Several times I've had experiences where I was running a fairly standard campaign, and I got a player who wanted to, essentially, watch the world burn.

  1. I had a character who offered up his life force to a dark god in order to release the god and destroy the world. (D&D-3.5e)
  2. I had a character who was supposed to be protecting a key NPC, but instead took her hostage against the villain; when the villain didn't back down fast enough, the character murdered the hostage "to show him I was serious". (Pathfinder)
  3. I had a character (in an undead horror campaign) who found a book of foul necromancy; the rest of the group wanted to burn it, but he told us he wanted to keep it and become a necromancer, and he was willing to leave the group in order to do so. (D&D-5e)

I've never been sure how to handle these. The first two times, I invited the party to vote on whether the thing would happen; the party voted to allow it, which ended one campaign and somewhat damaged the other.

The last time, with the book of necromancy, I straight-up told the player that he couldn't do that because it wouldn't be fun for the group. He told me he was doing it anyway, so I asked him to leave my table.

We've seen discussions of this sort of player in other questions (1, 2).

My goal is to make sure everyone at my table has a good experience. That includes this player, but it also includes all my other players, and it includes me as well.

In your experience, what's the right way to handle these sorts of issues?


11 Answers 11


From Experience:

I have players playing evil characters, both now and in the past, and I've played an evil character a few times as well, like every time I've ever played a Drow. (That's right, those characters were enforcing the stereotype from the monster manual!)

When I discover the player wants to play an evil character, I give them a simple choice:

Your evil character's motivations must work with the rest of the party OR you don't get to play an evil character.

If your character oversteps bounds or decides to run counter to the party's efforts, they will become an NPC, and you will need a new character.

That's the deal, and part of the social contract at my tables.

Here is some good advice- successful evil characters need good motivations, have a good goal, but are not adverse to using the wrong methods. They need to be valuable enough to the party that the party can look the other way.

Moustache-twirlingly evil characters is most easily solved by direct talking about it with the group. Is this play fun? Is this what we want? Can the evil character scale back the evil-for-evil's sake and get a goal or something to work towards, or know that those evil actions have consequences?

I'm also surprised that no characters (and therefore players) have indicated that (1) this is terrible, (2) how long will it be until he turns on us, and (3) their adventurers are out to kill monsters like this...

An Ounce of Prevention:

For those starting or about to start a table, the same page tool, or anything of its kind, can avoid this kind of thing. Even a simple statement at the start of session #1 could help.

  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. \$\endgroup\$
    – mxyzplk
    Commented Jul 7, 2017 at 3:40
  • 1
    \$\begingroup\$ I feel this answer isn't addressing the core issue here. You've listed some points about playing evil characters, but I think the heart of the question here is a sudden change of intent in a player, veering towards chaotic and/or evil actions. \$\endgroup\$
    – YogoZuno
    Commented Sep 10, 2018 at 21:16

This is one of the classic RPG motivations; in my class I call it Irresponsibility: Recklessly pushing buttons, opening doors, or antagonizing NPC's without real-world consequences. To some extent, this is present in most roleplayers - you and I don't go out into the wilderness and fight dangerous animals, nor would most sensible people, but RPGs are partly about being able to take risks and see what exciting things happen while simultaneously being safe in someone's basement.

However, as you've observed, this can be problematic when the players are looking for different things. In my experience, the best way to fix this is to prevent it by setting expectations at the start of the campaign. The most explicit way I've done so was right in the invitation:

What will the game NOT be like?


  • Lawless, where people do needlessly dangerous, violent, or obviously ill-advised things just to see what will happen or because they’re bored. This can significantly hamper the group’s goals and make the game less fun for those trying to accomplish them.
  • Strict about “realism”. [This game] operates mainly on the Rule of Cool; if you want to try something, and it would advance the party’s goals in a fun way, you can give it a shot, and I won’t be mean about the results.

Part of the reason I was so specific is that this was a Star Wars game - I couldn't just say "don't do anything stupid" without sacrificing the sort of campy, swashbuckling atmosphere we were going for. I settled on whether or not it advances the group's goals as the defining line between someone trying to burn the world down (killing NPCs the party needs, snitching to the Imps) and someone having fun with the genre (charging a corridor full of stormtroopers, launching a landspeeder off a ramp during a chase scene). The result was that everyone had jolly fun together with no real arguments over play style, because everyone knew what to expect going in, despite the fact that everyone was new to the system and many were new to RPGs altogether.

Note that there are games where becoming a necromancer or unleashing a dark god would be an interesting story development everyone was on board with - but it's fine to say you don't want yours to be one of them, and let people self-select out if that's really what they insist on.

On the other hand, Tim Grant makes some excellent points about the circumstances. Why would a party vote to unleash a dark god to destroy the world? Sounds like they may have been jonesing for a change. Whether or not you place restrictions like I've suggested, make sure the players have real choices to make and opportunities to influence the game world within whatever parameters you've agreed on, and if you're presenting a choice, try to think through all the options rather than assuming people will make the choice you want.


There’s a pattern here

“Bad apple” disruptive players pop up once in a while, but you’ve noticed a pattern where you get this a lot — per your figures, a couple times a year. While you do need to better figure out what to do when these things happen, you also need to figure out the underlying causes.

I often advise DM’s to feel free to jettison problematic players, but many players can be occasionally disruptive in this way, and disinviting them all could empty out your table.

Your other players

The fact that your other players had chances to stop these actions and declined to do so is a big hint that you may need to adjust something about your game-mastering. If they felt the player in question was just a jerk trying to ruin the game that they were enjoying, they would have put a stop to it.

A majority of the other players felt these actions were on some level reasonable. Some may also have thought they could have shaken things up for the better.

Note, I’m choosing my words, “disruptive actions” carefully here. You are seeing character actions that would disrupt your story and your game world. This is different that “disruptive behavior” by a player.

A “Contract” may treat the symptom, but not the cause

Proscriptive contracts recommended by some other answers can “put a lid on” the disruptions — but won’t solve any underlying discontent. You may just get different problems, such as player turnover, absenteeism, or inattentiveness.

Boredom and Frustration

If a player isn’t having fun, he might do something random and destructive just to mix things up. That player who wanted to “destroy the world” and his co-players who voted to allow it might have been getting tired of your campaign world or the current plot, and wanted either to cause big changes to your campaign, or end it.

Bored players act out especially when the action has slowed down, when they haven’t had a lot of time to do “their thing,” or when their attempts to influence the story or gameplay have been stymied. If the first real choice the players have to make is whether to let the demon out, that’s when they may show a little rebellion.

Accepting Little Changes

If your players feel railroaded, they may jump at any chance to affect the story or your game world. Are your players typically confronted with the choice or doing what you want, or letting the world come to an end? If so, be ready for the other choice.

You can usually avoid players pushing for a big change, by allowing little ones. Ask yourself if your players regularly have the opportunity to shape the play in a meaningful way. Do you choose the NPC’s they have to interact with, the places they have to go, etc.?

If you allows PC’s to affect little things, they will be less eager to jump at any opportunity to affect the world.

Accepting Big Changes

You ended one campaign, and considered the other “somewhat damaged” after a player acted contrary to your plot. The other players accepted their actions — clearly, they were OK with a big change.

You didn’t have to end your one campaign, and considering the other damaged is only your opinion. Did you ever consider how the campaign might have continued with an evil god loose in the world? Maybe your players would have dug that setting.

Expecting the Unexpected

Whatever the cause of these issues, you are seeing them. To keep from getting caught flat-footed, just prepare a little more.

So the party finds some evil artifact. Well, what does happen if a non-initiated person tries to use it? Decide ahead of game time. You don’t need to allow PC’s to bring your world to an end. An evil artifact might do something dreadful to the PC, for whatever reason — it’s an evil artifact. And that could include the character becoming dominated and becoming an NPC.

Just be ready for the story to continue.

  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. \$\endgroup\$
    – mxyzplk
    Commented Jul 7, 2017 at 3:40

tl;dr When life presents you with lemons, make lemonade

All of the things you mentioned can be worked into the game to make it a be tter game. That's what makes role playing feel alive and appealing. Try to find a way to incorporate those things; turn the problems into flavor.

Railroading is a valid game style, and some players even prefer it for various reasons, but that should be the exception not the rule. It doesn't sound like that's the case with your players.

I have edited this section since OP let us know that yes, the necromancy tome would unbalance their world and make it broken-powerful. See also this question about how to deal with a game where you accidentally allowed the players to acquire something that breaks the balance for them.

What is the real problem?

You did not say necromancer players would be difficult to allow in your setting and would make play impossible, so I'm going with the assumption that you just don't want it. The only reason I can think of to specifically disallow a necromancer is because having a mob of skeletons or zombies around you would be difficult to explain to the town guard.

Other than that, the rest of the abilities aren't as much of an issue. "Wow, you beat that goblin good; what was that technique?"
"Oh, just a battle spell I picked up along the way."
And what about when interacting with the dead? That's just flavor right there, especially if the player needs to try to do it discretely.

I have both DM'ed games with necromancers and I have played a necromancer. Sometimes the necromancers are not truly evil, but their moral standards are just sub-par. Think of grave-robbers, trying to benefit from the corpses in the graveyard because they are easy targets; a neutral necromancer could have the same mentality.

And for the first point, about the "I want to release the world-destroying evil," that one sounds like a dork move on the player's part, but it should not ruin the campaign. If it took a team of heroes a campaign of adventure to keep the evil from releasing, how is it so easy for this 1 player's character? Your vote to "allow" I assume means "Do you want to tell Bob he can't do that?" (that is, out-of-character).

In character, the rest of the party is likely to save the world again, from their own member; that's an interesting game right there. I have been part of games like that, and it adds flavor.

All these "problems" add flavor

More on necromancers... because they are cool (in games, anyway)

When you need to be discrete about your necromantic tendencies, sometimes it gets interesting. I once had to take only a small portion of my undead mob with me, only as many as I could cast disguise spells on and pass off as living to the commoner NPCs. Having a bunch of undead minions was necessary for what was going on as we needed many hands, yet we had to get by without having the town folk pitchfork us.

Also, having fewer undead in number that are higher level makes it easier at times. But sometimes it is just way cool to unleash a horde of minions, even if they are weak.

And back to the "Literally destroy everything by unleashing powerful evil"

I have DM'ed games with evil characters before, though none where they wanted to destroy everything. Even if you failed to stop the player from succeeding, there are still ways you can save your world.

  1. Maybe the power being unleashed is not so powerful that it can destroy the world instantly and there is time to try and find a way to stop it. (How was the evil locked up in the first place? Do that again.)

  2. Maybe the knowledge about what would happen if this evil were unleashed was not completely accurate (even if the evil entity itself said some of the things).

As a DM you need to metaphorically make a dodge roll so your world can take half-damage.*

Characters like lots of power too

I once DM'ed an adventure where the party went to a disaster zone to help people and to secure some magic that evil forces wanted. Among the ancient treasures that needed to be kept away from the evil forces was one major artifact that was powerful beyond the others. I knew the players and the characters well and trusted them to handle it accordingly.

The party did indeed save the day, beat the enemies, and acquire the powerful items. I assumed they would give the major artifact to one of the available magic organizations that could protect it, as it was a dangerous item the kind most sane people would not want to have in their possession. But one player character, a sorcerer, was so drawn to its power that he wanted it for himself.

In the final encounter the sorcerer teleported past the fight when just before it finished and took the artifact. There was little the rest of the party could do to stop him, as the sorcerer had more spell power left than the others. In the end, the rest of the party convinced him to let it go, that it was not safe. They all agreed that it needed to be destroyed completely, so that is what they did.

One player told me later on that that was the best adventure that he had ever played, for multiple reasons, the realistic party dynamics among them.

Now, even if the sorcerer had managed to take off with the major artifact, that still would not have ruined the game. This was a large game that two of us were DM'ing together, a vast, living world with dozens of player characters in it ranging from 1st level to epic level characters. The sorcerer was high level, so having a major artifact would have increased his power probably by 2 or 3 times but not broken the character.

During the dialog between the party members in the tense moments before the sorcerer gave in, I was seriously intrigued by the ramifications of him running away with the artifact. Even though the sorcerer could have been a few times more powerful with the artifact, it would have cut him off from the rest of his party, making him a party of 1 character that was effectively less powerful than the complete party he was with before.

Although I think it would have been fun to deal with the sorcerer had he left the group, I do admit there are plenty of facets to consider. Does this mean the game you are running gets split into two games? For the world I was DM'ing, that would not have been an issue since there were already multiple separate parties of characters which were changing and evolving over time anyway as characters shifted between adventures and parties. You need to answer that for your group.

For the normal campaign, this would be more difficult, but not without options. Could you separate a bad apple from the group and make that character a new villain? The DM does not always need to be in charge of the villains. This is my first choice in this situation. If you do this, you do not necessarily need to get rid of the previous villain(s) (assuming they haven't already been defeated by the party before this problem arises); there can be multiple villains, and villains can even be at odds with each other, or they can be allies.

Imagine a game where you do not need to set up the traps and enemy placements on your map because you have someone to do it for you? You generate the level appropriate list and just hand it to the player-turned-enemy to populate the map with them because he's in charge of it.

You could even let him control the enemy mini-bosses, or even all of the enemies if you wanted. Then you can concentrate on other DM tasks more. In this situation, the enemy player must understand fully ahead of time that the odds are being stacked slightly against the enemies, so that the enemy player's forces will likely be defeated; the player has to be OK with this.


If you don't mind splitting the game up into two sub-games, that works too. The point is, there are options. Even for the dork-move where your player wanted to unleash the world-destroying evil, there are still options.

RP DM'ing is not a "DM vs players" thing, however, the DM is controlling the world and so sometimes needs to be the world's player to save it and be crafty and scheming for it the same way the players are for their characters.

Sometimes saying "No" is necessary, like when you get the new player who you thought you were being nice to by letting him start at level 3 to match the existing players, and he responds with "And my guy is the son of the king and rides a huge fire breathing dragon!" Generally though, as long as the players are being reasonable, trying to work around the player's ambitions and fun is more interesting for everyone, and more fun for the players, than simply saying "No." And it sounds like your players were being reasonable, with the possible exception of "I suddenly do a 180 and want to unleash the world-destroying evil," though even that can be fun to incorporate.

If you cannot think of what those options are right at that moment in the game, sometimes you need to take a break. If you've been playing a while that day, call it the end of that session, even if it's early. If not, at least take a short break for restroom use, or for snacks, or whatever, just enough to give you 5 or 10 minutes to find your world's escape route.

Here are a some ideas for your first and third examples:

"I want to unleash the power that destroys the world."

Ok, go ahead. Oh wait, another party of heroes (NPCs) just walked in the door after you "saved the day" but changed your mind. They were on a similar quest as you but just a step behind. Now they are saving the world from you.

Yes, the previous one is a bit lame. Let's do a better one...

The evil force awakens, and the forces that held it captive all this time slowly fade. The dark being takes notice of this too and is enraged that its freedom is not immediate. It appears that the release will take some time.

And then you could have it take days, weeks, even months of in-game time. Or speed up the release but say that the entity does not yet possess its full power, so it can still be stopped. This leaves another session for the players (minus the dork who sacrificed their self to release the evil) to deal with it later.

For the necromantic tome, if it is so over-powering then that means others will want it to. Once word gets out that this player has it, suddenly the party will be on the defensive instead of the offensive. That player character is likely to be besieged by many creatures more powerful than their self, creatures they did not even know existed before.

Or the tome could have negative affects. It can be as simple as "the negative energy of the tome causes you to lose 2 levels," or it can be more intricate like every time the tome is used it attracts dark-energy monsters who attack its user - what a nuisance! Or maybe it constantly drains the user's energy while in use, so although it is overpowering it can only be called on sparingly without killing its user.

If you simply cannot come up with any ideas on how to allow something in your game, and it would break your game, then you can always fall back on "Hey guys, I simply cannot figure out how to work this into the game, and I think the game would break. As much as I want to respect your characters, I just don't think I can make the world work after that change. Could we step away from that just so we don't have to lose our game?" But that should be a last resort. If it does come to that, the voting (which you already did) is a good addition.

  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. \$\endgroup\$
    – mxyzplk
    Commented Jul 7, 2017 at 3:40

If a player doesn't want to play the same game as everybody else, they shouldn't play it. It is as simple as that.

That type of player tends to throw out unexpected, campaign wrecking actions are a high rate forcing the GM to spend their time compensating until the game collapses. This spoils everyone else's fun.

The Same Page Tool can help avoid this problem cropping up in the first place, but when it is too late for that…

There are three basic stages to dealing with them.

  1. Convince them to play the game everyone else wants to play
  2. Convince them to agree that the game you want to run isn't the game they want to play and work with you to let their character exit gracefully (and then stop playing)
  3. Kick them out

In my experience (roleplaying since about 1980, mostly not D&D but mostly The Fantasy Trip and GURPS, mostly un-scripted gritty simulationist sandbox games), the "right" way (your question) is to consider the game being played, and the players, and determine whether a player's stated actions are appropriate for the type of game I'm running, or not.

You and many of the other answers here seem to assume a typical modern cooperative group focus is desired, but I notice that you wrote:

I've never been sure how to handle these. The first two times, I invited the party to vote on whether the thing would happen; the party voted to allow it, which ended one campaign and somewhat damaged the other.

I emphasize that your players actually don't seem to have agreed that requiring cooperative behavior was what they wanted, at least at the point it came up.

Some players are interested in betrayals, surprising behavior, freedom of action, dark actions, or whatever. Where the line is drawn can take many forms, and I think it's really up to the GM to determine where to draw that line. And when over that line, to direct players appropriately, as needed.

I am clear about where I would typically draw that line, which for my typical playstyle is usually that players are allowed to have their characters do wild things that are in-character, but not to do out-of-character random actions. If a betrayal or strange or evil action is actually in-character (and all character traits are subject to GM approval), then it can actually be very interesting when a player is roleplaying wild behavior in ways that make sense, and I tend to prefer that to not-in-character party cooperation for the sake of cooperation.

If a player tries to run their PC too out-of-character, I intervene and ask them where they're coming from, and correct appropriately. There are a number of ways it may make sense to do this, depending on the player and what might be going on with them. The player may be new to roleplaying, or the type of play I run, or bored or misbehaving or tired or drunk, or sick of their character, or just having a lapse and not thinking in-character. What I ask and say to them will be based on what I can figure out about that, but may include asking them in private to explain the way they are roleplaying, or asking them to remember what their PC is like, pointing out considerations I think the PC would have, asking them to roleplay in-character, describing the situation vividly so it brings them into the situation, explaining to them why I question or am overruling their action, and/or asking if they want to have that PC become an NPC.

If the specific situations you mentioned came up in valid in-character ways, the way I'd handle your specific examples might be:

I had a character who offered up his life force to a dark god in order to release the god and destroy the world.

(After establishing the PC really is suicidal and wants the world to end, or this otherwise makes some sense.)

Ok, I'd consult the magic rules I were using for the situation, and play it out. The PC sounds liable to die unless the other PCs stop him somehow. (I do tend to avoid having situations where my game world is liable to be destroyed, in the first place.)

I had a character who was supposed to be protecting a key NPC, but instead took her hostage against the villain; when the villain didn't back down fast enough, the character murdered the hostage "to show him I was serious".

(After establishing this PC really would think this way, do this, follow through with it even with the helpless hostage, etc.)

If this is appropriate for a PC, that PC's a bit of an awful and/or foolish person in one or many ways. Or if the character isn't well-defined and the player is inexperienced and being foolish, and the other players as you say are into having this happen, then I'd let it happen. I'd have any NPCs including the villain behave appropriately, such as have them ask the PC what he's playing at, and if he's just being a murderous fool, treat him with the contempt he deserves. If the villain could, he might kill the PC. If the hostage had family or friends and they heard about it, they'd probably despise the PC and want the PC killed or imprisoned. I would tend to expect the non-murderous-idiot PCs to turn on the PC, or at the very least, take extreme measures to straighten him out. If the story got out and the PCs were still pal-ing around with that PC, they'd probably be treated as a band of murderous thugs. Non-evil PC allies who found out would probably turn on the PCs, etc.

I had a character (in an undead horror campaign) who found a book of foul necromancy; the rest of the group wanted to burn it, but he told us he wanted to keep it and become a necromancer, and he was willing to leave the group in order to do so.

(After establishing the PC really would do this.)

I've GM'd (and played in) many groups over the years with characters that get tempted by magic that other players think should be destroyed, is evil, etc. I tend to think this makes for interesting roleplaying, whether it is covert, overt, or involves splitting the group, having the PC become a party adversary and/or NPC, or arguments or PC vs PC combat. Personally, I'd much rather play in a game where this is a possibility and has serious actual in-play consequences, that one where an out-of-character meta-game consideration for "the group should get along and stay together" generically makes that not happen for no in-character reason. It does (or has, in the cases I've seen) tend to lead to such PCs leaving the group and/or deadly combat with the other PCs, but that's interesting to me as long as it makes sense and is in-character for characters approved by the GM.


There are three broad stroke solutions to this:

1) Talk to the player, then eject the player. You've done that. (This is my preferred solution, by the way. All three of those scenarios either are, or approach, vandalism.)

I consider "setting expectations beforehand" to be a subset of this, because the logical followup is still expulsion.

2) Let the game world run as it will in consequence. Remove aspects of script immunity. You've done that. (This is my second preferred method for minor cases. I find it turns into a battle of wills in a significant fraction of cases, and that is not fun for me the GM.)

3) Don't put game-breaking options in the game for your player to break the game with. (I don't do this. This can also be described as "child-proofing the game," and is nearly impossible against a dedicated vandal.)

The bottom line is that you can't make a player behave.


My goal is to make sure everyone at my table has a good experience.

While an admirable goal, it's probably not a realistic one. Not everyone enjoys the same style of play and clearly you've found several that would rather watch the world burn than save it.

Rather than wait until you are several sessions deep into a campaign, ideally you should find out what motivates the players and whether or not they will foster teamwork and good game-play before your first session. If you clearly have an outlier, then perhaps this isn't the gaming group for that person.


What I have done in my game is just let them do it, but amplify the consequences. Although I haven't really had to do deal with these scenarios that much.

An example of one scenario I did have to deal with though, was when the ranger wanted to stop for breakfast at a cafe when they were supposed to be defending the town. I said alright, you can stop at a cafe, but it will take you 3 extra rounds to reach the gate. The rest of the party reached the gate, but almost got killed by the time the rogue got there.

I probably would have let them release the god, and instantly say "Alright, the god destroys the world, at this point I would go into details describing the deaths of all NPCs the players ever met and the destruction of all the cities they ever went to and say congradulations, you all just failed all the goals you were ever given in my campaign and betrayed the trust of every NPC you met because of [insert player name]" then pack up my stuff. This would be said very calmly, not as a rant or tantrum but simply be stated as fact, with the hope of making the bad player realize the impact of their actions or at least make the other players stop them from doing something like that in the future.


Whether you choose to work it into the campaign, rework the campaign, stonewall or negotiate, please try to make it work in-game, between the PCs rather than the players. NPCs can also act on it, and there are many of them.
More roleplaying, less bureaucracy.

The offending players might find themselves dealt with like any other threat - in prison if possible. Reform or reroll. THAT's the threat, not scorn/expulsion from the game. Or they might continue to be a part of the game, just taking over the some of your (DM's) duties - see Aaron's answer.

Notice how modern reboots/remakes of classics (comic book / cartoon characters, etc) get more drama? Emotions? Friction? The party are not a hive mind - unless you've used the Same Page Tool to verify that they are - if so, just assign an encounter XP value to the obviously possessed offender. That is likely to result in fun times for all :).

DM is god, yes, but gods of rails are a niche thing.


While the concept of making the player conform to the game willingly is certainly a good fix, there's no reason you should've ever let this tool ruin your campaigns. To paraphrase Thackeray, GM is the name for GOD in the lips and hearts of all players. The GM puts the Deus in the gorram Machina. Crush. Him. As someone else said, "evil actions have consequences". Oh you want to become a necromancer? Well, someone hires the party to hunt you down. If that doesn't interest them, there's a militant group of paladins who are up for the job. Good luck, hoss. Show him there's a difference between outside-the-box thinking and out-of-bounds. If his behavior breaks the campaign, break the player instead. Sure, he'll likely leave, but that was already your solution. This at least gives him a chance to play like a big boy.


You must log in to answer this question.

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