Player survey: How often do you like to die?

[MaxWilson]

Obviously as players we all try not to die, but how difficult do you like the game to be, mathematically?

I've known some people who consider a 50% chance of dying over 30 game sessions unacceptable because then you "might not finish the campaign."

As a player I like games where 50% chance of death-from-math every single session exists. Sometimes this means playing with low stat rolls or "bad" subclasses. I.e. I want a game where the monsters are as strong as I am or stronger, so that managing to survive the session is a meaningful accomplishment, instead of having to survive 100 sessions. I just don't have the free time or interest to wait 100 sessions for a challenge. I want Darwin to show his face now! :)

Spoiler: Additional context: as a DM

Show

I tend to run games where a 20-60% chance of any given PC dying in every adventure is plausible, based on math (and in FTF games, an adventure tries to be no more than one or two sessions long), although I then also try to lean back in the players' favor in being very open to their attempts to beat the math or bypass the math via PC abilities or player creativity. So my actual PC death rate ranges from zero to about 1 per session, which is 0-25% fatality rate per session, not the 20-60% that raw math would indicate. E.g. four 8th level PCs vs. three CR 9 monsters looks like almost a 50% chance of death per player, but in practice it isn't that bad.

As a player, how often do you like Death to show his face to you in 5E?

[ProsecutorGodot]

It's hard to say, in our Mad Mage campaign, which is now bordering on 3 years in length... Nobody has died. The number of close calls can be counted on a single hand. I enjoy the campaign, but I would prefer if there was a higher degree of difficulty.

I would love to take credit and say my Redemption Paladin is just that effective, but I don't think that's it.

I think my DM is a bit hesitant to try and kill us and I also think I'm the only player at the table who dislikes that.

So I don't really know what kind of odds I'd want, but definitely more than 0%

[Seekergeek]

I prefer a deadlier game, both as a DM and a player. I would be happy if a game presented a challenge and a real risk of death on a session-by-session basis. It makes the game more rewarding for me if I am in a position where death and danger are overcome rather than plot points that get written around.

That sentiment is unfortunately not shared by my very long-running table. We've been playing together since we were teenagers (we're old and in our mid to late 30s and playing remotely across the country now). For narrative purposes most of my group would prefer to finish a run with the same character they started with. Our main DM would prefer to have the same PCs finish as started which in his home brew work tends to lead to plot armour and DMPCs running rampant. If he's running a module, his preference tends to manifest itself as well, and often clumsily. I sometimes find myself actively trying to die just to see how he will creatively avoid it for me. It's not an adversarial table, nor an optimized one - we've been playing together for so long that the competitive edge has long since dulled. But even the most derpy of builds tends to survive start to finish unless we actively ask to retire a PC.

edit: We are also currently doign the Mad Mage - we're less than a year in still and have had one close call for the party but we were all immediately stabalized by the NPCs. I'm not sure if that was a published decision or the DM pulling punches but it really bothered me. We messed up, had horrible tactics and paid the price...only we didn't. After that, there was a potentially deadly side effect of our capture which was essentially overcome by DM fiat. Again, I don't truly know if it was a written solution or our DM avoiding our deaths for us. Either way, it's not how I like my dnd.

[Vegan Squirrel]

This is a very interesting question to put numbers to, and ultimately I feel like it should be as much in the hands of the PCs as the DM. I think I prefer a curve of changing odds over time, where you're likely to survive for the first few sessions to get to know the characters (make them memorable), then you face a higher chance of death through adventuring, then that chance of death goes down as the campaign becomes a long-term campaign and you're more invested in things (and growing more capable), then the chance rises continuously from there, as you take on epic challenges that could easily kill you.

But maybe I'm overthinking things. The chance of death should make sense for the situation the characters are in, but the characters should have some control over what situation they're in. Eventually (for good characters, at least), there should be a strong enough in-world threat that the characters are willing to sacrifice themselves to stop it.

As a DM, I generally try to place threats before the players which I think they can handle, but then I'll not hold back from using their enemies' abilities to the fullest extent. And if it would make sense, in-world, for the threats to be beyond them, the threats will be beyond them. I expect them to avoid those dangers. My characters have gone unconscious and made death saves quite a few times, and I've killed off NPCs they were fighting alongside, but so far the PCs have all survived. There have been death saves made with two failures already, so fate could easily have changed that outcome. They know the risk of death is real, and they usually take precautions accordingly. And I suppose my villains usually die trying to escape rather than die trying to bring the PCs down with them—if they were more fanatical, they may very well have succeeded in mutual destruction.

I'm currently DMing 11th level characters, the highest level and longest campaign I've ever run, and the villains on the horizon will definitely be deadly. The characters will have the chance to collect intelligence, allies, and magic items to aid them, but their enemies will, too.

[da newt]

I like to aim for a 50-50 chance that an individual PC will live to complete the campaign, but I also think it's very important for there to be a compelling reason for a person to choose to put themselves in a situation where the threat of death is real.

I prefer a fairly light hearted game with episodic periods of dire consequences. I don't like slogs where death is always one bad roll away, and I don't like games where the PCs are neigh-immortal. It's a definite Goldilocks principal sort of a thing for me (not too hot, not too cold, but just right somewhere between).

[Dravda]

As a player, I like a 50-50 chance of death. Not at any random time, mind you: I want the opportunity for my character to die gloriously if the perfect opportunity arises. Being filled with arrows while covering everyone else's retreat, or in a suicidal rush to take on one last important bad guy with me before I go? Awesome. I don't mind my characters dying at all, so long as it has narrative weight and meaning.

As a DM, I typically shoot for 1-2 character deaths over the course of a campaign. Enough to remind my players that death can happen from poor planning or bad luck, but not present at every turn. That said, it's been a while since a character actually died in one of my campaigns: I feel like I'm still mastering 5e!

[Pex]

When the DM boasts of his PC death count he needs to step away from the chair. It's not the DM's job to kill PCs. He is not the players' enemy. The DM should be just as saddened about a PC's death as the player. There's no defined number to say X is fine, X + 1 is too much. It's easier to talk about the extremes. If a PC dies every session the DM is doing something wrong. If a PC dies every other session the DM is doing something wrong. Every two sessions. EVery three sessions. Eventually you reach a point where it's just part of the game. If no PC dies during the campaign, wonderful. If there's only one death the entire campaign, fine. Two deaths, fine. You eventually reach a number where something is wrong.

Too many deaths, however many is too many, I blame the DM, but the DM is not to blame for all PC deaths. A player who does something Stupid that leads to death is not the DM's fault. It's the player's fault. Stupid is defined as something that is obvious by objective fact and reality shouldn't be done. A death can also be no one's fault. A player could make the wrong choice but wasn't Stupid about it. He took a risk and it failed. A player can do everything right, but the bad luck of dice rolls makes the Thing a failure that leads to death. DM's fault comes in when it's deliberate, the situation is stacked heavily against the player it takes mind reading or extreme luck to succeed, or the mysterious X + 1 number is reached when you realize something is seriously wrong with the game.

[NecessaryWeevil]

Death as in "Ooops, time for Revivify" or "Time to find an NPC to cast Raise Dead"? Or death as in "You're not coming back, create a new character"?
We've had one or two of the former in every campaign (both of the most recent incidents the same player...I guess that's why he plays so conservatively lately). We've only had one of the latter ever, and that was only because the Silver Flame doesn't believe in resurrection and going toe-to-toe with Lady Vol herself was a suitably epic departure.
I could accept about a 1 in 5 chance of permadeath per campaign.

[MoiMagnus]

I'm assuming you're talking here about "unwanted PC elimination".
So we exclude:
+ deaths followed by resurrections
+ deaths happening when a player was already set on changing character
+ deaths happening at the last session of a campaign, or during the epilogue
But we include events that are not technically deaths, but result in the character no longer being available as a PC (without the player's consent).

With this definition, I'd say that for a ~30 sessions campaign, I'd expect it to have one session in average where this happen to one or more players.

I definitely don't play D&D to live through some Darwin's natural selection. That looks like stress to me, and I don't need more of it. Additionally, I like to get to know the other PCs session after session. One death is interesting, but if they keep dying, that makes campaigns far less memorable to me.

[Xervous]

It's not the DM's job to kill PCs.

The term I strive for is Suicide by Adventure. It’s not my job to kill the characters, but I do tell them what may kill them. If they decide to pursue such risks or refuse to back down from a fight it’s simply a matter of delivering on a promise and keeping the world consistent. The key part is that it’s informed Suicide by Adventure.

[MaxWilson]

Death as in "Ooops, time for Revivify"? Or death as in "You're not coming back, create a new character"?
We've had one or two of the former in every campaign (both of the most recent incidents the same player...I guess that's why he plays so conservatively lately). We've only had one of the latter ever, and that was only because the Silver Flame doesn't believe in resurrection and going toe-to-toe with Lady Vol herself was a suitably epic departure.
I could accept about a 1 in 5 chance of permadeath per campaign.

I am not drawing a distinction between death and permadeath for purposes of this survey. Whether PCs have the means to reverse death is a separate topic. Right now I'm more intent on getting a feel for players' preferred tactical difficulty (from a raw math standpoint).

[CheddarChampion]

I think I'd like to have a character death approximately once in every 10 - 20 sessions.
Maybe more if returning to life is an option, maybe less if it's a long running campaign.

Both death and the possibility of death make it more meaningful and exciting when you survive.

[LtPowers]

The term I strive for is Suicide by Adventure. It’s not my job to kill the characters, but I do tell them what may kill them. If they decide to pursue such risks or refuse to back down from a fight it’s simply a matter of delivering on a promise and keeping the world consistent. The key part is that it’s informed Suicide by Adventure.

But is there a viable alternative to pursuing the risk? Or is it "pursue the risk or that's the end of the adventure"?


Powers &8^]

[KorvinStarmast]

I could accept about a 1 in 5 chance of permadeath per campaign.

I am good with, on a campaign basis, 3 or 4 permadeaths if the campaign is a long one. New chars come and go.

It’s not my job to kill the characters, but I do tell them what may kill them. If they decide to pursue such risks or refuse to back down from a fight it’s simply a matter of delivering on a promise and keeping the world consistent. The key part is that it’s informed Suicide by Adventure.

Interesting approach, sounds a bit 'West Marches' in tone.

Max, about every three sessions if the chance of PC death isn't very real, my taste is "DM, please turn it up a notch and see how it goes for a few sessions"

A perma-death every other session, on the other hand, gets tiresome fast unless all of the players embrace that.

[Sherlockpwns]

I like to die once in a lifetime.

[PhoenixPhyre]

If every fight (or even every session) has a 50:50 chance of someone dying, then the chances of getting to T2 without a completely different party than you started with are...slim. And there's no resurrection in T1 at all (unless you're filthy rich).

In 6+ years, with 13 parties (not including one-shots), I've had 5 character deaths including those who were resurrected. Two permanent deaths.

1. A character died in session 1, from something that shouldn't have killed them--everybody screwed up at the player level, myself included. That ended up getting retconned with mutual consent.
2. A character perma-died when he, as a level 2 paladin, decided to solo a CR 9 dire yeti. After multiple rounds of "are you sure" and the monster stretching, being intimidating, etc. Massive damage is a pain.
3. A character died (but was resurrected) at about level 8.
4. A character perma-died when he was stunned and the mind flayer got a crit on the Extract Brain ability. No coming back from that without a much beefier resurrection spell than they had available. Especially since that was the party cleric.
5. A character died (but was resurrected) after he (with 98 HP max) faced a lich with power word: kill.

That's it. I've had lots of characters retire and get traded out during play though.

But I find death to be a boring failure state. I don't play to find out "will you survive", I take survival as a presumption. Instead, I'm more interested in "how does the world change because you've been adventuring in it? How have you changed? Who will you protect? Who will you sacrifice?" So failure states are more about the thing you wanted to do doesn't get done or backfires on you than you die and Bob #2 comes in to replace Bob #1.

[stoutstien]

For me it's more of a question of how much the chance of death fall on my decisions compared to relying on death based on RNG. If it's around a 80/20 split there I'm happy around a 20% chance of death everyday starting around T2.

[Rater202]

Not at all.

There should be risks where it feels like there's a chance I should die, but if my character dies it should be because I did something really stupid and failed.

Granted, this is becuase m first two-player character deaths were kind of crap.

Honestly, risk to the people around me(NPCs, other player characters, etc) makes the risk feel more real to me than danger to me. Danger to my character just annoys me becuase I'm here to have fun, not worry about every little thing.

[MaxWilson]

But is there a viable alternative to pursuing the risk? Or is it "pursue the risk or that's the end of the adventure"?

Powers &8^]

You never want the players feeling railroaded. But you can offer them interesting choices.

Spoiler: Evil Oni Empire Players read at your own risk of spoilers

Show

Consider: your 6th level players are sneaking back home after an illegal dungeon crawl. They run into a woman crying in the woods. Turns out a local demon has been courting her against her will, bringing her presents, attempting conversation, etc. It's grown angry at her obvious lack of reciprocation for its affections and after its last visit, it burned down her family's barn and killed the dog. She's terrified of what will happen next.

The players can take that as an interesting bit of world building and move on. Or they can choose to get involved, without knowing the precise risks (what are the demon's abilities? how many demonic allies?). This isn't part of any overarching plot, it's essentially a Gygaxian random encounter (i.e. one with context, not just a random monster attack that's over in a few seconds).

Would you classify this as "pursue the risk or that's the end of the adventure", or not?

[Xervous]

But is there a viable alternative to pursuing the risk? Or is it "pursue the risk or that's the end of the adventure"?


Powers &8^]

It depends on where they are for the given arc they’ve chosen to pursue. Sometimes it is indeed a matter of cutting things short or risking death. If the outcome is heavily desirable my players are willing to bet their characters’ lives on it. Though in true tabletop tradition the main near death experiences have been from optional encounters and (mostly but not wholly) well intentioned friendly fire in the most recent campaign.

Though hireling deaths. Party record for simultaneous losses is... 22?

[NecessaryWeevil]

I am not drawing a distinction between death and permadeath for purposes of this survey. Whether PCs have the means to reverse death is a separate topic. Right now I'm more intent on getting a feel for players' preferred tactical difficulty (from a raw math standpoint).

Ah, okay. In that case, 50% chance per ten sessions sounds about right, skewed toward the later levels when the party has access to resurrection (either internally or through cash or allies).

[KorvinStarmast]

Would you classify this as "pursue the risk or that's the end of the adventure", or not?

That depends on how many different ways there are to deal with that demon.
Examples:
1. Combat
2. Negotiation
3. Trade
4. Learn its name and banish it
5. Secret weakness: pine nuts coated in nutmeg and honey
Et cetera

There being multiple paths to pursue to address the impact that demon has on the game world is what makes for an interesting challenge to overcome.

[Vegan Squirrel]

I'll try to hazard a numbers guess here, for the sake of the original question, but it's still hard to quantify. I think, as a player, something like a 5–10% chance of character death per session sounds about right. That will vary; some sessions will be risk-free (especially those without combat) while others will involve much greater risks (intentionally challenging the big bad).

And honestly, if resurrection magic is on the table, then I'd rather not see any death, because resurrection magic breaks the meaning of the game so much for me. Just redefine 0 HP as unconscious and eliminate death saves at that point. If death is permanent, then I don't mind character death happening. I'm sure that's the reverse of most people's opinions, but there you go. I'll go along with resurrection magic for the sake of the other players, but I don't like it on a meta level.

I could also enjoy an old school deadly campaign where you cycle through characters as each one inevitably fails to survive, but that would be a very particular experience that runs against the expectations of a standard campaign these days. And I'd want to simplify character building for that, probably using a much, much lighter system, because building 5e characters takes way too long to do every session.

[Rukelnikov]

For my it depends on the level, since dying doesn't carry the same threat across all levels of play.

Lvls 1-2 Dying is extremely easy, an unlucky crit and you are rolling death saves or die alltogether, and at those levels, death is usually permanent. On the other hand those characters haven't been played that long, so its not as much of a downer as it can be dying after 20 sessions. This is, to my taste, a bit more deadly than I'd like it to be, but its ok, especially if playing a sandbox instead of a more story driven (which is what my group usually does).

Dying is a real threat here, and because of that I don't want it to happen as often as it would without fudging rolls or intelligent enemies taking unintelligent decisions in combat.

Lvls 3-6 Death from unlucky crit/damage roll is still a possibility since HP hasn't bloated that much yet for most PCs but not as likely as lvls 1-2, and if you die access to Raise Dead may not be as readily available yet, so the threat is still there. Revivify does come into play at 5, and I think it already starts watering down the threat of dying, and thus of combat in general. 10 rounds of grace means if someone in the party is able to cast said spell, we can just finish the encounter and Revivify afterwards, however that person needs to have the spell slot, and not be the one who died.

Dying is a threat the party may or may not be able to circumvent (This is the sweet spot for me), and it should happen often enough to remind players that combat is dangerous, maybe one PC or two dead across those levels which may come back or not, in that uncertainty there is thrill.

Level 7 onwards Death in 5e is not much of a prob sadly, its extremely cheap to cast raise dead, and even if no one in the party can do it, you are high level enough to find someone who can, your odds of coming back at this level are pretty high honestly, and while you do have a couple of days of weakness, you don't lose a level, or a point of Con, or have a cap to the amount of times you can be revived, or roll to see if you actually revive or not or some other "lasting" complication. So... death means only a couple days of extra caution, and thus the only regular threats of combat become TPK, which is kind of extreme, I wouldn't want that to happen in most campaigns, or a couple days of weakness which can be bad if time is of the essence, but its negligible in effect if its not, there's no "middle" threat like there was in previous editions (that's why we play with lasting injuries, even if you come back, you may roll bad on the table and lose an eye, or leg which raise dead will not restore)

Death needs to be somewhat commonplace for there to be any threat at all (outside doom clocks), in my group by level 10 or so, its uncommon that two sessions go by without someone dying (we usually play one or two Deadly x N encounters per day).

At really high levels(16+?) "Death" can become a problem again, cause it can come via soul stealing/sealing, trapped in time, or some such to make revivification processes more difficult or impossible. Those types of death, which are real threats, shouldn't be as commonplace as merely dying, thus going back to the frequency death had thru levels 3-6, maybe once every couple levels is ok.

My main gripe with this, is that levels 7-end of the campaign (usually), are the "safest" levels of play, which is something I constantly try to thwart, cause those are the levels where pcs start being able to do serious stuff, but the only combat threats you can throw at them are TPK or none.

[MaxWilson]

That depends on how many different ways there are to deal with that demon.
Examples:
1. Combat
2. Negotiation
3. Trade
4. Learn its name and banish it
5. Secret weakness: pine nuts coated in nutmeg and honey
Et cetera

There being multiple paths to pursue to address the impact that demon has on the game world is what makes for an interesting challenge to overcome.

Let's assume for the sake of argument that the DM hasn't put a ton of effort into enabling secret weaknesses and stuff for this demon. He's open to creative solutions if you've got them, but fundamentally this is an open-ended, optional bandit encounter against a super-bandit of unknown strength, unless you as players come up with a way to determine that strength. (E.g. pretend to be a rival suitor and challenge the demon to an arm wrestling contest.)

But the options the DM has prepared for you are limited to "ignore it and move on to the next scene" and "get involved, probably wind up fighting it."

Given that it's essentially a random encounter, would you classify this as "pursue the risk or that's the end of the adventure", or not?

[JonBeowulf]

I've been D&D since the very beginning and I miss the days when adventuring was a very dangerous occupation only chosen by the foolhardy, the desperate, or those sent on a quest. OD&D was an unforgiving system that quickly eliminated careless PCs and frequently brought down a PC whose only mistake was to be played by someone rolled poorly that one time. It also prevented annoying characters because it drove a mentality of, "My life is on the line here. Why am I traveling with this idiot?"

A lot of the mechanics of that system were terrible, but the games themselves were a lot of fun.

My biggest complaint about 5e is how quickly a single character death can turn into a full-on TPK. It's hard to kill a character and once it happens, it's even harder to keep the whole thing from falling apart.

How do I, as a DM, create an adequate sense of risk and danger without risking the entire campaign? Too much work. I don't like it so end up running games that are fun and challenging but not really all that dangerous.

So yeah, I'd enjoy a group that accepts that permanent death could be waiting in every cavern, ruin, or dark alley and plays accordingly. I can't define a range, so "deadly enough" is as close as I can get. As a DM, it allows me to run the world and trust the players to play intelligently. As a player, it forces me (and others) to pay attention to what's going on.

[DeTess]

It depends a lot on the style of the campaign. A highly tactical very combat focused dungeon-of-the-week game I'd be a lot more okay with high risk of death than a long-running narrative game. The exact consequences of death also matter, as does the theme of the game.

Thing is, death isn't the only kind of consequence you can use, and if you're running a narrative focused game it's best to use it sparingly because killing a long-running character generally hurts the story, not helps it. The most tense moments I've had in narrative games weren't related to the chance of my character dying but to the chance that my character would fail at something they considered very important. Losing family or friends, or just not living up to their own ideals have a far bigger impact than having to bring a different character to the table the next session.

So, for a narrative game I'd say characters actually dying shouldn't happen all that often, and most of the time it should happen due to active choices of the character, rather than just bad luck on the dice.

On the other hand, if I was playing a game that's heavily focused on combat and tactical gameplay, rather than a long-running narrative I'd be far more okay with characters dying. Likewise, if there's a tone to the game that corresponds to lots of death I'd be fine with that as well. I've got GM notes lying around for a game in which the players are an ill-fated expedition into the unknown. I'd have them actually stat up the entire crew consisting of 30-or-so people and I'd happily start killing off those characters as disaster strikes both to set the tone of the game, and because losing these characters has an actual long-term impact (as there are no reinforcements coming, so losing the expeditions main doctor has serious long-term consequences).

Anyway, these are my rambling thoughts tl;dr I think death shouldn't be sued much, unless it really plays into the game's theme or the consequences of a character dying are pretty minor. For narrative focused games there are far better buttons to press in case of failure than killing off a PC.

[Omni-Centrist]

Obviously as players we all try not to die, but how difficult do you like the game to be, mathematically?

I've known some people who consider a 50% chance of dying over 30 game sessions unacceptable because then you "might not finish the campaign."

As a player I like games where 50% chance of death-from-math every single session exists. Sometimes this means playing with low stat rolls or "bad" subclasses. I.e. I want a game where the monsters are as strong as I am or stronger, so that managing to survive the session is a meaningful accomplishment, instead of having to survive 100 sessions. I just don't have the free time or interest to wait 100 sessions for a challenge. I want Darwin to show his face now! :)

Spoiler: Additional context: as a DM

Show

I tend to run games where a 20-60% chance of any given PC dying in every adventure is plausible, based on math (and in FTF games, an adventure tries to be no more than one or two sessions long), although I then also try to lean back in the players' favor in being very open to their attempts to beat the math or bypass the math via PC abilities or player creativity. So my actual PC death rate ranges from zero to about 1 per session, which is 0-25% fatality rate per session, not the 20-60% that raw math would indicate. E.g. four 8th level PCs vs. three CR 9 monsters looks like almost a 50% chance of death per player, but in practice it isn't that bad.

As a player, how often do you like Death to show his face to you in 5E?

I want you to try to kill me in proportion to me taking risks or expressing myself in the world. If I'm doing laundry dont make me roll Initiative, if I'm traveling in a area known for kobolds and 100 of them pick up my trail, have them roll 100 attacks against me.

Make it real and rewarding.

[Waterdeep Merch]

I've got a reputation for being a killer DM among my players, but the permadeaths I've dealt out are extremely rare. Oh sure, I've had the party jump through some hoops once in a while to make a resurrection meaningful, but, hrm. Let's see...

*Player walked into an obvious beholder trap that the rest of his party told him was a trap and got disintegrated. They still laugh about it. Players had stables of characters, this meant the player had to grab a different character sheet.
*Two players got the wrong side of a random encounter table in unfavorable terrain and got eaten by eleven wyverns. This one's probably the most egregious to be honest, though we did agree to uphold strict rulings on things like this so that they could be exploited in turn (for things like leveling or deliberate hunting), and there was another method of travel that they willingly didn't use. Same campaign as above, the eaten characters were immediately replaced.
*I killed nine PC's in the Tomb of Horrors, but it was a one shot with multiple characters per player. No one completely ran out of characters.
*During a Christmas one-shot, a player got dominated and ended up using his very well-built barbarian crit fisher to absolutely murder his friend. This was the final boss battle.
*During a different Christmas one-shot with evil characters, two of them teleport-ditched the other three in front of the final boss. One then died while the other two pledged eternal servitude.

I think that's all of them. Five instances, three deaths that weren't one-shots, a total of fifteen, and none removed a player from being able to play for more than a single fight.

As a player, I've permadied three times.

*One-shot Cyberpunk where a fellow player spontaneously decided at the end when the GM started turning it into a full campaign to turn on and murder the rest of the party for no narrative reason. They weren't even against the campaign, they just thought it was funny to ruin it.
*One-shot 3.x hack of Call of Cthulhu where the DM had me as a traitor from the very start. The party killed me, but the damage was done and a Shoggoth finished what I started. I suppose that counts as a victory?
*3.5 game where I died due to a mutual narrative after returning to my home kingdom to discover my last living kin had joined my archnemesis. So rather than rolling dice, I agreed to an epic death after a duel.

So only one was outside of a one shot, and I personally agreed to it before it happened.

[firelistener]

I think it's best avoided as much as possible, both as player and DM, but I don't like fudging rolls or allowing do-overs. The only exceptions for me are when you just started a new campaign and the party wipes. I usually allow them to try a different strategy against whatever enemy wiped them out. My players usually don't bother with "plans" until that happens lol.