What to do when a game upsets someone - safety tool failures, and what to do next

There was a great conversation here two years ago, with people sharing difficult moments in games they have experienced:

What happens when, in your game, some upsetting content comes up, and a player is deeply affected?

Perhaps it is a “safety tool failure”, perhaps just an honest miscommunication, but, in either case, someone is hurt or upset. For the sake of this conversation, let’s assume that everyone means well: all the players are at least trying their best to care for each other and willing to make things better.

Safety tools helps us flag or avoid these situations, and we have so many now. It’s a wonderful improvement on a play culture, where people are starting to understand that the players are always more important than the game, and it’s spreading. Here are some local resources, for instance:

But what happens when you’re past that point?

What happens when that line has been crossed, the safety tools, such as they are, have failed or weren’t implemented, and now you must figure out where to go next?

What are your best resources or tips for handling the repercussions, handling the fallout, and repairing those situations? What are some good strategies or some great ways to handle such situations, once they have occurred?


For me, I do as with a wound. I want the wounded to gather their attention and assess the damage, feeling from the zone around to the wound.
At the scale of a game, this means letting the topic rest, then alluding to it, probably half-jokingly, threading lightly, and see when the topic is open. Then asking about each other’s experience using NVC.
I know half jokes are clearly not NVC, but if we are going to work on the wound, there must be some room for it without constant opportunities for retraction and resentment. This is very much like my process for calibration in “most useful safety tools” thread.

It’s my best method, and it has 2 big limitations : 1° in case of a bad behaviour, beating around the bush leads to a refuge in bad faith, when jumping at the conflict while it’s fresh is an opportunity for a flagrante delicto 2° one actual “play wound” is 3 months old now, and it’s still unaddressed, a clear example of resolution by decay.

I suppose thorn and roses would be more useful.


Thank, DeReel. But what is NVC?


Sorry, it’s Non Violent Communication. I mean it as a general principle, not as a brand of self help linguo. Like, expressing your feelings and perspective without making “mind reading” or “The Truth” assumptions about others.


There’s a book called Crucial Conversations that I would recommend on this topic. I’ll try to summarize some of their advice here.

One of the first things is that you have to discover that someone is upset. When they feel uncomfortable people tend to either react with (non-literal) “silence or violence”. I’ve found that the people who get quiet and withdraw are more common in my social circles, and also pose the greater risk of not noticing the problem and continuing to exacerbate it. Thus, it’s important to pay attention for when someone is participating less than their “normal”.

This is even harder with people you don’t know well (i.e. for one-shot games). In those cases, you may need to periodically check in. People in a “silence” reaction are more likely to answer check-in questions with short answers, and non-confrontational people will sometimes say that everything is fine if asked a direct yes/no question, especially if directed to the group. (Such as, “Is everyone doing OK?”) Thus, it’s best to ask open-ended questions. (“Is there anything you’d rather see less of?”) Often, starting a narrative answer can remove the block to talking, but not always.

I’ll pause there to avoid being long-winded, but am happy to elaborate further as the topic develops.

(I also have a few war stories, but I think the intent was to add tips / tricks / advice here, and keep the stories out. If that wasn’t the intent, I’m happy to share.)


I conducted anonymous research about people’s needs in emotional trouble. Around 200 young adults filled it out. The results are not representative but might be somewhat relevant here.

My main finding was that the two best things you could do to people who are sharing their personal problem with you are:

  1. Staying nonjudgmental about their issue (around 90% said they need this)

  2. Actively listening to them thoroughly and carefully, without any interjection (nearly 80% needed this)

Anything else (hugging, asking questions, saying it will be OK, giving advice, etc.) was near or under 50% of acceptance. These might be helpful for some, but generally not a good idea without an explicit request from them.

Of course, my whole research presumed that the person does actively share their problem with you, which is a serious limitation.


Wonderful answers, thank you. Great advice there on both counts!

I think that “war stories”, in terms of what went wrong and how, are better suited to another thread (like the one I linked above). However, stories about how someone handled the fallout/consequences, and how that went (or even some useful lessons or takeaways) would be fantastic for this thread. So, please do share if you have something like that!


Well… I’m not sure I have good advice, but let me highlight a failure and try to learn from it.

I ran a long-ish running Feng Shui game at an FLGS. I had a player who loved her character, to the point where she was writing fiction about him offline, incorporating him into a computer game script, etc. And then, after >1 year of play, he was killed in a climactic encounter. She was emotional, and I felt terrible.

Things I think I did well: I did pause the game at the time. I did offer to the group and the player the option of doing a retcon, etc. As far as I know, the player did not hold it against me and we are still friends, although less in touch. We did have a framework we were working within, having had an early conversation about script immunity vs “letting the dice fall where they may” at the start of the game.

Things I did poorly: guilt and shame. My emotions about the failure, not the player’s, were the reason that this incident killed both a year plus campaign and another unrelated game I was running. I made it about me when it should have been about the person whose emotional safety was violated.

Things I wish I’d done differently:

  • I wish I’d highlighted the additional risk of the climactic battle, and the potential that PCs could be killed, so people could be more emotionally prepared.
  • I wish I’d trusted the player to let me know what was needed, rather than assume what was going on in their head / heart. While they were upset in the moment, my impression was they mostly got over it by the next session, and I didn’t.
  • I wish I’d been more conscious of the escalating power levels in the game: throwing bigger and bigger threats made the game more “swingy” to try to create some sense of danger.
  • I wish I had been able to express that things had changed since the original conversation about script immunity, in terms of character investment. The player felt bound by a script immunity decision / commitment that had been made over a year before, with a brand new character as opposed to one they were very attached to. They expressed worry about ruining the game for everyone if they went back on their previous commitment to let the dice results stand, despite everyone at the table saying they would be OK with it.

Hope this is helpful. I’ll probably still be second-guessing this one for a while, and others advice or thoughts are welcome (although in transparency, it’s still a little raw, even a couple years later).


That’s a really frank and really thoughtful response, thank you! It’s lovely to hear about such stories, no matter how difficult they might be. Thank you!


One of my students’ character died in our OSR game a week ago. He was injured and fighting zombies on stairs in pitch black, while slowly suffocating from gas :slight_smile:

When he successfully disengaged, instead of fleeing, he tried to find and grab his dropped weapon in total darkness. He failed his saving throw and fainted. His comrade pulled him out he suffocated. The party failed their first aid test and was unable to bring him back to life.

While the player is 18, sonewhat experienced in RPGs and rationalized the event as the result of his own boldness, the other players were truely shocked.

Pro: I offered opportunities to flee the situation. The rolls were open. I showed consequences of bad decisions. After the death, I addressed their shock and shared my former bad experiences about it. I also let the other players act out their (characters) grief by asking ‘provocative’ questions about their emotions and memories of the fallen etc. I let the players mourn and do careful preparations for the funeral. We took time to let him go and not skipped these phases in favor of adventuring.

Con: I did not explicitly tell the players that this is a lethal gamestyle. Maybe I was not clear enough about the direness of the situation?


I’m going to sort of reiterate what I said in the linked thread.

It seems to me that most of the posters here, while well-meaning, are overly framing this issue from a aggressor-victim point of view, which is rather common in safety-culture circles. While there are cases where there is an aggressor (a malicious actor) and a victim, at this point we’re not talking about issues related to gaming safety, but just of social interaction. These issues are important, but don’t happen at the same level: I assume that a functional gaming group already doesn’t want to hurt each other and that anything that may happen of the sort is accidental and unwanted. If a person approaches the gaming group with a malicious intent (conscious or not), then I hope we can agree it’s an entirely different problem than what we’re talking about.

Inadvertently hurting someone is something that is itself hurtful to the person doing it, can generate the same type of response (silence or violence) and can carry the same feelings of shame and guilt that bottle up over time if unaddressed and eventually turn into resentment. “Being hurt” is something that can – not often, but it happens – as well be weaponized to cause this type of hurt and social shaming.

I don’t think anything of this kind can be approached without an attempt at seeing the entire thing as a miscommunication error between two people that need to be somewhat made to reconcile to heal as a unit. The only way to deal with this is having a social culture where it’s OK to feel hurt, but it’s also OK to say what comes to mind and potentially hurt people as long as one is willing to learn after stumbling. I don’t buy the ‘impact not intent’ mindset. This activity is inherently unsafe, as all creative endeavors.

As a diagnosed neurodivergent person – I really wish that I wouldn’t have to frame it like this, however it does probably strengthen my point to this crowd – my ability to read social cues, notice when others are hurt, and all of the other skills that you all are describing as necessary things, are difficult or in some cases impossible. I don’t want to make people feel bad, and if I do and I’m put into a spot, past feelings of shame and trauma come back – I’m working on reacting better as time goes on, but this is the reality of things. Saying “suck it up, it’s not about you” is completely unhelpful.

Every time I read something like what some posters have said in this thread, I just feel I should just stop roleplaying and never come back, as it’s obviously becoming a social minefield impossible for me to navigate, people don’t care about issues that are visually invisible and not trendy to talk about, I’m not going to go around with a hat saying “please forgive me, I have X”, and I have better things to do than to submit to this type of anxiety.

Kind regards everybody.


Hello, I followed perfectly until Froggy introduced a third voice, and there they lost me :
In “Saying “suck it up, it’s not about you” is completely unhelpful”, who’s talking to whom about what hurt?
I suppose the same is at work in “something like what some posters have said in this thread”. I can make do with not understanding part of the message, but I would like to know over what Froggy’s giving up TTRPGs. If only to see where the mines are ?

Mainly, this line, and a bit of the surrounding text. I didn’t mention it as I didn’t want to call anyone out specifically, as I didn’t feel that was particularly productive and I don’t think Psyke had any ill intent in writing it. I realize the result of this is that it sounds like I’m vaguebooking or being passive-aggressive, which was not my intent.

This brings to mind some of the experiences I’ve quoted above, which elicit from me a strong response. Creating narratives where there is a offending party and an injured party that that must be cared for by the group is incredibly unhealthy.

Essentially, what I’m trying to say is: the experience of roleplaying is inherently unsafe. However safe you want to make it, when producing creatively we present an inner side of ourselves that might be badly received, for whatever reason. People may get offended and hurt. Filtering the creative process introduces a certain phonyness to the experience which I’m not interested in pursuing.

Presenting a side of yourself through the creative process and being told that’s hurtful is itself incredibly hurtful. All you can do when you’re creating together is establish trust as much as possible and talk through problems when they happen, treating them as misunderstandings between mutually trusting and well-meaning individuals, that are probably mutually hurt by the misunderstanding.

Regarding your question about where the mines are: I don’t know and can’t tell, otherwise I wouldn’t call it a minefield. That’s my impression and experience, you can accept it or not.


I understand your points and they are worthy of being made, but I think you may have misunderstood the context of my original statements, so I hope you’ll let me clarify. In particular, I hear you saying that there should not be a villain in (many / most of) these situations, and that’s actually in agreement with my point from the paragraph you cited.

Quoting myself for context, from two separate parts of the post:

The reason I felt that I handled this poorly was that I stopped listening to the other party, and assumed the role of the villain in the narrative within my own head. As a result, I was unable to deal maturely with the situation and find a resolution, instead making this a bigger deal than it had to be. (Literally relationship destroying in that by ending the campaigns I lost touch with several of my players, although not this particular player.) I didn’t allow for healing or moving past the situation precisely because, as you pointed out, I was myself hurt by the fact that I accidentally hurt my player. To use the Power of TED vocabulary, I fell into the Persecutor / Victim mindset.


Thank you for staying and clarifying @Froggy and @Pyske too. I now realize that Pyske was never a target, but his words a trigger.
I agree with your analysis, seeing no fundamental difference between the 2. Shout at me if I’m mistaken.

Your formulation of the evils of White knighting, Froggy specially, is enlightening. It makes me reinterprete some borderline reactions about overprotective behaviors at the table I’ve heard elsewhere, as ill articulated versions of your argument.

For safety in game, providing a quick summary of the Bully Victim Savior triangle seems easy to do.

1 Like

I think the Persecutor / Victim / Rescuer triangle you guys quoted is a good place to start, trying to be careful not to fall too much into armchair psychology. It’s important to acknowledge that all elements of the triangle are responsible for perpetuating it, regardless of who ‘started’, and that all need to take some level of responsibility and change their behaviour and mindset to break the pattern.

Yes, I think I know what you mean. Sometimes people have a strong reaction to overprotective behaviour, which makes them sound callous or mean. It took me a while to be able to say this in a way that doesn’t sound like that, but more or less it’s the same train of thought.


To continue with what Froggy has described, I share a story where a player was very very disappointed and upset, but we managed to avoid the persecutor / victim / rescuer triangle.

It happened in the very first scene of our new campaign. I wanted to try out DramaSystem and every player seemed to be okay with the idea. I proposed a mix of GoT / blood opera / dark fantasy and they created very cool, tragic, epic characters. Every PC was in alliance with a victorious sorcerer king. I started the game with the death of this central NPC and was curious about how the fate of the alliance will turn out without him in the middle. Even though it was only the very first scene of a new campaign, and we only created characters a half an hour ago, one of the players, who played the lover and bodyguard of the sorcerer king, was so outraged about the death of this NPC that she had to act out.

We talked it through and I realized that we cannot continue to play. I think no one really made a mistake. It just misfired. We still play together sometimes. I also ran the same campaign concept for another group with high success.


With my gaming groups, I’ve had people object to content a few times. For example, Swords of the Serpentine has a skullduggery skill that lets you produce dirt on people, and I started making something up porting pizzagate into the setting. The group wasn’t OK with anything involving child abuse, even “off camera” by NPCs, so I quickly changed it to a rumor about basil being watered with e coli infected swamp water.

In this case we’d all been gaming together long enough that we were OK talking about it openly and changing the story. I think that level of trust is the fundamental requirement?