I am running a game of 5th Edition Dungeons and Dragons.

Due to the players' manipulations of events, it is becoming clear that I will have to run a climactic battle that has three factions- one will be the players, one will be their main antagonist, and the third will be an additional antagonist that is also trying to beat the main villain. All three factions are after a powerful Macguffin that has serious implications for the world. Armies from all three sides will rage, mostly as set dressing for a confrontation among the leaders of the three factions.

I am wondering if there are any best practices when designing and running such an encounter. I don't want to slow down what should be an epic and climactic battle by having a large number of villain turns, and it may feel like if one set of bad guys attacks the other set, that there is a lack of interaction with the players. Does anyone have experience running multi-faction fights? What were the takeaways that you learned?

  • 3
    \$\begingroup\$ The current answer assumes you would force your players into the combat. Is that accurate? If the players notice the second faction going in, would you let them hang back? \$\endgroup\$
    – lucasvw
    Commented Mar 27, 2020 at 11:50
  • 1
    \$\begingroup\$ Also, how many total NPCs would be involved in the fight? If there are only 2 main antagonists and 2 additional antagonists, that's not that different than a regular fight \$\endgroup\$
    – lucasvw
    Commented Mar 27, 2020 at 12:15
  • \$\begingroup\$ @lucasvw It is different. They can fight 2x2 while PCs are just watching. \$\endgroup\$
    – enkryptor
    Commented Mar 27, 2020 at 14:13
  • \$\begingroup\$ @lucasvw Not sure if I should edit to add this to the original question- I say "I will run" to indicate that the players have set themselves on a course that has resulted in them fighting a three way battle. As suggested below, this fight has a goal other than "kill everyone". Of course, players are allowed to engage (or not engage) in this fight however they see fit, but not intervening means one of two very bad people has the potential to get the All Powerful Macguffin. Does that help answer your clarifying question? \$\endgroup\$
    – sillyputty
    Commented Mar 27, 2020 at 16:44
  • 1
    \$\begingroup\$ Also, there are scores if baddie on both sides, but I will take the approach of one of the answers below and make armies clashing setting, rather than actual minis on the board, as I have done in other large scale fights. \$\endgroup\$
    – sillyputty
    Commented Mar 27, 2020 at 16:45

3 Answers 3


Get a second GM

I have been in a game where there was a three-way fight. One team was the players' party. Our GM ran the second team. And our GM brought in her partner (also an experienced GM) to run the third team.

There are a couple of benefits to having another person run the third team. One major benefit is that it reduces the load on you as the GM since you only have to control half as many NPCs. This helps streamline combat somewhat.

Another very important benefit is that this third team will be able to act truly independently. If you run both teams, you will be subconsciously (or consciously) colluding with yourself and the three-way battle won't feel as authentic. The different play-style of your assistant GM will be evident to your players and this extra set of NPCs will feel different to if you had run them. And by having another person running the third team, there will be the added thrill of all three sides genuinely fighting to win.

Have rich goals

This fight will be long and slow, even with a second GM. As such, you need to make the fight narratively rewarding and more engaging than 'I hit them until they die'. Enkryptor was spot-on when he said that every combat should have a goal.

In my particular example, the goal of all three parties was to 'get the girl', a plot-critical NPC who was a long-time member of our party (so we were protecting her while the other parties both tried to take her for themselves). Because we had a goal which was more complicated than simply 'kill everything' the combat was more dynamic and exciting. There was the added complication that no side wanted the girl to die, even though the girl was quite capable of fighting back, so extra care had to be taken. It was a goal that we the players were quite invested in, which motivated us through the battle. And it meant that some NPCs fled when heavily wounded and left their underlings take care of the mission, which made the battle more authentic than people pointlessly fighting to the death.

The objectives of the three factions should also be such that there can be only one winner. If, as in your example, the goal of both the party and the secondary antagonist is to kill the main antagonist, then the most logical arrangement is for the secondary antagonist to temporarily side with the players, reducing the battle to a two-sided one. However, in my scenario, only one team could get the girl, so we all fought tooth and nail to make sure it was ours. There could be no parley, not without losing her.

Allow lots of time

This battle will take an entire session. It will likely be a long session. Plan accordingly. You should end the previous session just as you've set up the battle and combat seems inevitable, even if that means ending early, otherwise you will not have time for your planned battle. This also builds up the hype for your players.

Abstract away unimportant creatures

In such a large battle, you are likely to have mooks fighting mooks. However, your players don't care about the mooks, they care about themselves and the major NPCs.

You can substantially reduce the number of active characters in the encounter while maintaining the scale of the battle by going "this group of soldiers is fighting this other group of soldiers and they are both busy doing that" with an implicit "they won't bother you if you don't bother them". The outcome can either be determined as is narratively convenient or with a few quick dice rolls, although the result should be withheld until either the end of the battle or when it becomes relevant. This puts greater emphasis on the characters which matter while keeping a cinematic feel to your epic battle.

(This advice applies to all battles with many NPCs, not just three-sided ones.)


I recommend getting a second GM to run the extra faction. I recommend that each team have mutually exclusive goals which means that they all must fight against each other, and that these goals should be more elaborate than "I want everyone else to die".

If none of these options work for you, you may need to reconsider doing a true three-way battle.

You might want to run the NPC vs NPC battle beforehand and let the players intervene when they see fit (if it is appropriate for the players to ambush the NPCs).

You might have the players fight just the main antagonist, then when the battle is almost over the secondary antagonist swoops in and ambushes both parties. This makes it a two phase battle. Because one NPC team will be mostly (or entirely) dead by this point, the second phase would run quite similarly to a normal two-sided battle.

Or, if the narrative allows, you might have the secondary antagonist make a temporary truce with the players. Then the battle will be reduced to two factions and end up being simpler to run.


Don't do that

There is a couple of reasons why this probably won't go well.

I am going to run a climactic battle that has three factions- one will be the players, one will be their main antagonist, and the third will be an additional antagonist

It seems you've already established an important detail: players will engage the fight between their foes. Although it won't be very smart in such a "free for all" skirmish — it's probably better to let the foes kill each other first, then finish the survivors — you don't allow such an option for players. That means you're railroading them, forcing them into the fight, thus taking away their agency. This alone is probably not a good thing.

it may feel like if one set of bad guys attacks the other set, that there is a lack of interaction with the players

It is indeed. There will be several people just waiting at the table, wasting their time (real time, not game time), while the DM is literally playing with himself. Combat in D&D can be notoriously slow, there is no need to make it even slower.

Normally a DM shouldn't ever play NPC vs. NPC combat. Your game should be a story about players, not a story about main antagonist fighting another one. It might be a decent plot for a book, but unfortunately works poorly with a TTRPG.

What you can do

If you want to experience a full-featured combat between NPCs, you can run it before the session. Then just narrate it to the players and ask, if (when) they want to intervene. That helps you to save a lot of players' time.

Give your players a possibility to find their own approach to solving in-game problems. Do not think for them, forcing the story to go a particular way. I know, the final fight is supposed to be epic and climatic, but still. Players are more important than story, they will be grateful, believe me.

Also, remember that every combat should have a goal, especially a combat between organized groups. Yes they can fight till death just because they fight if you say so, and this paradigm works okay in computer games, but a TTRPG combat should be a little more meaningful. Placing other objectives which are not "kill all enemies" is a good idea.

  • \$\begingroup\$ First, thank you for taking the time to answer this so thoroughly. Were I a new GM, or if this battle were my plan from the start, I would absolutely agree with you. But this scenario has been brought about through direct action by the players. I am hoping to find ways to run this scenario in an interesting and engaging way. Your advice should absolutely be kept here for others, but the specifics of my situation, I think, require something else. \$\endgroup\$
    – sillyputty
    Commented Mar 27, 2020 at 17:08
  • \$\begingroup\$ @sillyputty no problem. Anyways, that's how I understand the situation based on the question text. You probably want to edit the question a bit and add this important detail. \$\endgroup\$
    – enkryptor
    Commented Mar 27, 2020 at 17:39

Engage the PCs first, then add the second antagonist.

This prevents players from just waiting out until both antagonists have nearly killed each other. PCs could then just finish them off. It also feels less railroad-y for players, the second antagonist was just waiting for their chance.

Antagonists should assume PCs are also enemies.

Your big baddie should position Fireball to hit both the opposing antagonist as well as the PCs. This creates danger from both antagonists to PCs, as antagonists are damaging them as well.

Allow for short-lived truces.

During the fight, the weaker antagonist can ask PCs for favors to help beat the stronger one. They can also double-cross PCs after they have gained the upper hand. If both are low, they can also ally against the party for extra fun.

Don't worry too much about having more turns than usual. Just make them memorable.

If turns are creating memorable experiences, then they are fine. These include spectacular attacks (dangerous), terrain-shaping effects (lead to forced movement) and strategic effects (if one antagonist goes invisible and hides, the party either has to snuff them out or focus on the other).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .