How to Implement Save or Dies in a Balanced D&D System

[Kaelik]

I don't buy it. At least not in the abstract.

I play in a game where one player frequently chooses non-optimal battle choices for character reasons. The character is a sneaky rogue and COULD get sneak attack every round but instead spends 2-3 rounds setting up those attacks. Even without 'death flag' we have characters that choose suboptimal routines and the game adapts.

Stomping encounters harder than you would have is not a problem. You were going to win easily and you won even more easily - that's not hurting anything.

All these arguments are not very good for defending "Fighters vs Wizards" where there's some problem in the game that needs fixing.

They are significantly less compelling as an argument for adding a new bad thing that didn't exist!

There is no good reason to give people a numerical boost for raising their death flag! It's very bad!

[deaddmwalking]

If two characters have an ability (like Rage) and one chooses to use it and one doesn't, one character will be 'more powerful' than the other for a given fight. When abilities like rage are limited to a number of times per day, players that are BETTER at calculating how many times they'll need it are at an advantage versus players that don't.

But the characters themselves are 'equal' because both characters COULD activate the ability. Sometimes choosing not to activate it on 'easy stomp' encounters is the right choice because that means you won't have the ability available when you REALLY need it.

Death Flag, in the abstract, could be like rage (a limited ability that you choose to activate). Likewise, it could be a defensive ability you choose to activate (I can't be killed NOW). As an ability that is activatable 'at will', there's a POTENTIAL for some amount of 'power disparity' between people that use it all the time versus people that don't, but there's also still a cost (potential for death). That is a real and serious cost and you can't pretend that it isn't.

When all players have THE CHOICE to use the same ability, but they choose to use it at different times/circumstances, it's hard to state clearly that one player is 'more powerful'. In the abstract we can agree that a player might be REALLY GOOD at knowing when to use a power optimally, and an optimally played character is going to be more powerful than a non-optimally played character, but ensuring 'absolute balance' isn't even a desirable goal. 'General Balance' is a good goal. Death Flag (in the abstract) fails to create 'absolute balance' because you could imagine a situation where one player is 'always right' and one player is 'always wrong', but since it comes down to a player choice (and their personal values) that's not always a bad thing.

If you offered your players to play a character of Level x with the caveat that THEY CANNOT DIE or you offer to let them play a character of x+1 with 'death and dying as normal rules', that's defensible from a design standpoint - as long as the choices are communicated clearly. That's very clearly a 'I have a defensive ability and it costs me something in raw power' choice. It doesn't become a less meaningful choice if you can temporarily suppress your ability for +1 level

Having a mixed party of characters with a 1 level difference isn't even uncommon in the game. Yes, it is a power-mix. That's not always DESIREABLE, but it also isn't always wrong. The power discrepancy between characters of 1-level difference is OFTEN not enough to really matter in terms of outputs - especially considering the variety of inputs.

Ultimately, you're playing a game. 'Plot Armor' is often 'genre appropriate'. Taking off your plot armor to 'raise the stakes' may also be appropriate. There's no reason to EVER take off your plot armor if there's no benefit to doing so. Therefore, if you want to create a situation where stakes feel higher (because death is on the table) but you also want to respect player choice, giving them a mechanical incentive is a viable choice.

I'm imagining myself as a player that never 'raises the death flag' and I don't mind that other people do. If my contributions became meaningless because I didn't, that might be an issue, but there's a lot of ground between 'not contributing as much as I potentially could' and 'not contributing anything at all'. There are LOTS OF WAYS to fail to 'play optimally', and that's kind of important because players need to make choices. If the optimal choices were always known in advance and players always took those choices, you're playing Candyland and you know how the game is going to end by looking at how the cards are stacked. That's not the game that you want to be emulating.

[Neo Phonelobster Prime]

I don't buy it.

So. You don't just insert new characters except when you do. Making it a cost not for the player exploiting death flags, but for everyone, except when you don't.

You use a mechanic not remotely the same as Death Flags. And you will defend it. By giving an example where it isn't actually used. And which looks exactly like a game without it and a GM typically low balling a near TPK event. But that is presented as a good thing and an example of the mechanic working as intended?

And that failed defense of a different system is a defense of the death flag system as long as you squint hard enough?

Then you think that makes giving SOME players a giant bonus for free and denying it from others because they don't like the fundamental thematic difference of losing their current character. Somehow is supporting players "valuing different experiences" without punishing them?

I'm seeing a lot of low effort hand wavium.

"If you want to keep your character it will be significantly numerically inferior." is a bad thing and obviously so. If you want to defend that you need to try a lot harder.

[Neo Phonelobster Prime]

If two characters have an ability (like Rage) and one chooses to use it and one doesn't, one character will be 'more powerful' than the other for a given fight.

FFS. Only using the rage ability causes your character to die at the end of the encounter.

Otherwise you are waffling with false comparisons and nonsense.

[MGuy]

I think losing your character is a significantly higher price to pay for some numerical advantages and reminds me of a number of games where you have a random chance of dying/becoming an npc as punishment for actually using your abilities. I do not like rules that include that personally but I can understand their inclusion if your characters are supposed to be throwaway, dying is more an inconvenience than a game ender, or the idea is to match the theme of the setting and that's valued above reasonable hand design.

This death flag thing is harder to defend I think because the price you pay for its usage is not like Rage or whatever your x per day ability of choice might be. It encourages players to open up the ability for their character to die. I do not see the gameplay benefit in doing that. If you want to imitate genre conventions where a character really only dies in important situations it would more likely come in the form of "boss" enemies being able to actually execute characters (probably in a telegraphed way to allow others to do something about it) or letting a character definitely be able to successfully do a thing, maybe take something off the board where they normally couldn't in exchange for their character dying or a hefty, permanent, debilitation. I don't think being able to just carry around buffed stats in exchange for a detriment that is imperceptible unless it comes up does the trope well.

[Thaluikhain]

This death flag thing is harder to defend I think because the price you pay for its usage is not like Rage or whatever your x per day ability of choice might be. It encourages players to open up the ability for their character to die. I do not see the gameplay benefit in doing that.

What about comparing it to Reckless Attack instead? You are more likely to hit, but more likely to be hit. Now, that doesn't necessarily equate to being more likely to die, but someone who did that all the time would be putting themselves at risk in order to achieve more.

[MGuy]

Trading defense for greater attack is fine I think. The glass cannon archetype is very common in games of all kinds. Trading effectiveness for a greater chance to lose your character is still a very different thing.

[deaddmwalking]

Trading defense for greater attack is fine I think. The glass cannon archetype is very common in games of all kinds. Trading effectiveness for a greater chance to lose your character is still a very different thing.

It's not, really.

'Immunity to Death' is a defense, similar to a high Armor Class. Lowering your defense for an increase in offensive power (like Reckless Strike) makes you more likely to die - you're more likely to hit/take damage and if you miscalculate how many people will target you, you're more likely to die. The difference, such as it is, is that immunity to death is a strict immunity - lowering your AC by 5 points doesn't guarantee that you'll take even a single additional hit and/or that you'll take any more damage - if your opponents had no trouble hitting you to begin with, the 'lower AC' is probably not much of a disadvantage. But becoming vulnerable to death also doesn't guarantee that you'll be at any additional risk - just that the possibility exists.

From a genre emulation standpoint, Wheel of Time appears to have something like this: Rand uses a sword technique that leaves him defenseless against a serious attack but allows him to be certain to succeed against his opponent. While it is a book and still has 'plot immunity' (we know Rand didn't die) and we can't derive a specific game mechanic, we could potentially say that creating scenes of that type are something we want to encourage in game and we can start considering options to create it.

Immunity to Death is a valuable ability that may be prized by a large number of people, but alternatively, it may feel like 'training wheels'. Does it make sense for characters to risk jumping over a pit of bubbling lava? If a character knows they CANNOT DIE, they may feel justified in taking a risk with a low-probability of success. They may literally jump from flying airplanes and crash into the ground face-first to save casting a minor spell. That's not behavior that most games want to encourage. Explaining away 'certain death' may cause issues with maintaining verisimilitude, which is critically important for the success of a long-running game. Adding it to the game should not be taken lightly, and the implications should be addressed.

But if you do add it into the game, offering players an incentive to 'turn it off' is the only way that it will be turned off. If your goal is to create situations where players consider 'turning off' the ability in order to increase their likelihood of achieving success, and they do, you've succeeded. At the end of the day, that's what design is about. If you want to encourage something and it doesn't happen, you've failed. Now, specific implementations run risks of failing for one reason or another. Lowering your immunity and then immediately raising it again before anyone can take advantage would be one possible failure - clearly if you're going to lower it, it would have to be for a period long enough to make it meaningful. Similarly, the 'benefit' would have to be tailored; a single +1 to an attack (or even a single +20) is unlikely to meaningfully change the nature of an encounter - using this ability has to be meaningful because of the narrative stakes it represents.

Phonelobster & Kaelik seem to believe it can't work. I believe it can. Based on PL's original objection, it seems that lowering your flag should provide a powerful short-turn benefit (one scene) but raising your flag should require a significant investment (like a long rest). This would prevent one player from always running their character with the flag-down (and thus always gaining the benefit). Such an interpretation is not contrary to Tussock's original suggestion (which was a 'here's an idea with no specific implementation') - it's just refining and clarifying how it could/would work in practice.

I still fail to see how giving everyone a power that they can choose to use if/when they sit fit is automatically a problem when some people will use it effectively and some people won't. Games give people choices; designers should build character choices to ensure that characters have similar power options - but they can't also MAKE characters use the powers. Some wizards don't ever cast spells. That's stupid and lame, but if Gandalf is having fun waving around Glamdring, it's okay that the designer didn't think about 'how to make wizards EVEN MORE POWERFUL in the event that they CHOOSE not to cast their spells'.

When everyone has the same power, and some people choose to use it and some don't, that's balanced. They all have the same freakin' power! That's so much easier than giving people DIFFERENT powers that are GENERALLY equally effective, but NEVER EXACTLY as effective.

[Neo Phonelobster Prime]

Lowering your defense for an increase in offensive power (like Reckless Strike) makes you more likely to die

Death flag mechanic says no.

[MGuy]

Whether or not you die in the system from losing a fight or getting hit too many times is still different from a mechanic that enhances your character in trade for being more likely to die. Losing a fight doesn't necessarily lead to death. This thing where the enhancement specifically leads to death necessarily does. I don't see a good reason to do that unless you want to encourage people to die.

[deaddmwalking]

A system where you can't die under normal circumstances reduces character death.

We're not comparing systems where death doesn't exist and we're introducing a new and novel way to die - we're comparing systems where death exists and you can choose immunity.

Activating 'death shield' or lowering 'death shield' is a semantic difference but either way you have a death shield you didn't have before.

[MGuy]

Alright so let me ask a question to see if we can get on the same wavelength. Why do you want to encourage players to die vs just not letting death be an option for those that don't want to?

[Neo Phonelobster Prime]

We're not comparing systems where death doesn't exist and we're introducing a new and novel way to die - we're comparing systems where death exists and you can choose immunity.

Or rather you are introducing a mechanic where some, and in my estimate many, groups will collectively choose not to die ever. And in the process be rendered mechanically inferior to the supposed dramatic challenges that matter.

You don't get to say you are comparing grey to grey when you are defending a mechanic which entirely within itself has the capacity to generate black and white outcomes.

[MGuy]

If the player gets the bonus for opting in to being killable then this is a context where normally they could opt not to get this bonus and be unkillable. So the context here is "death doesn't exist and you're introducing a way to die". That's the lens I'm looking at this through. Whether it works or not depends on if that is the outcome that's being sought or if that is an intolerable consequence of its implementation. I have said it in other places and I'll repeat it here. I do not personally care much about what arbitrary goal a person has. I'm only interested in whether or not the rules used to get to that goal are a good way to get to that goal once it is decided. I know that death is desired in some way based on where this discussion came from. I figured that beyond that what was desired was a way to make possible death, or eventual death, feel meaningful while giving players a reason to worry about it. I offered what I thought were better ways to get that because I do not think simple numerical bonuses are a cool or interesting way to achieve that goal and essentially getting nerfed for not wanting your character to die does not seem like good design based on what I thought the design goal was.

It does feel like a waste of time to keep up with claiming that X isn't X. An ability that trades defense numbers for offense numbers is not the same as choosing to open yourself up to permanent character loss in exchange for numbers. One is taking on a penalty for a bonus, both dealing with numbers that are in the game. The other is trading character loss for numbers. These are not the same thing. To even get to the point where you equate one to another requires some mental gymnastics that aren't really required for this discussion and are very unconvincing. Especially with the added context that outside of taking this risk characters do not die.

Let's be reasonable and not waste more text over a thing like that. You want characters to die. I do not think this is an interesting way to get there. You do. Why do you think this way is more effective, interesting, or whatever than just letting players choose when or where their characters die? How is this better than not giving them the choice at all and just deciding that death is on the table should the required circumstances be met?

[The Adventurer's Almanac]

The real "issue", if you can call it that, is that some players are okay with their PCs dying and others are not. These viewpoints cannot coexist within the same game. Some people will never do anything to risk their character and others don't give a single flying fuck about changing Gorb Headsmasher to Blorb Headsmasher when Gorb gets wasted by a bodak. These people should probably not be playing the same game.

[deaddmwalking]

We're not comparing systems where death doesn't exist and we're introducing a new and novel way to die - we're comparing systems where death exists and you can choose immunity.

Or rather you are introducing a mechanic where some, and in my estimate many, groups will collectively choose not to die ever. And in the process be rendered mechanically inferior to the supposed dramatic challenges that matter.

What's wrong with that?

If nobody chooses to take a power-up there's no discrepancy in power-level among the players. I thought the objection was that somebody would choose to be vulnerable and overshadow everyone else?

In movies and TV shows, there are consequences to failure besides death. In The Princess Bride, the Man in Black gets captured and tortured - but because he didn't die he has a chance to participate in 'saving the princess' adventure, even though he is wounded and seriously reduced in effectiveness. Translating it to game terms, the Man in Black didn't raise his death flag. The game continues, but it is due to player choice (therefore agency) rather than GM 'pity'.

[Omegonthesane]

We're not comparing systems where death doesn't exist and we're introducing a new and novel way to die - we're comparing systems where death exists and you can choose immunity.

Or rather you are introducing a mechanic where some, and in my estimate many, groups will collectively choose not to die ever. And in the process be rendered mechanically inferior to the supposed dramatic challenges that matter.

What's wrong with that?

If nobody chooses to take a power-up there's no discrepancy in power-level among the players. I thought the objection was that somebody would choose to be vulnerable and overshadow everyone else?

In movies and TV shows, there are consequences to failure besides death. In The Princess Bride, the Man in Black gets captured and tortured - but because he didn't die he has a chance to participate in 'saving the princess' adventure, even though he is wounded and seriously reduced in effectiveness. Translating it to game terms, the Man in Black didn't raise his death flag. The game continues, but it is due to player choice (therefore agency) rather than GM 'pity'.

What's wrong with a system that is just simply not used is that it consumes some amount of your time, attention, and ultimately ability to understand enough to play the rest of the game. Not a large amount, but every drip makes it less likely that people will bother with whatever remains. For tables who reject the "character death is on the table, but you get +3" bargain, the mechanic is better off not existing.

What precise outcome are you after which is achieved by "get a +3 if you opt into death" and not by "player can choose if it is thematically appropriate for their character to die at the point where defeat happens"?

[MGuy]

The real "issue", if you can call it that, is that some players are okay with their PCs dying and others are not. These viewpoints cannot coexist within the same game. Some people will never do anything to risk their character and others don't give a single flying fuck about changing Gorb Headsmasher to Blorb Headsmasher when Gorb gets wasted by a bodak. These people should probably not be playing the same game.

This is false. If a person doesn't care if their character dies just let them opt to let their character die. The people who don't want their characters dying just don't have their characters die. It really can be that simple. I've done this exact thing.

[deaddmwalking]

What's wrong with a system that is just simply not used is that it consumes some amount of your time, attention, and ultimately ability to understand enough to play the rest of the game. Not a large amount, but every drip makes it less likely that people will bother with whatever remains. For tables who reject the "character death is on the table, but you get +3" bargain, the mechanic is better off not existing.

What precise outcome are you after which is achieved by "get a +3 if you opt into death" and not by "player can choose if it is thematically appropriate for their character to die at the point where defeat happens"?

A system of 'you can't die, unless you raise your death flag (providing benefit x' is really easy to explain and requires less understanding than most Feats. Since it is used by all players it's also one that you can just say 'what's the deal with raising your death flag' and someone will be able to answer it, the way that something like 'are iron golems immune to flamestrike' isn't.

In normal D&D, death is always on the table. Characters make choices knowing that they could die. Mostly, they make those choices while trying to avoid death. If you give some immunity to death, but offer someone a chance to 'turn off' that immunity for a benefit, they're automatically 'buying in' to the possibility of death, but not necessarily the guarantee that it happens.

Perhaps another example from a TV series... In Avatar, the Last Airbender, the Avatar is certain to be reborn. However, Aang can activate 'The Avatar State', gaining powers from all the past Avatars at the risk of potentially destroying the cycle of rebirth (permadeath). That's a serious enough concern that Aang doesn't proceed to activate the Avatar State in every other fight where he's generally confident in victory. Importantly, at no time does Aang actually die while in the Avatar State. To us, the audience, it signals that the stakes are significantly higher, and in game, it shows that the player believes that 'winning' the encounter is important enough to risk more. It's the equivalent of going 'all in' in Texas Hold 'em.

[MGuy]

So the thing about referencing a narrative is that it is a narrative and not a game. In Dragon Age using magic can mean turning into a demon but for the purposes of the videogame that's not a thing you worry about because of course it isn't and I've heard that in the ttrpg it is indeed dangerous similar to being a psyker in Warhammer. I've already referenced that kind of mechanic earlier and if you want it to be like that that's fine but it has to be highlighted that in this case the context is if you don't do that you never die. We are not talking about what happens in DnD we are talking about the thing you're presenting for everyone to examine and criticize.

Worse the Avatar state you're describing is not effectively represented by the thing you're proposing. The Avatar state radically changes the power level of the single character that uses it. We're not talking about a little number boost but straight up giving abilities the character doesn't and at a scale they can't seem to otherwise achieve. The thing you have just tweaks some numbers.

[Neo Phonelobster Prime]

What's wrong with that?

Lots of things.

Lets start with the third sentence in the three sentence post you cut off to remove the context of me pointing out you were misrepresenting what death flags are and what you are comparing them against.

[Neo Phonelobster Prime]

These viewpoints cannot coexist within the same game.

The problem is they absolutely can coexist.

But if you give one of those viewpoints a hard mechanical numeric bonus and deny it to the other you are basically trying to set fire to a small scale culture war in your player group for no fucking reason.

So lets forget I just typed that last sentence and here's my death flag culture war/elitist hardcore gamerz posturing counter proposal.

If you want to encourage character death, lets ACTUALLY encourage character death for drama and high stakes play. When the death flag is raised characters become vulnerable to death and the only numeric changes they get in return are significant across the board penalties on all their defenses.

Better than that, lets run it in the "are they incompatible view points" mode, and your flag is raised or not based on the "type of player you are" irreversibly and forever and not by the type of player you claim you are in any given encounter.

I mean. At least it makes character death not just possible but actually more likely (I mean is this a mechanic trying to make it happen or not?). And the only players experiencing it are the hardcore edgy ok with character death ones that want to. Right?

I mean it is in the end just the death flag proposal but even better at proposed goals right?

[Omegonthesane]

What's wrong with a system that is just simply not used is that it consumes some amount of your time, attention, and ultimately ability to understand enough to play the rest of the game. Not a large amount, but every drip makes it less likely that people will bother with whatever remains. For tables who reject the "character death is on the table, but you get +3" bargain, the mechanic is better off not existing.

What precise outcome are you after which is achieved by "get a +3 if you opt into death" and not by "player can choose if it is thematically appropriate for their character to die at the point where defeat happens"?

A system of 'you can't die, unless you raise your death flag (providing benefit x' is really easy to explain and requires less understanding than most Feats. Since it is used by all players it's also one that you can just say 'what's the deal with raising your death flag' and someone will be able to answer it, the way that something like 'are iron golems immune to flamestrike' isn't.

In normal D&D, death is always on the table. Characters make choices knowing that they could die. Mostly, they make those choices while trying to avoid death. If you give some immunity to death, but offer someone a chance to 'turn off' that immunity for a benefit, they're automatically 'buying in' to the possibility of death, but not necessarily the guarantee that it happens.

Perhaps another example from a TV series... In Avatar, the Last Airbender, the Avatar is certain to be reborn. However, Aang can activate 'The Avatar State', gaining powers from all the past Avatars at the risk of potentially destroying the cycle of rebirth (permadeath). That's a serious enough concern that Aang doesn't proceed to activate the Avatar State in every other fight where he's generally confident in victory. Importantly, at no time does Aang actually die while in the Avatar State. To us, the audience, it signals that the stakes are significantly higher, and in game, it shows that the player believes that 'winning' the encounter is important enough to risk more. It's the equivalent of going 'all in' in Texas Hold 'em.

Not a single word you said even approaches answering my question.

Why does there have to be a bullshit +3 involved? Why can't it just be "character dies if player thinks that's appropriate" with no carrot or stick nudging them one way or another?

[deaddmwalking]

Not a single word you said even approaches answering my question.

Why does there have to be a bullshit +3 involved? Why can't it just be "character dies if player thinks that's appropriate" with no carrot or stick nudging them one way or another?

Most people like to believe that they 'win' because they are smart and lucky, not because it was ordained. We use dice to create randomness in the way events work out - even though the GM is stacking the deck in favor of the players the dice appear to be neutral arbiters that decide outcomes. When you replace randomness with a script, people tend to object. Not just with character death, but with everything. At least, if they notice. Maintaining the facade that events cannot be known in advance is actually important to playing a game like D&D.

Please note, I'm not ADVOCATING for a death flag - Tussock brought it up, people do the thing where they assume that everything he says is stupid and wrong and I don't see it. A deathflag mechanic seems to make sense for design goals - often discussion of character death comes out as an 'undesired result'. Coyote and Crow has a firm 'characters don't die unless they agree to it', so that's certainly another option. But if you want to have characters raise stakes, putting death on the line (when it normally wouldn't be) does that. One problem with D&D in general is that in every fight you're 'all in' - you know that losing the fight is bad, and you usually can't run away (being slower, having fewer movement options, etc) so every fight is a duel to the death. Taking the worst possible consequence of defeat out of the equation (using the rules) seems like a defensible way to promote 'more fun play'.

Have you ever straight up killed a defenseless opponent because you thought it was possible that they might sneak up and kill a party member while they were sleeping? One of those opponents that is 'too dangerous to let live'? Would you have felt compelled to do so if you KNEW that you COULDN'T get murdered in your sleep - that you had 'plot armor' against such a death? And if that doesn't apply to you, do you think it might to others?

[The Adventurer's Almanac]

Have you ever straight up killed a defenseless opponent because you thought it was possible that they might sneak up and kill a party member while they were sleeping? One of those opponents that is 'too dangerous to let live'? Would you have felt compelled to do so if you KNEW that you COULDN'T get murdered in your sleep - that you had 'plot armor' against such a death? And if that doesn't apply to you, do you think it might to others?

That's dumb. Of course I would still kill them. My character doesn't know he can only die during the opt-in killing sessions.