Tuesday, May 09, 2006

When should a Character Die?

Heya,

When designing an RPG, the issue of character death always comes up. Some people feel it is a necessary part of design (untrue), while others feel it should be avoided because it is un-fun play (also untrue). It can be hard sometimes to know when to include and when to leave out character death. To my way of thinking, there are only four reasons to include it in your design: 1) it adds to the atmosphere, 2) it increases the stakes, 3) it fulfills the player’s goal, and 4) it escalates the overall conflict/scenario all players are participating in. Let me break them down one by one.

1. It adds to the atmosphere
It certain games, mainly those that explore the setting or situation, character death can be an integral part of play. In fact, character death is expected, and the lack of it will diminish the fun. I am reminded of many Call of Cthullu games I have participated in and read actual play reports about. There was even a really hilarious thread on RPGnet many years back where people posted the most outlandish and fun way their characters died. For this game, at least the way I’ve seen it played, character death is well accepted, and honestly it’s anticipated. The thing that really makes it work, IMO, is the speed of Chargen. I know many CoC players who can draw up a brand new character in 5 to 10 minutes. So when their character dies, they jump right back in during the next scene no biggie. To me, that’s the key for using character death in this way. If your game has a time-intensive character creation and advancement process (imagine recreating a level 15 character in DnD 3.5 from scratch) then I would counsel against using character death to add atmosphere to your game. Three things to remember for this use of death: Frequent, Fast Chargen, Expected.

2. It increases the stakes
In my game, The Holmes and Watson Committee, players have the option to put their character’s life on the line. Choosing to do so grants them an additional bonus, but also makes the situation a lot more crucial. This is an example of character death increasing the Stakes. Sometimes it’s important to put everything you have on the line to accomplish something important. In RPGs, that can mean everything from nabbing the bad guy (like in Holmes and Watson) or saving a town from corruption (like in Dogs in the Vineyard). For designers, using death to increase the stakes means limiting the conditions where a character death is possible. Usually, that means making it a player option of some kind. Choosing that option ought to grant the player beneficial but dangerous bonuses of some kind. Sorcerer, Dogs in the Vineyard, and The Holmes and Watson Committee are the best examples I can give you of games that do this. Three things to remember for this use of death: Optional, Bonus, and Lethal.

3. Fulfills a Player’s Goal
Some games have players create a destiny or goal for their character to achieve by the end of play. The death of a beloved character can be a very moving experience. I can easily envision a game that incorporates that fact into the mechanics. The payoff would be the character dies with *something*. For instance, a character could die in redeeming himself. He could die with dignity in a war. He could die saving another person. Or he could die in order to allow another character ascend to a great position. The point of using character death in this way is to allow the player to choose the outcomes, but not necessarily the means of the character’s death. I can see a functional design that allows another player to choose the way a character dies so long as the character’s player gets to choose the conditions leading to and the consequences of that death. Three things to remember: Resolves the Character’s Story, Player Controlled, Consequences.

4. It Escalates the Conflict
Escalating the conflict means that the death of one PC impacts all the other PCs and their enemies in a way that makes the conflict more meaningful. Using character death in this way requires that all players be invested in everyone’s characters in some way or fashion. A dungeon crawl where everyone is out for himself to grab as much loot as possible and then skedaddle is not an effective atmosphere for character death that will escalate the conflict. A game where characters are dependant upon each other or where characters are related to each other in some way (blood, sex, loyalty, duty, etc) is the type of game where the death of one will add meaning to the final victory. In cases like this, character death IMO should still be at the option of the players playing the characters. However, the conditions that the game puts the characters in can make and should make death an attractive option. For instance, if I sacrifice my character, everyone else gains two bonus die and has enough time to rescue the princess from the evil sorcerer. Without my character’s death, the mission might fail. Of course now the sorcerer is really pissed and will take his vengeance out on the whole kingdom. Three things to remember: Sacrifice, Relationship, Meaningful.

In any of the four examples, character death needs to be explicitly mechanically supported by your game. Just including a rule like “When your hit points equal zero, your character dies” is not good enough. What does that death mean? Why might a player want his character to die? What does the character’s player get in return for putting his character’s life on the line? What does everyone else get if he does die? Answering these questions will add a lot of depth and meaning to the death mechanics of your game. Consider them carefully as you design.

Peace,

-Troy

18 comments:

Frank said...

This is one I've been thinking about for a while. My second 2nd blog post is on the idea of challenging the assumption that character death should be permanent in D&D like games.

I'm trying to figure which category character death in D&D like games falls into. It's partly atmosphere, but that doesn't quite fit. And perhaps that "doesn't quite fit" is why I've been exploring whether character death should be permanent, so the player has a say in when to retire the PC.

Having just had the first death in my new RuneQuest campaign last night also gets me thinking. I think RQ fits better into the atmosphere category (and chargen in RQ should be pretty quick).

Frank

Troy_Costisick said...

Heya Frank!

You've hit the nail right on the head. Death in DnD fits none of the four criteria I mentioned up above. That's why dying sucks so bad in that game.

DnD bases character death on the assumtion that "Character death should be used to punish players who fail to make optimal tactical decisions and/or roll exceptionally poor." I personally do not think death should be used as a paddle to punish the players. Instead, it should be used as a tool to enhance the play experience.

Think about how much more fun DnD would be if you got to pass down your dead character's gear + give your new character a Feat-like ability that gives him bonuses when he works at avenging the old character's death. Then once the death is aveanged, he gets a permanent bonus of some kind. Such a mechanic is far more compelling than "at -10 hit points, you're character is toast."

Peace,

-Troy

Frank said...

Crap... a big comment just got lost... GRRRRRR....

Now I've totally lost my train of thought...

Frank said...

Ok, let's try again...at least with one part of my response.

There does seem to be a set of folks for whom death in D&D is important. It's definitely gamist, probably hard core, play, and the players expect bad tactics, or bad planning for potential bad luck, should have consequences. I've definitely encountered these players.

I'm not sure a vengeance mechanic would work for these players.

I've also generally seen that players generally are ready to just move on with a totally new character.

Frank

Ricky Donato said...

Hi, guys,

At first, I thought that D&D death is associated with Troy's first reason: "character death is an integral part of play". Then I realized that this is not true. There is a thread on the official D&D boards discussing the most outlandish ways characters died in D&D, same as Troy's CoC example. But that isn't enough to make it integral to play; instead, it's something you laugh about afterward but you could very well be angry about as it happens.

The solution is not, I think, to rely on resurrection spells, because of a combination of two reasons:
1) Resurrection spells cost you a level or XP;
2) Everyone gets XP at the same rate, so there is no way for you to catch up.

If either one of those two reasons is not true in a particular D&D campaign, then resurrection magic could be a viable solution to character death.

Troy_Costisick said...

Heya,

There does seem to be a set of folks for whom death in D&D is important.

Okay, let's examine why it's important. From what I've read and from my own experience (let me know if yours is different) character death in DnD is seen as a dis-incentive. It's a punishment (or reward for losing, however you want to look at it) and the only main form of punishment the game really provides.

Your character loses the fight, your character dies, you lose, buh-bye. No mechanics for losing the stakes. No mechanics for retreat really (other than your opponent gets another AOO). No mechanics for capture. Indeed, the DM may do all those things, but that's him making things up as he goes- not something explictly and mechanically supported by the text.

Also, let's look at what happens as a result of character death in DnD. I have seen and played in environments where all the following options were enforced: Option A) your character dies and you are out of the campaign. You've lost, you're done. Option B) your character dies and you roll up a new one at an XP and Gear penalty compared to the other PCs. Basically you miss the rest of play that night while fiddling with your character and then when you do get to start playing again, you are at a disadvantage. Option C) Get a rez, lose a level, jump back in. Ricky pointed out the issues with this method.

Anyway, (sheesh! This is a whole 'nother article!) death in DnD is not dealt with very well by the text IMO. It's left up to the participants just how much it hurts or wether or not the DM will fudge his rolls just to avoid dealing with it. It is not a helpful mechanic, but is instead a hurtful one.

Peace,

-Troy

Frank said...

A disclaimer before I continue... For myself, I'm in the camp that D&D style PC death does kind of suck. My interest in exploring this is to try and understand how I can improve my gaming that will include miniatures style tactical combat, that at least requires the possibility that a PC is knocked out of the fight, if not knocked out of the campaign.

Certainly interesting points, but I still wonder, is it really a hurtfull mechanic for gamism? If you and your friends decide to play Risk some evening, some players will be out of the game before the end, and have to sit around for the rest of the evening. If you play in a single elimination tournament of some game, once you lose, you're out, and don't get to play in further games.

So is this bad? If it's not bad for Risk, why is it bad for D&D? Especially early D&D which was barely more than a wargame.

Of course D&D in the 70s didn't take but a few minutes to create a new PC. Of course if your 10th level PC died, and you had to start a new 1st level PC, that kind of sucked. And perhaps the death mechanic is misplaced in D&D 3.0/3.5 which has grown quite far from that original idea that grew out of campaign wargaming.

But another piece that got lost in the great lost post was a comparison with Dogs in the Vinyard. Dogs still has this element where PC death results in the player not having a way to continue formal game input (they can informally contribute by kibitzing, but so can the D&D player - both of course adjusted by social contract about kibitzing - though Dogs addresses kibitzing in the text). Dogs chargen is also going to be slower than 70's D&D, and requires GM attention (for the initiation conflict). I'm not sure what that time is down to for experienced players (but of course we also have to compare that to experienced D&D players, though my assumption is that a D&D 3.5 PC is going to take longer to create than a new Dog).

I guess one argument that can be made is that perhaps one should separate war gaming and role play. But darn it, war gaming is more fun with the role play component added, and there are definitely folks who enjoy that (and I've run wargames with a role playing component for large groups at conventions and they've been well received - in the form of Evil Stevie's Pirate Game using LEGO pirate ships).

Frank

Frank said...

I should also add that I'm glad to see that designers are giving more thought to whether PC death, or even "combat" belongs in their games. And if it does, what does it look like.

Frank

Ricky Donato said...

Frank said:

"If you and your friends decide to play Risk some evening, some players will be out of the game before the end, and have to sit around for the rest of the evening.

[snip]

So is this bad? If it's not bad for Risk, why is it bad for D&D?"

I think this is really bad; in fact, I've named it "the twiddling thumbs problem". It's boring for players to sit out of a game waiting to jump back in. Typically in Risk, it's not too bad because by the time someone dies, someone else (usually his killer) is poised to win the game, so the loser doesn't twiddle his thumbs too long.

I play a lot of Magic: the Gathering and the same problem applies when playing with more than 2 players. The most obvious way to play Magic as a multiplayer game is that when someone dies, he leaves the game, and the last man standing wins. This is known as "chaos" or "free-for-all". The twiddling thumbs problem is quite bad here, and gets worse as the number of players goes up. I've played some multiplayer variants that eliminate the problem completely, and they are so interesting that I now flatly refuse to play chaos.

I think in general an RPG should avoid twiddling thumbs as much as possible. If you play D&D with Frank's variant of "You're knocked out instead of dead, if you want", then you only twiddle your thumbs until the end of the fight. That is a huge improvement. Even better would be, "You're knocked out instead of dead, and for the rest of the combat you can play someone's warhorse, or a henchman, or help the DM play the bad guys." This completely eliminates twiddling thumbs, while at the same time reinforcing that the player has lost because he has to stop playing his character for the rest of the combat.

Frank said...

I think this is really bad; in fact, I've named it "the twiddling thumbs problem". It's boring for players to sit out of a game waiting to jump back in.

Is this bad for you, or is it just plain bad? I happen to agree with you (generally) that this twiddling thumbs problem is bad (and for Risk, I have seen a player knocked out and the game go on for quite some tome). I prefer not to play multi-player games of this sort. But it seems like lots of folks do enjoy this sort of game. Are they wrong to enjoy it? I'm inclined to say no. Also, consider, sometimes the social atmosphere is that it's still fun to sit around watching the game, and partially that brings me back to the kibitzing thought. If it's ok to kibitz, it can still be fun (in fact, I've spent hours watching people play games that I didn't play in at all - for some of those games, I've eventually decided the game is fun, and the next time someone plays, I'll join).

So basically I totally disagree that players being out of a game is inherently bad. It's badness depends on personal preferences and specific situations.

And given that, I don't see why RPGs must be different. Can they be different? Sure. So can board games.

One of my favority multi-player games is the open scenario in Barbarian Kingdom and Empire. It's a "barbarians sack Rome" game, where the barbarians eventually become kingdoms, and then empires. At any time, you can decide your position sucks, punt, converting your surviving units into static "NPC" garrisons, and start a new barbarian tribe.

Many times in RPGs, a player whose PC goes down takes over an NPC (NPC party member, PC of a non-present player, mount or other animal companion, or even some of the opposition). That can be a fun solution. I've also seen players say "no, I don't want to inflict my bad luck on someone else" or some players will decide to leave a bit early (happened the other night in my RQ game). Others will happily start creating a replacement PC (happened with the first death in my first Arcana Unearthed campaign - a 10 year old kid). Others will read a book, play a video game, nap, kibitz, or whatever.

As a closely related, but separate issue, the PC that went down might or might not be subject to "death."

Something to consider, if you want an activity that involves more than two people, almost always there is necessarily time when one or more people are not 100% engaged. In general, we learn to enjoy those times. Of course it's always worthwhile to explore possibilities for making those times we sit on the sideline and watch more fun, either by reducing their occurence and duration, or by making it more interesting to watch what's going on in the spotlight.

To bring this back to game design, I think the important thing is to consider what "death" means to your game. When are people sidelined? If they're sidelined, what can they do, how do they have fun? Can a player be required to "start over" (create a new PC or whatever)? How can that be made more fun? Why does your game require "deat" and "starting over?"

Frank

Frank said...

Another thought -

There are two things that bother me a lot more than the possibility of "PC death" and having to "start over."

The first, and biggest one is games where it "takes a long time to lose." Games where you know it's pretty much hopeless, but social contract/system doesn't allow you to just quit (or perhaps the game dangles false hope). This is unfun play (often involving a death spiral) where your ability to contribute is minimized, but you are forced to keep marching.

The second is when "starting over" leaves you in a much worse position, especially when it may be close to impossible to gain parity of play with the other players. The old "you have to start over at 1st level" is an example of the worst kind of problem here (note that in Barbarian, Kingdom, and Empire, "starting over" (mostly) does not prevent you from gaining parity (I have seen a situation where two players, one a kingdom, and one an empire, were able to conspire to support each other, and kept anyone else effectively out of the game for over an hour - but that's easily solved at the social contract level, and basically amounts to dickery).

D&D like games tend not to be too bad on the first problem, and the second problem is easily addressed by allowing PCs who are created after the game is in progress to be started with more game resources than a true "starting" PC.

Frank

Troy_Costisick said...

Heya,

The main problem I have with DnD's death system is that it forces people out of the game but gives them nothing in return. Let's face it, even if there is an incentive for dying (like in Riddle of Steel), people would still rather keep playing the character they started with. DnD and many many other games fail to recognize this and keep punishing players for taking chances. In my mind, it's very anti-gamist. It keeps people playing it safe and taking on enemies they're assured a good chance of defeating.

Peace,

-Troy

Frank said...

I'm not sure how that's anti-gamist. Gamism in fact requires that there not be a payoff for losing, you need to know you lost. Now there's different degrees of what you risk. Some people play poker for play money, some for pennies, some for nickels, and some for serious bucks.

Now it is valid to raise the question of payoff. Obviously the poker group that's playing with real money, the payoff is winning several times your stake. Assuming the players are reasonably equal in skill, of course, the net will actually be zero. In fact, for poker to have any reward, you have to sucker players who are not as good as you into playing with you.

So the question is what's the payoff in D&D? That goes right to the heart of what is the reward cycle? What I take away from the Forge, and my examination of gaming in a new light is that the reward cycle of D&D is probably NOT levelling up. For my players, I saw the reward cycle in the non-mechanical satisfaction of having created a kick-ass character. Does PC death threaten that satisfaction? Yes and no. Assuming not too much randomness in the system (and D&D's hit points minimize randomness), a dead PC usually results from a poor decision, or a poor build. Obviously the player lost. The player is then incented to do better next time (use better tactics, design a better character, etc.).

But I think the designer does have to think about how death happens in the game. And the players need to understand how death happens in the game. And they have to decide if that's the type of challenge they want. Of course for a non-gamist game, death should be driven differently. A narativist game should have address of premise and theme drive death (Dogs in the Vinyard is a great example here of course). For a simulationist game, the death has to fit the dream (Call of Cthulhu is probably a good sim example).

Of course a game (even a gamist game) needn't have death on the table. In fact, in relation to gamism, and game play in general, one of my thoughts is that ultimately game play is an alternative to risking actual death. I.e., instead of proving ourselves to each other in the colliseum, we play chess, or D&D, or whatever.

But what I do know is that gamism absolutely requires the existence of identifyable loss conditions. PC death is just the most extreme of those conditions.

And as I noted in my blog, it certainly isn't required.

But I can't see it's existence in D&D as bad.

Now one angle to play would be a handicaping system. Watch play over a long period of time, and to make the players more equal, handicap the better players. Golf and Go are both examples of games where handicapping is used in recreational play by at least some players.

We've really beat on D&D and gamist play here. I'm curious about thoughts of death in other creative agendas, or even different styles of gamism.

Frank

Ricky Donato said...

Frank said:

Is this bad for you, or is it just plain bad? I happen to agree with you (generally) that this twiddling thumbs problem is bad (and for Risk, I have seen a player knocked out and the game go on for quite some tome). I prefer not to play multi-player games of this sort. But it seems like lots of folks do enjoy this sort of game. Are they wrong to enjoy it?

Hmmm, I never thought of this. If I'm understanding this right, you're basically saying, "Some people find watching the game just as entertaining as playing the game, and so we shouldn't deny them that opportunity."

My reply to that is that sounds reasonable, but they can watch the game without needing their character to die; they just watch rather than participate (or they watch whenever it's not their turn to participate). The death mechanic that forces a player out of the game therefore allows watching to occur, but that isn't needed; at the same time, for someone who enjoys playing more than watching, he is prevented from enjoying the experience if he dies.

Frank said:
So the question is what's the payoff in D&D? That goes right to the heart of what is the reward cycle? What I take away from the Forge, and my examination of gaming in a new light is that the reward cycle of D&D is probably NOT levelling up.

Dang, you beat me to it. :-)

I think the way you win in D&D is at the social contract level. You demonstrate to everyone else that you can play this game well, that you know how to exploit the system for maximum gain. A player whose character keeps dying might be made fun of, for example, or otherwise indicated that the player is not as good in the competition as someone else; this depends on the social contract.

If you've ever visited the D&D boards, there is a long-standing thread about trying to build broken characters. The idea is people use every supplement they want, and create a character optimized in some way, such as most damage in a round, or highest AC. The results are staggering, like a character with 110 AC, or who can deal a million points of damage in a round. The most interesting part is that the people who participate in this thread never intend to introduce this character to an actual game; they understand that this character renders the game unfun for everyone, and thus breaks the social contract. Instead, they do this simply to see if they can; in other words, they Step On Up in the thread to exploit the system as much as possible.

Frank said...

My reply to that is that sounds reasonable, but they can watch the game without needing their character to die; they just watch rather than participate (or they watch whenever it's not their turn to participate). The death mechanic that forces a player out of the game therefore allows watching to occur, but that isn't needed; at the same time, for someone who enjoys playing more than watching, he is prevented from enjoying the experience if he dies.

No, that's not quite right. Look at single elimination tournaments. Compare those to D&D. Yes, it can be enjoyable to watch the rest of the games after you're out, but the primary desire is to play yourself (and some of the interest in watching the rest of the tournament may be to learn how to play better oneself).

This tends to get to a gamist agenda. But death, or other non-participation, and watching can be a part of any creative agenda. Again, before condemning D&D for making a player sit out for the rest of the session, and starting over, look at Dogs in the Vinyard. It's quite possible for a PC in Dogs to die early in the running of a town. Of course the PC will have died because the conflict was that important to the player (or the player has yet to comprehend the implications of fallout). So in D&D, the PC died because the player "lost." In Dogs, the PC died because the player addressed premise and made a thematic statement.

Frank

Troy_Costisick said...

Heya,

I'm not sure how that's anti-gamist. Gamism in fact requires that there not be a payoff for losing, you need to know you lost.

I somewhat differ from what I think your idea of Gamism is, Frank. But that's okay. Creative Agedas really need to be defined for each individual by each individual, IMO.

But for me, Gamism is about addressing Challenge not necessarily winning and losing. If a mechanic keeps a Gamist from being able addressing Challenge for an extened period of time, then to me it's anti-Gamist.

You and Ricky also differ a bit in your perspectives. And that's fine too. I imagine the three of us would still have a blast playing RPGs together. Like I said, the exact nature of each Creative Agenda will vary from person to person even under the same catagory :)

Peace,

-Troy

checking said...

Setting is one of those game elements that trips people up (including me) because it seems like it isn't necessary, just a nice add-on. "This game doesn't need a setting! It's generic!" I think that if that's true, then you don't actually have a game. In fact, what actually happens is that the players create a setting (with or without realizing it) to play this "setting-less" game in. So in fact the game does need a setting - it just doesn't provide one.bohyme

Diogo de Souza said...

Greetings,

I am a veteran RPG gamer and have recently found your blog at SocraticDesign, where I thoroughly enjoyed reading this article.

In fact, I enjoyed it so much that I'd like to translate it into my native language (portuguese) and post it on my blog (metagamers.com.br) with all links and credit to you, Troy.

I couldn't find in your site any license under which your material is published, so I needs must write you this email to explicitly request for your permission to translate this article.

Thanks,

Diogo.