Flying with a Co-Pilot

The post comes from the "slush pile" – the unwritten pile of articles – from Dwiz over at Knight at the Opera. (You can see my slush pile here).

This would have been a post about the rare practice of running D&D with two GMs. I've heard of people doing it before, but I've never tried it myself (nor do I know anyone who has). It sounds awful tricky, but I keep coming back to it because of some interesting possibilities. -Dwiz

This inspired me to write this post and share my experience because I have tried running D&D with two GMs. This post is going to tell that story, the lessons learned, and why I'm absolutely never going to do it again.

The Pitch

I am a huge fan of Star Wars, as anyone who knows me in real life knows. In my junior year of college, wanting a pause from my normal game, a close friend and I decided to co-run a Star Wars game using a hack of D&D, set during the sequel trilogy that was coming out while I was in college. Could we have used a Star Wars-specific system? Yes. Were we deep in the "why learn a new system when you can hack D&D to do something it wasn't designed for" ethos that still pervades a lot of TTRPG players? Also yes.

The idea was to co-run a campaign by running stories on opposite sides, with players moving fluidly between the groups. I'd run a First Order, "evil" campaign, while my friend would run a campaign about the Resistance. Our sides would mostly be acting independently but in the same area, and so their campaigns would influence each other. And every now and then, we'd truly co-run a session, with both groups coming together and either having to work together against a third party (a classic trope) or racing to accomplish their goal first.

We also wanted the opportunity for characters to betray their faction and defect with valuable secrets. And we had the idea of character death being more impactful, by having players switch sides of their character died, literally weakening your group's ability to go on because of being understaffed... at least until the numbers evened back out.

It sounded like a cool pitch when we started. The hack was mostly in character creation, rather than changing major rules about D&D, so revisions to classes and features slowly turned the originally clumsy classes into something that was actually fun to play.

The Problems of Co-GMing

Information Asymmetry

My co-GM and I would regularly have scheduled meetings to share everything that happened during our session so that we knew where the other group's plot was. We'd co-plan both our individual storylines and any larger crossover components. But GMing involves way more tiny decisions than you think, and certainly more than I'd think to share with the co-GM.

The most noticeable case in point: a character in my group was staying with my faction mostly because they were providing medical care for an ailing and sickly brother. Ultimately, the character felt as though lines were crossed, and he decided to defect, and in negotiating with the other party (secretly), he secured a promise that the other group would help rescue the brother so that he wouldn't be a pawn.

The only issue was that I had, as a throwaway line weeks earlier, mentioned that the medical center was in orbit around a planet. It was not relevant enough to bother reporting, but when the defection happened and my co-GM was building his dungeon crawl, he placed it in an asteroid belt. And suddenly, it was clear that this was not actually in one universe. Continuity error.

It's not game-breaking, certainly, and a solo GM can also make continuity errors. But you're more likely to have these errors when you're co-GMing, at least in the parallel GMing structure that me and my friend were doing. There are so many small decisions that you make while GMing a game like D&D, and often you need to make a call on the fly. These insignificant decisions are not something you'd think to share, but they can prove to be moments of illusion-breaking for your players.

You can also simply not care about these sorts of continuity errors. Dave Clark at Full Moon Storytelling shares his story about "co-DMing"--or, more precisely, handing over your world and playing in someone else's campaign using your world: "The details don’t matter if they don’t matter. Does the now co-DM get the name of a city wrong? That doesn’t matter. Keep playing at the table. Don’t even bother to correct it." That sort of thing would drive me up a wall, but that might be more a me-problem than necessarily a problem for everyone.

More Work

I'm not talking about the situation where we were running the campaign together but running separate sessions. I'm talking about those big crossover sessions more akin to a traditional co-GMing model that Dwiz focuses on, where each person takes the lead at different moments.

But even just having one extra GM is cutting the amount of work you do in half: You take turns narrating. You split up control of monsters in combat. One will work on bookkeeping or prepping the next bit of material while the other is talking. You can better act out a conversation with multiple NPCs. You have twice the creative energy to improvise and worldbuild and tell the story. You can occasionally intentionally separate information from each other to avoid metagaming issues, like by retaining control of NPC factions separately and not informing the other of their plans. You might even split up pretty big roles, like having one be "the rules GM" and the other be "the lore GM." -Dwiz

But in practice, these sessions felt like more work and more mental difficulty.

Yes, you take turns narrating, but often you want to sort of clear your narration with your partner. You don't want to step on their toes; you don't want your narration to cover a topic that your partner was going to focus on, and you don't want to say something that contradicts what your partner said. So you're doing half as much actual narrating, but the time saved by not narrating is still filled with listening to that narration and making sure you're absorbing it fully. You're not saving that much time or effort. Certainly, it would be hard to do bookkeeping or prepping while your partner is talking.

Splitting up control of combat does make it easier and more realistic. Having two-NPC conversations is a lot easier, particularly if you can have some NPCs "belong" to each GM so that you don't get weird personality swaps. Those I fully agree with.

Bouncing ideas off your partner, the whole creative/worldbuilding help component, does happen... but I wouldn't necessarily say it takes less work. It is just a different type of work.

Having separate information to avoid metagaming though, does not entirely work if you're truly co-DMing. If you're alternating narrative duties, for example, to the extent that one of you can focus on bookkeeping, then what happens if someone asks a question about the faction that the co-GM doesn't know about? It's awkward.

Same thing with rules and lore. While having a sort of "rules expert" assistant can be helpful, that sort of division of labor would be clunky in practice.

A potential solution

The solution here is GM assistants, rather than a true co-GM that is equal. An assistant can run monsters. An assistant can be a sounding board for creative ideas. An assistant can be kept in the dark to provide a faction's response to events without seeing the whole picture. An assistant can come in to play a recurring NPC, to help with two-NPC dialogue. That is something that I love doing, and having some sort of "guest star" NPC who can also serve as a sounding board is something that I do in pretty much every campaign. But I wouldn't call them a co-GM.

Simply put, especially in a traditional RPG like D&D where there's an emphasis on GM-as-moderator (as opposed to some more story game RPGs), someone needs to have the ultimate GM authority. There are too many micro-decisions being made to split power fully equally without it muddling things--or at least, without it actually increasing your workload rather than decreasing it.

Dwiz somewhat has hints of this in the section of the article stub.

It's somewhat similar to having a Caller in the party. I would benefit from another set of eyes and ears keeping track of the chaos. -Dwiz

And yeah, I agree, but for an assistant. The Caller is not a co-GM; it's a different role. The caller is not asserting narrative power in the way a GM does in a traditional game. An assistant can serve that role, absolutely. But unlike with co-teaching, where you can trade off the role of leader, doing that in D&D can lead to major stylistic shifts that disrupt the flow of the narrative (which can be good for a classroom, to break up any monotony, but is less good for any sort of consistent narrative).

The tradeoff is that you usually get a good 20-30 minutes to think of how to resolve each player action rather than being under pressure to make an immediate judgment. -Dwiz

Dwiz's comments about Model UN hit home for this. Really co-GMing takes time to assess decisions. You need time to discuss and collaborate. You need time to react, to clear your reaction with everyone involved, and to make sure you're not creating continuity holes. You need time to make sure that your offhand comment isn't going to throw a wrench that you didn't expect into someone else's plans. MUN has this time built in, and D&D does not.

Someone needs to have the GM cap, and if you want a team to help with specific tasks--managing monsters, being a sounding board, helping with complex factional relationships, and avoiding metagaming--bring on an assistant for those tasks.