Splitting the Party

I have a guilty pleasure when it comes to DMing. In fact, I probably have several, but one came up recently as I was browsing the Pathfinder Reddit: splitting the party.

Splitting the party is generally considered A Bad Thing™. The Reddit denizens (Redenizens? Renizens? I’ll workshop it.) are against it almost universally, impressive for any online community. TV Tropes agrees (…usually). I could swear it was one of the taglines for D&D 4E or Pathfinder promotional material, though I don’t have too much of that lying around. People have even written songs about it.

Why is splitting the party bad? A few reasons, all related:

  1. It divides the party’s abilities. In an archetypal party, splitting in two means only one side has a healer, or a slashing weapon, or something else. Characters form parties specifically because each member offers something the rest of the group doesn’t have, and splitting the party negates that advantage.
  2. It breaks the power balance. Most adventures are designed for a group of X people, each with a certain number of actions per turn and a certain set of resources. Charging into a battle like this with only X / 2 players is a good way to run into a TPK.
  3. It divides player attention. When the game is focused on group A, group B isn’t doing anything. The players are watching other people game and waiting for their turn. It’s boring.
  4. It divides DM attention. Now he or she has to remember two groups, their progression, their location, and what the world around them is doing. Multitasking like this leads to slower, more confused play and frustration.

Reading them in a list they seem like good reasons: splitting the party leads to a bad gaming experience. But I’m not convinced these are problems with splitting the party as such.

  1. Resource consumption is going to happen whether the DM intends it or not. Eventually the cleric will run out of healing, or the ranger will shoot her last arrow, or the fighter will go unconscious. That’s why characters have backups (a wand of cures, the rogue’s quiver, a summoned monster to hold off the enemies for a few rounds). One of the DM’s responsibilities is to make sure the game never gets to a point where it’s impossible to advance, and part of that is not basing survival or progression on a specific player having a specific resource at a specific time.
  2. Power balance in combat design is mutable, almost to the point of being unbreakable. If half the party stumbles into a room where a DM planned to have six orcs, she’s well within her right to change it to three orcs. Only the DM really knows how many fireballs the enemy wizard has left, or where the pit trap is, or whether the cult leader has Dodge or Toughness or Skill Focus (Perform [oratory]). A TPK isn’t any more likely for half a six-person party than it is for a three-person party when the challenges are in flux. Only the most rigid DMs can’t handle any change in their plans.
  3. The nature of having specializations means that as some point somebody is going to sit out. If the party is fighting undead, the enchanter can’t do much. If the party needs to disarm traps, only the rogue gets to play. If the party is talking to NPCs, the face of the party is doing it, whether that’s the actual party diplomat or the de facto diplomat by way of being the player most comfortable with it. That’s built into the game, and we’re not even getting into situations where a player simply isn’t interested in what’s happening at the moment. If everybody is acting at all times, you don’t have a tabletop game, you have a shouting match.
  4. The DM’s job is to manage monsters, NPCs, plotlines, maps, settings, the players, etc. all at once, and do it in an entertaining, engaging way. If the players split up, the cognitive load of trying to handle two rooms at the same time isn’t going to drive a DM to tears and force him to start killing PCs to lighten the load (which is, and this is true, an actual consequence that players legitimately discussed on the above Reddit).
There’s a trend here: the DM should be able to handle a split party. If he or she can’t, that’s not the players’ fault.*

I’ll admit to having a bit of a formative experience here. In the first session of one of my earliest campaigns as a player, we split up to explore an absurdly spacious sewer and look for…something. Along the way, half the group stumbled upon bandits, and these bandits promptly squashed that half of the group. The fracas alerted the melee-capable members of the party who ran to the rescue in time to save us from dying. I distinctly remember the DM complaining about it the next week, completely nonplussed that the party (who had all met twenty minutes ago, who’d by that point only taken a walk and killed a spider together, and who hadn’t shared both their first and last names with each other) would do anything but walk joined at the hip, trusting each other with our lives in a place where we did not believe lurked any immediate danger.

First off: never tell the players dying (or nearly dying) was their fault. Wait for them to say it first. More relevantly, it wouldn’t surprise me if some of my love for splitting the party came from this. That is, if a bad DM DM we didn’t like was adamant that splitting the party was bad, maybe it was good when handled by a DM we did like. I ended up with a lot of opinions from that campaign, some of which I still hold and all of which came from watching this DM and doing the opposite. So I probably feel more strongly about this than is wise or adult.

Still, I’ve managed to split the party several times over the years, and for the most part they’ve been great successes. I am aware of the above problems (even if I think players over-inflate them, that doesn’t mean they don’t exist), and I try to counter them like so:

  1. Divide the party across very specific lines, or let them divide themselves, so the most necessary resources are split appropriately. If I tell the party “you’re splitting into two groups of three”, they will make certain both sides have somebody capable of healing. If I give them more information, like “group A will be protecting the stationary teleport crystal while group B will be sneaking through the catacombs to find the source of the corruption,” they can specialize even further, and usually in a more helpful and character-relevant way than pushing the whole group into one or the other.
  2. Design for the half-group. Often I leave myself some wiggle room in the encounter design. For example, whichever group has the wizard gets a bunch of minions thrown at it while the other gets fewer normal guys, and the group with the fighter gets attacked by her mortal enemy while the other gets a summoned devil. A little change like that gives both groups different encounters while staying within a balanced challenge range.
  3. Switch at meaningful points. If one group is entering combat, I try to time the other group to hit combat at the same time. Then we run a single combat, just spread over two maps, and the only time somebody’s out of action is at the very end when one group has finished. If there’s no combat to be found I jump at dramatic times, the same place I would put a commercial break to increase tension. Every once in a while I have a player run an NPC so they have something to do during another player’s turn, and almost every time it’s amazing. But in general players are mature enough to know not every thing that happens must involved them, and as long as I don’t test their patience they’re unlikely to revolt.
  4. Know what’s happening. Usually I can keep track of all the players and their actions, but if I get lost, I just ask where we were. My players aren’t the sort who would lie to me for a game advantage, and sometimes hearing their version of what’s going on creates a new path for the following scene. The extra cognitive load isn’t so frustrating that it outweighs the joy of giving every character and player a chance to show off, do what they want without putting it through a party vote, and explore the setting in a more personal, dynamic way than they would as a group.

Splitting the party isn’t for everyone, or every session (looking at you, every story game ever). Certain campaigns and DMs handle it better than others. It’s almost always a good idea to stick together in pull-no-punches situations like Lair Assault or the Temple of Elemental Evil. A hardline simulationist DM would never tweak the setting to match the situation; if he’s decided the lieutenant never leaves the general’s side, the druid who stumbles across the general is going to have to deal with the lieutenant because that’s what makes sense. And no matter what, dividing a group takes some tweaking and a certain amount of thinking on your feet beyond a normal session. D&D is a game about parties, it’s written with that assumption, and usually it works best that way.

But it’s not a hard-and-fast rule that the party should never, ever split up, any more than it’s a hard and fast rule that goblins can’t be bards. Some of my favorite sessions have come from splitting the party, sometimes without their consent and sometimes for far longer than they expected, and not once has a player told me it led to a negative play experience. It’s as good a storytelling tool as any, and we need to stop rejecting it out of hand just because it takes a little finesse.

By the way, if you’re in my upcoming campaign, feel free to treat this post as foreshadowing.

* — Unless the DM flat-out says “Guys, I don’t want to / don’t feet like / can’t handle splitting the party right now. I really prefer / suggest / need that you all stick together”. That happens sometimes, and often it’s a legitimate request. If the players ignore that and split up anyway, then it is their fault.

This entry was posted in DMing. Bookmark the permalink.

One Response to Splitting the Party

  1. Kurt says:

    Another reason not to split the party: Lanchester’s Laws. Ignoring party composition issues (i.e. even if all combatants are otherwise indistinguishable and fungible), a fighting group of two combatants is *less* than half as combat effective as a group of four combatants.

Comments are closed.