the PC’s, and whether the PC’s do something to stop them or slow them down. In a minute, we’ll discuss the notion of random screw-ups, but those are intended to apply to the PC’s, not their enemies, and there’s a reason for that.

So, let’s move on to the kinds of screw-ups that the PC’s can suffer. Essentially, if any of the PC’s suffer a screw-up, then the PC’s that screwed up will be caught up to by the pursuers. Of course, their comrades can decide to stop and help them, or just keep running. I would hope that most adventuring parties wouldn’t just leave a fallen hero to be killed by whatever they’re running from, but it’s always a possibility.

In any case, the rest of the party can decide immediately whether to stop and help, at which point anyone who suffered a screw-up or stopped to help goes back into initiative order for combat. As in the situation where some of the PC’s stayed behind to cover the escape, we need to require that the pursuing enemies get at least one chance to attack any PC’s who fled and got caught before those PC’s can decide to try to flee again.

So, here are the possibilities for PC screw-ups.

So those are the basic PC screw-ups that allow the pursuers to catch up to them and resume combat. Remember, a lot of the enemy screw-ups and player screw-ups can work in the opposite direction if it’s the enemies fleeing from the players.

However, there’s a completely separate type of PC screw-up, and I intend it to only apply to PC’s, and not to their enemies. This type of screw-up is based on just getting unlucky and running into an obstacle that might stop a PC from continuing to run away. Why shouldn’t this apply to enemies? Mostly because it would be extremely inconvenient to work out which enemies would be the unlucky ones, but also because it’s safe to assume that, if the PC’s are chasing enemies, they are probably going to catch them by hindering them in some way. PC’s can pull off that trick because they almost always have greater resources than their enemies. The enemies, without as much access to spells and special equipment, need an edge to keep things interesting. That’s where the bad luck PC screw-ups come in. And as with anything in D&D that has to do with luck, that means we’ll be rolling some dice.

Bad Luck Screw-ups

In order to create the sub-mechanic for screw-ups that are just bad luck, we need to have a few things worked out:

If we can get all of that straightened out, we should have a decent way to adjudicate screw-ups that happen just due to misfortune.

The first question is a simple one: we don’t even check to see if a bad-luck event happens until after all of the other screw-up circumstances have been handled. In other words, we don’t even need to think about possible misfortunes unless the PC’s are about to escape from a determined pursuer without any trouble.

How often these misfortunes should happen is ultimately up to the DM. If they happen too much, they become an annoyance. If they don’t happen often enough, the players forget about them, and that’s a detriment to their decision-making regarding fleeing from combat. My recommendation is to roll the frequency issue in with the issue of who is affected. Whenever I need to pick a random character for any reason, I just assign each player a number, starting from my left and going all the way around the table. Then I choose a die with enough sides, and roll to see whose number comes up, and then I have my randomly selected PC. This will work for deciding who gets the bad luck as well, as long as you choose a die with a few more sides than you have players. So, if you have four players, and you want a random misfortune about 65% of the time, roll a d6: a result of 1 to 4 selects a player, and a result of 5 or 6 means nobody encounters a chase misfortune. If you want your misfortunes to be a little less frequent, use a d8 instead to get a random misfortune about 50% of the time instead.

I would recommend against having the possibility for multiple PC’s to suffer misfortunes, in order to make the rest of this make sense. But, if you really wanted to, you could have 1 to 4 select a player, 5 selects nobody, and 6 means you roll twice more and ignore further results of 6.

So, now we have a way to figure out which PC to target with a random misfortune, and we’ve integrated a way to determine about how often a misfortune occurs into the same die roll. So far, it’s been very clean and simple, and that’s good. The next question is to figure out what sorts of misfortunes can occur, and we’re going to use the last question about possible mitigation to come up with the basis for a list.

Deciding on Specific Misfortunes

We’ll work backwards a bit on this, because choosing events for misfortunes is ultimately a question of DM narration. Once we figure out the general sort of thing that happens, the DM can come up with an in-world event to go along with the dice rolling. And, not to worry, I’ll provide a starter list of events a little later on.

When we come up with a random character to get a bit of bad luck when running from a battle, we want to make sure that we provide a way to save out of the detrimental effect. We’ll be talking mostly about environmental hazards: types of things you might encounter in a dungeon environment that would hamper your escape. It’s pretty reasonable to say that we’ll be using STR, DEX, and CON saves for almost all of our bad luck mitigations; there’s probably nothing you would encounter by running that would require those more cerebral abilities to overcome.

And, because different PC’s have different ability scores and different save proficiencies, we want to randomize which misfortunes occur in order to give all the PC’s a chance to use their best skills. I decided to roll a d4: 1 requires a STR save, 2 a DEX save, and 3 a CON save. 4 is a special case, which we’ll get to in a moment.

Any save needs a DC, of course. I generally set these at about 12 or 13, and I leave them there as the PC’s level up. If you wanted to increase the DC with PC level, that’s your call. I don’t really see how chase misfortunes would become more serious at higher levels, but you might want to raise the DC in order to keep them relevant.

Once you’ve chosen a character to have the bit of bad luck, and rolled to find out what save is needed to overcome that bad luck and get away, you’ll just need to narrate what troublesome thing happened as the PC was running. And the narrated event will have to match the type of save required. I promised I would give some samples, so here they are:

So there are some samples for you. You can use your imagination and the environment your adventure is taking place in to come up with more of your own. You can see a little better now why I suggest only one PC encounter a misfortune: if you have more than one PC affected, you suddenly have these little problems crowding around everywhere, especially if you roll for the same save for each of the PC’s involved. I think it’s much better to keep it down to one PC, for the sake of simplicity. If that PC gets the screw-up, probably everyone will be stopping to help, and if that PC evades the screw-up, then everyone gets away.

I promised to get back to what happens if you get a 4 on that d4 roll to determine the nature of the misfortune, and the answer is… anything you want. 4 is a wildcard, and you just pick a hazard that works with the environment. It could be an arrow trap, falling into a sinkhole, breaking open toxic gas spores, or anything else you can think of. There are only a couple of restrictions you should observe. If it would count as a screw-up and end the chase, it should have some sort of a saving throw. If it’s something that creates a lasting effect (like gas spores now floating in the passageway), consider whether it might be a screw-up for the pursuers even if it might not be one for the PC’s.

Summarizing the Mechanic

From the beginning, here’s a brief summary of how it works. The PC’s can choose to flee combat all at once at any point in the initiative order. If any choose to stay, they have to wait until an enemy’s turn has passed before they have the option to flee.

If the PC’s get chased at all (and sometimes, depending on the enemy, they won’t), they can escape in two basic ways. Either they do something to screw up the pursuing enemy, or they avoid screwing up themselves. If they can pull off one or the other, then they get away. They can also always decide to turn and fight.

If nobody does anything along those lines, there’s a secondary mechanic which allows the PC’s to encounter bad luck as they run away. Using die rolls, the DM selects a random player to run into an obstacle which has a random saving throw to avoid a screw-up, and narrates the nature of the obstacle to the player. If the PC saves out of the danger, then everyone gets away; if they don’t save out, then that’s a screw-up and they get caught. The bad-luck mechanic doesn’t usually apply to enemies, because they have far less capability to force a screw-up on the fleeing PC’s, due to lack of spells, abilities, and equipment; the bad-luck mechanic evens the field a bit.

And that’s how it works. Remember that for the players this looks very simple: let them know the general circumstances in which they can decide to flee combat, and just narrate what happens. The players will have the opportunity to use their various resources to get away at multiple points, because after you narrate, they get their chance to act. When enemies flee from the PC’s, it works the same way.

Don’t forget that enemies can also surrender at any time in the combat, just like they can flee at any time. I suppose the PC’s could surrender as well, but I wouldn’t ever expect that to happen.

I hope this gives you DM’s out there something to work with so that your players can exercise the better part of valor at least some of the time. I know it’s been helpful for me, especially since I don’t have to worry as much about any particular encounter being too difficult for the party… they can always turn and run if things are going really badly. Granted, they’ll probably still fight to the death most of the time, but at least now, they don’t have to.