Monday, June 20, 2011

A Death Penalty

The last post was about death penalties in general. Among some boring properties I stated that
The perfect death penalty is an extremely effective deterrent, but completely harmless once it has actually happened.

Scrusi, in his comment, finally pointed the dilemma out. Isn't there a perfect symmetry? Isn't it exactly the repercussion, the aftermath, which deters one from dying? The answer is that for a perfectly rational person, this is correct. But fortunately, players aren't perfectly rational.

The trick is to exploit a psychological fault within the human brain; a fallacy, to create a death penalty that is very deterrent, but rather harmless once it it has actually happened. But this won't make the perfect death penalty, of course. Please remember that until your computer mind-controls you, there won't be a perfect death penalty. Humans are just too smart for that.

First, I need to very superficially describe the game the death penalty is embedded in.

There's a player-run castle with lots of player-owned shops. In these shops you buy equipment. You can buy very cheap equipment or very expensive equipment. Expensive equipment gives a rather small boost to your character's effectiveness. There's also an entire virtual world outside that castle, but that is mostly irrelevant for this post.

Now, in the basement of that castle there is the entrance to a vast labyrinth. That's very similar to Diablo, I know. But this is far from an action RPG. You want to venture into this labyrinth to gather resources and stuff you can sell. By doing so you can improve your castle, you own house, finance a war and lots of other stuff. On a side note, you also want to venture down there, because it's an unpredictable adventure and fun in itself.

There's also a very credible weight limit. So, while you venture down there, you will not pick up every plate mail and two-hand sword you see. Instead, you will rather look for gems or whatever is valuable and carry-able. Even if you stay down there for a long time, your backpack will stay rather clear.

And there's the need to carry supplies with you, because your equipment will eventually break. Unless you can repair or replace it down there, you are in trouble. So, remember: All your equipment eventually breaks! All of it. Equipment is just a tool! Nothing lasts forever.

Of course, while you venture down there, there is the usual character power progression. It is highly compressed in this game and a lot of it runs horizontally. That is, you further customize your character, but he doesn't necessarily become more powerful.

The deeper you venture, the more dangerous the enemies are, of course. Also, the deeper you venture the more potential rewards there are. But it is perfectly normal to just venture down there for half an hour. The labyrinth is procedurally generated and the enemies use some basic AI to move around in it. You can explore endlessly without going deeper, if you want. To go really deep, you want to take a well organized, well supplied, group. But it shouldn't be too large, because making a lot of noise is not a good thing when you're really deep.

I know, you love that game already. But please, please! Now comes the topic of this post. If you want to comment, do so on the topic, please ;)

When you die in the labyrinth you lose your stuff and get teleported back into the castle. I know, you hate it and are very disappointed. But here's why that is actually a good death penalty once you emotionally disconnected yourself from WoW.

Shefrin & Thaler, back in 1988, wrote
"[Humans] mentally frame assets as belonging to either current income, current wealth or future income and this has implications for their behavior [..]"
That's called mental accounting.

Weight considerations make players necessarily start their adventure only with what they really need and with what they absolutely expect to lose. There's no real emotional punishment once you do lose it.
When you bought that axe to use down there, you already knew that it is going to break eventually. That's why you also had some equipment to fix it once it does. But since it does become ever less effective, you already wrote it off - mentally. A second trip would be started with a new axe, anyway! Nobody would want to buy a used axe! The axe was an investment.

The stuff you found in the labyrinth, may be in your backpack, but it is not safe yet! You need to get to the surface to actually own it! It is potential future economic gain. It is not something that you can lose, because you don't emotionally own it yet!

It's like a guy unexpectedly promising you to give you 100€ if you beat him at poker right now. But unfortunately you lose. From an purely rational point of view (Gevlon will insist!) this is exactly the same like losing the 100€ note you just found on the street. But from an emotional point of view, it's something completely different.
Think about this example for a second.

You'd hate to be prevented from playing the guy, because that felt like losing 100€. (You'd absolutely hate to die, because that would prevent you from carrying your stuff to the surface). But if you actually do play and lose, it's really not that much of a problem, because you didn't actually lose 100€ - at least you don't feel like it. (If you actually do die, it's really not that much of a problem, because you didn't really lose this stuff - at least you don't feel like it.)

Now, this isn't perfect. It might, in fact, make sense to really empathize that your stuff is not truly yours yet. For example, by using a gamey separate backpack, that's called "stuff found so far". Maybe there are ways to do so that are more compatible with the simulation.

This death penalty also has another interesting characteristic. In the last post's comment thread, Eteocles pointed out that the same death penalty can actually be differently harsh. Of course, he's right and this effect should be used to create a good death penalty.

You want the death penalty to be more harsh to those players who like harsher death penalties. Now, some commenters will point out that nobody actually likes a harsh death penalty. The only thing they like about it is that other players have to face it. Call me naive, but while I do agree that this is true for a few players and party true for most players, I don't think it is entirely true.

The proposed death penalty becomes ever harsher the deeper you go. You lose more invested time and you probably lose more valuable items should you die.
And since you are likely to be in a group, if you venture beyond a certain depth, your loss will be shared. "Do you remember when we found that big red gem at level 50? And then we went on to level 51 and the big troll army ran us over from behind? We were so like ooggh!".
Shared loss isn't even necessarily bad! It creates memorable moments. Read Elder Game if you want to know why that is very important. Memorable moments make players resubscribe after a break.

Finally, this death penalty is auto-adjusting. Just like the challenge level of the entire game is. You remember this post?
If you want to optimize your financial gain, you don't take a too large group with you. But going really deep with just your buddy is .. hardcore. On the other hand, if you just want to see the content and don't care all that much about optimizing the financial gain, you go down there in a large group. Which makes dying improbable or even impossible. .. If you don't go too deep that is. But with a casual mindset you don't want to spend too much time on the game anyway, and consequently will never make it very deep for time-reasons.

The death penalty is a clear cut. No aftermath. You are not weaker for some time, you don't have to regain (re-grind) stats or money. In fact, nothing in the game reminds you about your death. Once you are back on the surface you are free to experience new adventures. In the labyrinth - or outside of the castle.
The game makes it easy to get some basic equipment, of course.

The death penalty, in this game, exploits mental accounting to make people afraid to lose potential future gains. But since it's just potential future gains, they aren't that painful to actually lose.

The death penalty is auto-adjusting: the more hardcore a player is, and the more risk he is willing to take, the harsher it becomes.

The death penalty is a clear cut. Nothing in the game reminds the player about a past death.

The death penalty is impossible to abuse, unless the player wants to get back to the surface and lose all his stuff.

Finally, assuming the lore supports the naked-teleport, it makes a lot of sense from a simulation point of view.

Is it perfect? No.
But even if you don't like this specific death penalty, I hope I could spark your imagination. Maybe you can come up with more cognitive biases to exploit to create an even better death penalty.

For purely personal reasons I removed a blog from the "Blogs I follow".


  1. I'm honored by such a detailed response and I like the psychological approach. If you can manage to make players dread death but not actually hate the repercussions you are golden.

    I'm not completely convinced though that that is possible ;)

    I played a game recently that allowed me to build my own house (check) on top of a dungeon (check) in which I could go deeper which would increase difficulty (check) and where I would lose part of the spoils gathered if I died (semi-check). The game was Terraria and man did I get frustrated by dying in that game.

    I would make my way down in the dungeon and then randomly die to not paying attention (or whatever). Following the death I would respawn at the top and make my way down again. I got close to quitting the game for good a couple of times because of that.

    I suppose I'm a hard one to please in that regard as I have a deeply ingrained understanding of opportunity cost and I also absolutely hate waiting or redoing content I have already done.

  2. > It's like a guy unexpectedly promising you to give you
    > 100 E if you beat him at poker right now. But unfortunately
    > you lose. From an purely rational point of view this is
    > exactly the same like losing the 100 E note you just found
    > on the street.

    It's not at all the same. Losing a 100 Euro note is your failure. It's not that easy to lose a 100 E note and if you do you must have made something really stupid.

    On the other side not beating someone in poker, who offers you 100 E if you beat him, is very likely as he's probably a much better poker player. There's not much you can do about it.

    - One situation is very likely that you have the money (keep it).
    - One situation is very likely that you will not have the money (win it).

    You can't compare them.


    On topic. Your DP is very hard. Investing 3 hours to walk down is a very high investment to lose, time is valuable. And losing my 3 hours for a lag spike would probably make me rage quit.

    And in any case I would stop playing for the day because after "wasting" 3 hours my day won't have enough time left to try again what I tried before. If there's one thing I really really really hate in a game it's when the game tells me when I have to stop for the day. Did I already mention that I hate daily quests?

  3. Oh, and if you're down there together with a buddy, after you've died your buddy has two options.

    - Doing suicide to continue playing together with you.
    - Trying to reach the surface, alone, which he is unlikely to succeed and if he does it'll probably take much more then the 3 hours it took the two of you together to get down. The game has a very harsh DP for your buddy (unless the game has some kind of teleport to return safely to the surface).

  4. Hmm the main thing imho is human are risk averse. If you make harsh death penalty they will avoid it to irrational lengths. While this is a game and you supposed to be able to do stuff you cant do in RL without risking much

    Even in your example - people will min/max to run dungeons which provide best reward while having least likely chance to die

    For example even in WoW people bitched and moaned about gold repair cost and "ghost runs".

    I was thinking more in lines of travel time penalties:

    World would be big. With binding stones all around the place. Player cities would be naturally concentrated around those stones.

    But there also would be big world beyond the area with the initial stones. Traveling further and further away into unknown would naturally result in bigger risk.

    I would expect 90% of player activities concentrated in hub area. While explorers would be free to venture far away. The explorers would be like pioneers - they would go and discover new locations.

  5. Scrusi, to be honest, it wasn't so much a response, as something I planed to write anyway. You just happened to be the first who pointed out that there should be a symmetry. ;)

    About Terraria. It guess it doesn't really try to exploit your mental accounting. We, the players, are used to think of our backpack as the stuff we OWN. The game I described would make very clear (tips at loading screes etc) that this is not the case. Nothing is yours until you stored it in your house/bank.

    The guards when you enter the labyrinth will remember you that you will lose everything you carry with you.

    The stuff you wear will become damaged slightly already during the first fight; and damaged equipment is not worth much and there's no reason to not buy new equipment before you go anywhere.

    Moreover, to just remove parts of what you own is worse than taking everything away, I think. Sounds contra-intuitive, I know. But the point really is that the player must consider this dungeon as a sealed-off, closed adventure that ends when he dies.

    The clear cut after death is important. If, after death, you can try to get back to were you were, there's no clear cut. The teleport is just annoying. And the item loss is especially annoying.

    If you lose the poker game and the guy tells you that you can try again if you pay him 5€ (invest time to run back), this game can very soon become very annoying.

    I'd be perfectly happy if the player logged off after death, by the way. It's an MMORPG, not a single player game. There enough other stuff to do and enough other stuff that keeps you playing.

    But if you venture into that labyrinth then you risk all you wear (not much, unless want to wear slightly more powerful, expensive stuff). And you will only be able to profit if you get back to the surface alive. Of course, you always advance your character, and nodody can take this from you, ever.

  6. Kring, it's funny to see you argue that it's not the same. I mean, you are right in some way - and then you are not. From a pure business point of view this is the same. You started without a 100€ note. You had a chance to gain one, but you failed. And so you ended up without one.

    It's very similar to the Sunken Cost Fallacy. You felt like you owned the €100 (which you did.. but does it matter?).

    It's smilar to this:

    "You are going to see a movie, and the ticket will cost you £10. Right outside the cinema, you realise that you have lost a £10 bill. Would you still buy the ticket?"

    "Now imagine that you have already bought the ticket – but outside the cinema you realise that you have lost the ticket. Would you buy another ticket at £10?"?

    If your answer is "YES" to the first question and "NO" to the second you are irrational. But so is almost everybody.

    Of course, this feels different, and from this point of view it is different. But from a rational economic point of view, it's all the same. It's just that the one scenario is so much more annoying.

    The game would make very sure that you don't die to lag spikes - well unless you have 10 second lags; then there's nothing I can do.

    A game with a harsh death penalthy, must always feel fair. That means the player must always blame himself for dying and not something 'unfair'. He must feel like he should have know and should have been smarter.

    For groups content: You cannot die unless everyone in your group is dead. The monsters will then finish you off. Before that you are (most likely) only 'defeated'. Good point.

  7. Max, you aren't free to choose the best dungeon. Another guild has put their own castle on the 'better' dungeon. Now, you can try to besiege them. But for that you need ressources. The dungeon is a way to get those ..

    Humans are risk averse; correct. But they are also curious and greedy. And, as I said, you can't lose anything by ventureing into the labyrinth. The gear you enter the dungeon with is cheap! You can only ever lose what you collect while in the dungeon.

    On the one hand side you want to explore the entire dungeon (curious) and find all the treasures (greedy). But on the other hand, you are risk averse; and maybe rightfully so ..

    This is a constant interesting decision. The more hardcore you are the more risk you accept. The higher you want to fly, the further you can fall.

  8. The example with the cinema ticket is excellent. Because the chance for losing a 10 pound bill or a cinema ticket is (nearly) the same. I put both in my wallet. Therefore you can compare those two cases.

    But your example is different. Everyone is able to "not lose 100 Euro". I do that all the time. I'm actually very good in not losing money. And if I ever lose money I'll really hate it. Because it's my fault!

    On the other side I do not play poker and therefore don't have experience in that game. I have a snowballs chance in hell to beat a poker player. I would probably not even try to beat him for 100 Euro because I would just lose time without a chance of winning. I might play for the game but not for the chance to win. I don't play the lotery for the same reason.

    Those two cases have a completely different probability of happening (chance of losing money: 0.0001%; chance of beating the poker player in poker because of a very lucky draw: 0.1%)*, that's why you can't compare them.

    *) scientifically made up numbers.


    > The game would make very sure that you don't die to lag
    > spikes - well unless you have 10 second lags; then there's
    > nothing I can do.

    I sometimes play over WLAN which loses connection from time to time. That sucks and is completely my fault. The other side is that I won't be able to play any game where this is a huge problem. And if you want my money you have to consider that. :)

    That's why WoW uses such an "outdated" graphics engine. Not because they can't do better, but because they love to get 2x 13 Euro per household which means it must also run on the second newest PC in every family.


    I like the defeated idea although it's very un-immersive, isn't it?

    A possibility would be to give everyone a rezz.

  9. Kring, I'll re-invent the example for you:

    You find 100€. Some guy approaches and asks you to play a round of dice (perfectly fair 50% chance to win). If you win, you get another 100€, if you lose, you have to give him the 100€ you just found. You agree and lose. You have to give him the 100€ you just found.

    Now imagine this same guy, but you don't carry any money with you. He still offers to give you 100€ if you win a game of dice. You lose.

    Point being: If you at some point felt like you OWNED the 100€, it's much more annoying to not have them in the end.
    On the other hand, if you considered the 100€ to be potential future income that you didn't turn out to get, it's not so bad.

    I agree with the graphics. Stylized art is still the best option. Narrow dungeons are good for graphics, by the way.

    You think everybody rezzing everybody is immersive? You played to many modern RPGs ;).

    In fact, the 'defeated feature is more complex and includes an interesting decision. But that's off topic now.

  10. On the one hand side you want to explore the entire dungeon (curious) and find all the treasures (greedy). But on the other hand, you are risk averse; and maybe rightfully so

    So what exactly your death penalty would be?

    I would say repair costs (which could be tuned as basically "cheap equimpent" -sans the hassles of getting your "spare set" ,re-equiping it etc.) plus lost "treasure" .

    Thought what if dungeon/location took 1h+? Would you want player get absolutely nothing out of it? Wouldn't it be discouraging?

    I d say he should get some token reward so he feels like time not totally wasted (same like you get points even for a loss in LoL, WoT , WoW battlegrounds). With more substantial rewards being for actual success

    As about free to choose. That I think is actually hard problem. It stems from the fact that people needs to be matched to appropriate challenge. If you have harsh death penalty combined with mismatched difficulty its a recipe for disaster - people will quit in droves in frustration

    So I was kinda thinking the PvE rating for dungeons. Of course in my ideal game the PvE dungeons would be actual mostly run by other players (playing a dungeon master role , but having no say in doling out rewards, DMs ultimate goal would be to kill the party, but not earlier than they complete x% of dungeon and DMs would have limited scaling tools in chain so they couldnt just put imba boss at the end while having easy trash all way to it)

    So it would be like ELO for pve dungeons. One for players and one for DMs. Pro players would be matched with pro DMs and so on

  11. What am I doing with this stuff I find in the dungeon? It seems like a big part of your plan is to make me not think of my equipment as stuff I own, so obviously I'm not getting this stuff from the dungeon to get equipment.

    Instead, the adventure is in service of getting a better house or castle, but those things don't help me on my adventure. I also hit a limit to my character power progression (since it is horizontal) so I'm not going into the dungeon to make my character better at going into the dungeon.

    And so I think you might be missing out on the fundamental appeal of RPGs for me, and I think for a large segment of the population that plays them. You do a task to get better at doing that task so you can do the task more and get better and doing it more. If dungeon delving doesn't make you better at dungeon delving then as soon as the progression runs out you are going to feel like you've hit a wall. If you never make it past level 5 of the dungeon, eventually you stop getting excited to get to level 5.

  12. Sthenno, rewards that make you better at getting rewards lead to characters that become exponentially better at getting rewards. It's possible and usual, of course. But it brings a along a lot of other problem that have been discussed at this blog.

    If you think that building a bigger house is boring then this game is just not for you.

    Now, your charcter does get better. Always. There's just heavy diminishing returns. You keep the better charcter after death, obviously.

    In addition you get better yourself and got to know the labyrinth better (that won't be on the internet, but is still not randomly generated whenever you enter; hehe).

    You will have become more skilled yourself. And you might have made friends. With these friends - and theirs - you might decinde to explore more of the labyrinth - or any other part of the game.

    I think there's plenty of reason to explore the labyrith in the end. But if the only thing that keeps you playing is getting (much) better at getting rewards, then this game is not for you.

  13. Max, compared to what you had when you entered the labyrinth you never lose much. Looking at it this way, this is a very lenient death penalty.

    However, if you die your current adventure ends and you lose everything you gained during it. Looking at it this way, the DP is rather serious.

    In the end it just depends on how long you have been in the labyrinth. 30 minutes? Then dying is really just annoying. But the labyrinth is tuned in a way that you likely won't die after 30 min, unless the only thing you did is going ever deeper. It's perfectly reasonable to farm resources at higher levels. It's not risky and does give 'rewards'.

    If you have been in there for several days in a row (you can safely log out), then the death penalty might become really annoying. But if you do that you are hardcore enough to live with it.

    You know this blog, Max. Do you think I like badges? ;)
    There are more immersive ways to reward people. I just listed some in the last comment.

    The labyrinth is self-balaning. No need for ELO ratings. The deeper you can go, the more determined, risk-loving and, obviously skilled, you are. You can mostly choose your fights - until very deep when the mobs will also start to move around a lot and there's generally a high mob density.

  14. Is it just me or is part of the human engineering the "transaction" aspect. I spend x (risking time and lost gear) in the hope of gaining y. Versus ganking.

    Buying Apple stock and seeing it decline is annoying but part of the risk/reward "transaction." Losing the same money because someone broke into my house would be far more annoying. (i.e. ganking/griefing)

    So if there is a higher level NPC by a resource node, I may die three times before I mine it. For me, that will generate minor frustration and less than getting randomly killed three times running around a dungeon.

    I maintain a non-dying % of about 80-85% is the least fun. If you win there is little satisfaction and yet you lose a non-trivial % of the time.

    I do think for many in your example, the most annoying will not be the lost loot but the "having to redo upper levels while knowing the only reason you are doing so is because you died."

  15. Hagu, you write
    I do think for many in your example, the most annoying will not be the lost loot but the "having to redo upper levels while knowing the only reason you are doing so is because you died."

    Point is that this is not just a game about this dungeon. The dungeon is one feature of this game; it's an MMORPG. It's not like Diablo or any dungeon crawl, or Terarria. Once you died there' really no incentive at all to get back into that dungeon. That's really very important!

    You went into that dungeon, you tried to get some stuff, you advanced your character. After your died, your character is now still a bit more advanced, but you didn't get anything else out of the dungeon. That's sad, but now you can do something else.

    Sure, you can try it again, but that's not encouraged at all. There are probably new monsters around. You could now decide to just stay at the (rather safe) upper levels and farm some resources. Or get out into the world and do some trade; anything.

  16. I like your ideas about death penalties because they take into account what I feel is an important point:
    a death penalty should be there as a potential consequence for taking a risk and not just to slap the player on the wrist for messing up.

    I believe that death penalties should be there in order to add balanced risk to the game, and they should have an understandable context.

    I have a problem with death penalties when it feels like I don't have enough control in order to avoid them. If I get penalized for death I want to be able to feel that the penalty is balanced for the amount of risk I was taking, as well as rationalize that there was room for aversion (either from skill or better planning).

    I think many games are too keen on the "penalty" part, and kicking players while they're down.

  17. In addition you get better yourself and got to know the labyrinth better (that won't be on the internet, but is still not randomly generated whenever you enter; hehe).

    Hehe and there we get into nitty gritty detail. So labyrinth will be magically generated so its always new? And still be interesting to actually go in and explore? :)

    Here is my problem - to make interesting dungeon it needs to be pretty finite and well designed by actual humans (we do not get tools to give us interesting procedurally generated dungeons yet).

    So you cant have "increase difficulty" going deeper. -how many levels you want it be "deeper" -cant be infinite? how many % of your people will see last levels? and so on

    And all the basic scaling problems come in - designing for various composition, group sizes ,etc.

    Therefore I pulled another "magic" trick -let players do it. Reason I think it will work better than "procedural" trick just because we already have working examples. Players do create great content . For free. (e.g. mod communities)

    Big design problem is give players tools simple enough for them to create content complex enough to be interesting. You give them tools to simple - nothing interesting comes out. Too complex? - too few will bother

    And you have to have boundaries so those tools do not simply become cheat shortcuts (like some of player created missions in COH)

    On paper my solution is partly in the main feature (minecraft like ability to design buildings and environment from elementary blocks) , tied in with certain tools to match players with dungeons and rate dungeons by quality

  18. Hehe and there we get into nitty gritty detail. So labyrinth will be magically generated so its always new? And still be interesting to actually go in and explore? :)

    Yeah, that's quite a designer-deed, is it not ? :)

    The labyrinth isn't always new. It's always the same, although the monsters behave according to rudimentary AI and thus unpredictably, yet not arbitrarily.

    Point is that the people in the castle, have a very strong incentive not make a map about it public. They might want to create one for own use .. but they don't want it to get into the wrong hands ..

    Since there's relatively few people in control of such a labyrinth's entrance, there were won't be enouth potential for crowd-sourcing. So even the best maps will be pretty bad. Which is a design goal.

  19. You make a lot of excellent points in this post, and I really like the idea you put forth for a "death penalty." I'm not even sure WoW couldn't utilize something like that (you mention that something like this would be too harsh for WoW in your previous post).

    What if you lost everything in your backpack but not on your character? Since we have access to bank vaults in many situations now, people could store their "off sets" in bank vaults (or even make a feature for guild bank vaults to allow a "closet" section for individuals or the like). You'd only bring enough food and flasks as what you'd need. If people had to nip back to town, we have PLENTY of ways to summon. This would allow people to carry only what they could reasonably expect to lose even when raiding. During leveling, it'd reward living for a long time, and it would certainly induce some terror in farmers who spend hours upon hours just farming without every checking in.

    All of this is just off the top of my head, so I'm sure it's riddled with problems, but you made a very interesting point and I've been writing about perma-death recently (and my character finally died after living for 38 levels - bummer) so I thought the topic was too good to pass up as a WoW topic.

    Great post!

  20. The death penalty, in this game, exploits mental accounting to make people afraid to lose potential future gains. But since it's just potential future gains, they aren't that painful to actually lose.

    Imagined pain is still real pain. The soldier who feels phantom limb pain in an arm that is no longer there still feels pain. There is no distinction. The only people going down into that dungeon after dying the first time are the people who enjoy playing Fallout Tactics or Diablo 2 on Hardcore mode, e.g. masochists. The only difference between your proposed death penalty and games with absurdly distant save points is (presumably) the carryover in terms of XP or skill point gain such that you did not *completely* waste your time.

    And time is exactly the factor that you are ignoring here. Death in any videogame already has built-in deterrents: it indicates failure, and you spend X amount of time getting back to where you were. Hell, even when you look at Battlegrounds in WoW you see how much death is a deterrent despite there being mass-resurrection every 30 seconds. Are there people who "suicide" into zergs in BGs or try and defend a flag to the death? Yes, of course. But those kind of deaths are not the sort of thing you want to deter people from doing. Why would you?

    And so it really comes down to something not expressed in this article, vis-a-vis the goal of death penalties. A (even purely psychological) harsh death penalty does *what* for your game? Make people more paranoid? Stress the player more? Put more emphasis on quick decisions (since if you can see death approaching a long way off it's less of a concern, unless it's inevitable)? Make the game more "realistic?" When you look at WoW, do you see anyone *not* caring about getting ganked? I don't. I feel miserable getting ganked on an alt by a high-level character I have no chance of touching, let alone a main vs main scenario where there is the added embarrassment of losing in a (more) fair fight. And I technically lose nothing more than time!

    People are still afraid to die in games with quick-save features where you can literally reload 5 seconds after any bad decision, let alone in games with "soft" death penalties in WoW, so I am not sure what this is all about. It is an interesting discussion, but I am not convinced anything needs "improved."

    If you really want a psychological solution to death penalties with no practical long-term effects, I have a much easier suggestion: make death embarrassing. In your dungeon delving game, let players keep everything they find down there even after they die... just give them a "black mark" of some sort. "Times killed by slimes: 5." Maybe give forced titles like "Bob, the Defeated." In PvP situations, let the winner take the loser's dog tags, or have a personal announcement when you die like "xxSephirothlolxx has killed you twice in a row. Ouch." Perhaps have some way of showing off the dog tags you have taken, including being able to link them in global chat for everyone to see. See someone talking smack about how good they are in Trade chat? Link the dog tag(s) you took from their corpse. Maybe put the top-scoring (or richest from delving dungeons) players up on a public leaderboard, thus encouraging people to kill those players to knock them down while simultaneously increasing the psychological death penalty for those same players (who are obviously the most serious about the game, and implicitly need a greater challenge). Bam! Done.

  21. Azurial, you say:
    Imagined pain is still real pain.

    and I agree. If this imagined pain is less 'ex post', but higher 'a priori' then the death penalty is a good one.

    The goal of death penalties? They should make you not want to die really hard. As I said: The perfect death penalty is an extremely effective deterrent, but completely harmless once it has actually happened.

    In addition to that, many people like death penalties that make a lot of sense from a simulation point of view. Most of them like harsher death penalties, because it makes them feel more at risk when exploring.

    Now, I guess, you are not one of them. But these people exist. You can't just write them off as masochists. And even if they were masochists; they are paying masochists.

    Your proposed death penalty is interesting. It scales with the pride of the player. And that pride usally grows with involvement in the game. Which leads to hardcore players being more heavily punished than casuals. Nice. However, I have problems thinking of a simulation background, in which this makes sense.

    Also, this is the opposite from a clear cut. Means you are not free to experiement ever, because people will see the result of this experiment for forever on the char.

    Still an interesting idea, thanks.

  22. How can only being defeated with a dagger through your heart be immersive? :)

  23. Your rewriting of the 100€ example made clear what bothered me about it the first time around: The two examples are functionally different (unlike the cinema ticket example).

    In case A I stand to lose €100 in the game of dice and will, if the game is fair, break even on average. Deciding to participate is a net neutral option.

    On the other hand, deciding to participate in case B is net positive as it will gain me €50 on average.

    Now, if the two events ("find €100" and "play dice") were connected, I might see what you mean. Maybe if you found a €100 voucher for a game of dice, compared to being offered a free game. But then those really are the same thing so maybe I'm not on the right track either.

    I suppose the point is that the "play dice" decision seen independently (as it should) is a good one in one case and a neutral one in the other case. In your cinema example, the "buy ticket" option, seen individually, is the same in either scenario.

  24. Kring, Scrusi, I think you are right. Thanks for pointing this out.

    Perhaps it is a mistake to use outside examples at all. So, let me try once again ;)

    Imagine you are about to venture into the labyrinth. Your king approaches you and says: "Kill these three beasts and if you return alive I will reward you with a bag of gold. "
    You venture into the darkness, kill two of them, but the third kills you. Your King tells you: "You didn't do as I asked. Therefore you get no gold.

    On the other hand, imagine you venture into the labyrinth (no king). You kill two beasts and they each drop 1/3 bag of gold. You put that gold into your backpack. Unfortunately, the third beasts defeats you and you lose it all.

    Both systems can be designed in the game to make the player receive 1 bag of gold only if he defeats all three beasts without dying. Thus, it is really the same from a reward point of view.

    Which failure feels worse ?

  25. Azuriel already made my point -- that "time is exactly the factor that you are ignoring here.". But you didn't really address it in your response to him imo.

    I love the idea of your death penalty, and agree with what you are basically trying to argue. But for me, loss of time is an extremely harsh penalty. If I only have 3 hours to play today, and I spend those three hours with no reward, then there's really nothing harsher.

    What would you think about augmenting this concept such that there are 2 distinct types of things you gain while exploring the dungeon (I won't specifically suggest what they might be because it isn't relevant). One type of thing would be lost completely if you die without getting back to the surface, but the other type of thing would remain with you. Depending on what this second thing is, make up your own rationale as to why or how it persists after death.


  26. Thanks for the comment, John.

    The system you propose is already in place. You keep your character progression (gained skills/exp/etc). You also keep the knowledge of the labyrinth and the game in general. You keep the social ties you created. Finally, you had fun, I hope, and nobody can take that away from you.
    Focus on the character progression, if that's the most important thing for you ;)

    About the time argument: Most of this post is about trying to exploit a cognitive bias to make the player fear death more than he regrets it after it happened.

    To lose time in some indirect way is almost impossible to avoid. Especially, if you want death to be a clear cut that players forget after it has happened and only feared before it was about to happen.

    Finally, the simulation aspect of a MMORPG is important to me. I understand that coming from WoW this may be hard to understand, but to be revived seconds after you were killed and to empower other players to magically see how often you died, doesn't sit well with me.
    I want a virtual world, not an arcade game.

  27. Nils, a much better example. Here, losing the money you already picked up should feel worse than losing the option on money.

    That said, visibly losing "quest progress" might feel just as bad or even worse than simply losing gold. One has to be careful how one presents the deal.

  28. I think I misunderstood part of this. It seemed to me you were saying you enter a completely new dungeon every time. If you can learn about the dungeon by exploring it, an the dungeon is persistent, then I feel a bit differently about it.

    Just to clarify, though, if everyone has the same dungeon then it will be on the internet. If everyone has a different dungeon then when we play together do we get to decide whose dungeon we go into?

  29. Scrusi, I agree. I seems that exploiting cognitive biases is possible, but just not very powerful. Still something to keep in mind, I think.

    Sthenno, I probably expressed myself poorly.
    The labyrinth is perfectly persistent (and huge). The players have an incentive to restrict access to the dungeon to the own social group. And they have a strong incentive to prevent players outside their social group get to know the layout of the dungeon beneath their castle.

    Thus, nobody would want to put the labyrinth online and more importantly, not enough people have access to make crowd-sourcing feasible.

  30. Hehe, some of your game's rules and suppositions are genuine fantasy :) the mechanisms don't do much to encourage you to keep playing. That is, it would only attract the kind of gamer who enjoys gambling.

    here's what I mean. You say the game explains that even though you loot an item, you don't own it until you store it in your house. this is so irrelevant to the player that its hard to attack this with enough fire. If I, as the player, pick up an item, it is mine. it doesn't matter that Ive been briefed that I dont have anything. I saw it, i picked it up. Dying and starting without it does not erase the sense that I have lost all I gained. I did, in fact, lose all I gained. I agree with Kring: if I sit down at a poker table with a shot at winning 100 bucks in exchange for the pleasure of my company, I lose nothing. If I find 100 bucks on the ground, but get robbed on the way home before I can spend it, I still lost 100 bucks. And it still sucks.

    There is a modern game on the market which hit about 2 years ago, called Demon's Souls. It plays almost exactly like your scenario. You start at level 1 and kill enemies, find objects, etc. If you die at any point before completing level 1, you start over with nothing. What keeps players playing that game? Not much, and that is why to this day its a very niche game. Few players want to *spend time* on that sort of game. That's the ultimate penalty if your game doesn't give more for effort.

  31. Doone, Demon's Soul has sold over 1.1 million boxes by now. That's not something I'd call niche for a PS3 game. (Link)

    Sure, it's not a Call of Duty with 10 mio, but it's still extremely profitable. Add in how much more CoD cost to produce and add in the much higher popularity of the series, and add in the money spend on adds(!).

    Anyway, I could live with Demon's Soul being niche. What I don't understand is where your hate and despise come from ?

  32. I don't think you can make a world with any kind of persistent information and not expect that information to end up online. That being said, I think just the fact that you can learn about the dungeon solves the problem I was raising. As I said, I think if there is no positive feedback then you will hit a wall at some point. If you can explore and make maps and plan how to get further in next time or look for things you haven't seen then you won't hit that wall (if it's randomly generated every time rather than persistent then the things you haven't seen quickly become the same as the things you've seen since they are all random).

    I would just hope that the dungeons would have really interesting things in them, not just corridors and monsters.

  33. The goal of death penalties? They should make you not want to die really hard. As I said: The perfect death penalty is an extremely effective deterrent, but completely harmless once it has actually happened.

    That isn't good enough. And I think you understand this, because your objection to my "embarrassment death penalty" is:

    Also, this is the opposite from a clear cut. Means you are not free to experiement ever, because people will see the result of this experiment for forever on the char.

    There is nothing to say that experiment = death, unless we are operating under the assumption that death comes quickly to people who aren't playing at 100% capacity. You can experiment in WoW for example (new rotation, specs, etc), and the chance of dying is pretty low. Indeed, WoW could make it so you lost everything you carried when you died, and while everyone's blood pressure would be high, the odds of casually dying is remote.

    The "simulation" argument is pretty weak, because we are already assuming a fantasy setting with fireballs and castles built on labyrinths full of monsters (who does that, really?), and infinite resurrection. As the old joke goes, we have already established what you are madam, we are simply haggling over the price.

    So, again, what is the goal of a harsh (or lenient) death penalty? Having one for the sake of having one is not particularly compelling game design. The penalty should fit the mood of the game, but also how you want players to play. EVE, for example, has a "realistic" death penalty, but it also serves another purpose: to prevent veteran players from rampaging across the galaxy with their capital ships and wiping out everyone. They would likely win every encounter, but there is always the outside chance they would run into someone else's capital ship, or a large collection of newbies who could potentially gang up and wipe them out. That threat (presumably) keeps capital ships outside trivial battles, while also forcing players to pool their resources since it's difficult for just one player to shoulder such a risk by themselves.

    You implicitly agreed with me that harsh death penalties reduce exploration/experimentation, which are two of the things you LEAST want to discourage in a game. So... why advocate for harsh penalties? My "embarrassment penalty" wouldn't really affect new players, or even many veteran players who have decided not to strive for the leaderboards (just like Arena rating isn't a deterrent from casual players wanting easy Conquest). And everyone gets to keep their stuff. Win-win, right?

  34. Azuriel, I won't reply to your entire comment. If I did I could just as well make a new post ;)

    I thank you for the time you invest to discuss this, but to use this

    The "simulation" argument is pretty weak, because we are already assuming a fantasy setting with fireballs and castles built on labyrinths full of monsters (who does that, really?), and infinite resurrection.

    argument on my blog is really .. brave, you know. Have you even noted the subtitle of the blog? ;)

    I'll give you the answer I give to everybody who starts to argue the way you do there:

    I am talking about simulation, not about realism. No fantasy game is ever realistic, but it can be good simulation. In fact, it should be.
    If the simulation were not important, then you could just as replace the character models by squares and the NPCs by circles.

    I want a credible and consistent simulation of a fantasy world. Not a copy/paste of (1) real life and not an (2) arcade game, but a (3) fantasy world simulation.

  35. @Nils: Point taken on boxes sold. But boxes sold "by now" means its quality over time has continued to attract the players would otherwise not try a difficult game. Its an excellent RPG through and through and it's "multiplayer" ghosting system should be stolen throughout the industry.

    Also ...hate? there's no hate! :) Im sorry if I said something that wasn't clear. But fear not, there was no hate or anything like that. I just think death penalties which don't encourage game play, but which over time discourage gameplay, don't work.

  36. I just think death penalties which don't encourage game play, but which over time discourage gameplay, don't work.

    That sounds like a very logical statement. The question, of course, is how death penalty can encourage play. ;)

    I think a death penalty should make a game fun. And a MMOROG is not only an abstract game, but also a simulation (I'm not just a guy with gameplay abilities but I'm also a guy in a fantasy world).

    When the game's rules break the rules of the simulation, that harms the fun. And fun is how a death penalty can encourage play ;)

  37. @Nils: I agree with you on that last point, unconditionally.

    In an MMO, I think older games addressed penalties for death best, tbh. But I don't think newer games will every turn back. Designers dont even remember why we had it in the first place and seem to arbitrarily toss it out as something grassroots gamers merely tolerated, not enjoyed. I think many of the things mentioned in the comments and article are some great ideas.