Tuesday, December 27, 2011

Game-Mastering Dilemma: Sinking the Player's Ship

Sea ice off the coast of Svalbard
Holiday break is over, and it's back to work.  Hope everyone had a nice break and is charged up for another year as we countdown to 2012.  Let's kick things off with a game mastering dilemma.  I'd love to hear how different DM's would approach the issue outlined below.  I don't know whether folks still think in terms of GNS* and games theory (see notes below), but my first thought is this can also be analyzed as a GNS problem - your priorities would inform your own solution.

The Problem
I've been working on a campaign setting on the blog on and off for a while, called The Black City.  It's a ruined alien city on the shore of a frozen northern island; I'm calling the island Thule, but the real world inspiration is Spitsbergen island (part of the Svalbard archipelago in the Arctic circle north of Norway).  Viking adventurers have discovered the ruined city and the lucrative treasures found in the icy ruins.

Ocean travel during the Dark Ages was dangerous, and ships were frequently lost.  For example, the records of the Viking expedition from Iceland to Greenland report that nearly 50% of the initial colonization fleet didn't make it (either turned back or were lost at sea).  The trip from Lappland to Svalbard is equally as long - about two weeks across open ocean, with the only possible stop at Bear Island in the Barents Sea.

During summer, June-August, the ocean north of Lappland should be passable and navigable with little or no sea ice.  The Atlantic current keeps the southwestern Barents Sea warmer than comparable latitudes in the Arctic.  The confluence of Atlantic currents and cold Arctic waters does create strong currents and frequent mists and fogs, complicating sailing conditions even when the sea ice is gone.  There's only a few months of favorable weather for exploration of the ruins each summer. 

We can calculate chances that ships will sink while making the 2-3 week journey, and I'd hazard to guess it's not a trivial percentage.  Sinking is a death sentence in those cold northern waters for everyone on board.  We could quickly identify factors that might cause a ship to sink  - foundering in high waves, getting caught in a squall or storm, crashing into rocks or sea ice while sailing blind in heavy mist and fog.  One could identify factors that would reduce the chances of the same; the experience of the captain and crew, the load and cargo, the presence of beneficial magic, and how early in the season the ship sails (there would be more ice in May, like the picture).

Here's the dilemma for old school dungeon masters:  When the adventure requires a hazardous journey just to arrive at the adventuring site, should the characters have a chance at failing to make it?  Killing the entire party, by sinking their ship on the way to the dungeon, would be identified as a "negative play experience" by modern designers.  Aren't the player characters special unique snowflakes that should be granted sufficient 'plot immunity' to at least make it to the dungeon?

Let me hear how you'd handle this type of ocean travel in your game.  My thought is most DM's would give player ships a chance for sinking or blowing tragically off course based on whatever weather charts they're using, but would then let the players get shipwrecked, washed ashore, or clinging to wreckage, rather than instantly killing everyone.  But washed ashore or shipwrecked is not an option in the arctic waters.  So what would you do?

I had my own solution written as part of the post, but its better to hold that back a few days and first hear how other folks would handle the situation without establishing any precedents.  I've also got a new poll up on the right to get the discussion started.  Let's hear some reasons why we should or shouldn't sink the player's ship in the comments.

*GNS Theory
GNS theory was used on The Forge to analyze approaches in games based on whether the rules were trying to advance a particular agenda - gamist, narrativist, or simulationist.  A simulationist approach might declare that ships will always have a chance to sink and the game rules will try to model a world with internal consistency; the gamist approach might consider it a mere resource problem, and by applying the right choices to the percentages, the players can "win"; a narrativist approach might let the players handle the description of their own journey (along with precautions) and just resume the story at the island.

**Sea Ice pictures
Atmospheric and glacier research is done at Svalbard, along with limited mining, and lots of folks seem to post interesting blogs while they're up there, full of excellent pictures; these sea ice pictures came from this blog (link) and show sea ice conditions in May.


  1. The phrase "The location for the adventure lies here" is not very old school.

    If they have to ride the train to the dungeon, then unless the adventure is about surviving a train wreck, then don't wreck the train.

    However if there is countless wealth over the seas and the players choose to go there over other options, then yes, the risk of death should exist.

  2. A chance for them to sink at sea and die is basically this sequence:

    1) Make up guys.
    2) Roll to see if they die on the way to the dungeon. If so, go back to 1.
    3) Play out the dungeon.

    Repeat, if anyone wants to.

    Saying they make it automatically is basically this sequence:

    1) Roll to see if they die on the way to the dungeon. If so, repeat this step until an expedition makes it.
    2) Roll up the guys for the expedition that makes it.
    3) Play out the dungeon.

    That's just for the first expedition. In that case, it's better to go with #2 regardless of your bent in play style. Why make up crusaders and see if they die on the way to the crusades game you want to play? Why not just play the guys who made it? I mean, how much free time do you have?

    Traveller gave you a chance to die in chargen, but that was a way of saying "Do you want to push your luck or stick with this guy as is?" IMO. It wasn't "roll to see if the party lives to play tonight." Which would be lame.

    I'd require a roll, personally, but make it a "does bad stuff happen?" roll. Maybe a damaged ship (can't sail home without fixing it, repairs need stuff in the Black City), maybe lost food (cook up the apes in the BC), maybe NPCs were washed overboard, maybe people got sick, etc.

    Seriously, why roll to see if the game is over before it starts? How is that fun? Rolling for calamities is fun, rolling for party death isn't. IMO.

  3. -C: it depends whether it's the players deciding to go or the DM 'making them'. For purposes of removing that factor from the discussion, if the first game starts with the players choosing to hire on as marines on a longship, to make it out to this ruined city for some looting, you're still back to the same place: Would the DM sink the ship along the way?

    Peter - I was also thinking about 1970's Traveler as I was writing my own approach (and how this situation differs).

  4. Well, isn't that the primary difference in the new school and the old school?

    If it is what the players are choosing to do in a sandbox, then of course you should sink the ship. If you've got a sequence of precious encounters that you are forcing them to go on, then don't.

    I would think that would be the metric.

  5. No, I wouldn't sink the ship and kill everyone.

    a. You could outline the chances (as a percentage, or roll of 1 on 1d6, or whatnot) of players getting there successfully.

    b. You could start the campaign during the safer ice-free summer and not have ships lost at sea then

    c. You could sink the ship by having it crash into an iceberg riddled with the tunnels of the hideous Ocean-Going Yeti rather than lose all hands

    You put all this work into your megadungeon right? So making it impossible to get to makes no sense. Any player with enough information to evaluate the risk is going to decline.

  6. didn't think that last comment fully through. Option "a" is only good if you want to discourage the players from going, because they're really not likely to volunteer for a TPK sea voyage with no way to alter the outcome.

  7. I think it depends on when the adventure/campaign starts. In a "pure" sandbox, pick a start location, and then if the players choose to sail the oceans they should be notified of the risks ahead; if they choose to take the risk, then roll for sinking.

    On the other hand, if your adventure/campaign is focused on the Black City itself and is not a sandbox of the Far North per se, then just quickly narrate the journey and start the adventuring careers (and the risks) at the trading post assuming that the PCs got there safely before level 1.

  8. I think it all depends in where the focus of the game is. If it is all about the city, then start with the PCs arriving in port and go from there. If it is a place that characters can travel to, from another location, then allow for some chance of sinking; but even so, I'd shipwreck or otherwise penalise the characters, rather than kill them off simply because of a ship travelling at sea.

  9. I'm on board (rimshot) with the others who say that if the game is the Black City, then let the players get there safely. If the game is a world that includes the Black City, then the trip there can be hazardous. At least if they're passengers.

    If the players are the ship's crew, then if you want to send them into a storm and let them fight the elements, fine. That's another matter.

  10. My rule of thumb is always Do What is Fun.

    Is having the characters die in a ship-wreck before they get to the actual adventure fun? In my opinion, no. So I wouldn't do that.

  11. For purposes of removing that factor from the discussion, if the first game starts with the players choosing to hire on as marines on a longship, to make it out to this ruined city for some looting, you're still back to the same place: Would the DM sink the ship along the way?

    Just talking in terms of the first game, why does the game start on the coast ("with the players hiring on as marines") instead of on the island of the black city. Both are equally arbitrary as starting points. I have always been puzzled by the investment so many DMs seem to have, given the presumption the PCs can go wherever, in starting the players off in a boring village instead of a place with a lot of potential for an adventure. If they want to take a perilous sea voyage during the game, sure, have the risk there - and if you start off the game in the black city don't be surprised if no one wants to risk the sea voyage to go to the boring village.

  12. Let them start at the trading post near the Black City, otherwise what is the point?

    I do like Pat's suggestion above about the floating iceberg with tunnels inhabited by Yeti. That would make a good side-trek if the ship crashed into it.

  13. Let the boat sink if that's what the dice say, but don't kill them off as a result. Let them decide what to do, because "we're stuck in the arctic without transport" sounds like one icy heck of an adventure.

  14. If it's an informed choice, the players know that there is such a high chance of total death, and the voyage is part of the at-table play then yes I think it's fine (though likely to be either something very funny or a campaign killer if it happens).

    If you make an informed choice I would think about giving the players some in-setting options for lessening the chance through their own actions (of course with some risk of its own). Maybe their is a magical horn on a nearby mountain or a certain rune that is only known by a skald in such and such place.

    If it's just background to where the sandbox really begins (all campaigns have some kind of assumed frame) then hell no. Presumably you wouldn't have players roll for their chances to die as infants or from contagious disease or starvation because there is historical precedent

  15. I wouldn't roll a die to instantly kill PCs in a shipwreck any more than I'd roll one die to instantly kill them in a battle.

    In your particular case, if the campaign is the dungeon, I'd start them arriving on Thule. I might roll for weather dangers on the way back but it'd either be a lot of rolls to get to "everybody dies" or there would be an 'out', such as an uncharted island. Given the tenor of your campaign I'd likely go with a "Boats of the Glen Carrig" approach and have the shipwreck be a jumping off point to more battles with horrid monsters.

  16. Irrespective of the particulars of the situation (i.e., ship sinking), If the players choose to have their PCs take a risk, and then the risky situation goes bad, then their PCs may die. Full stop.

    That said, we're talking about a fantasy world, so the PCs have ways to reduce/eliminate the potential for problems. This could include pre-journey divinations and sacrifices to sea gods, for example.

    And if they do run into problems, they probably have some ways (spells, potions, magic items) to reduce the danger of a shipwreck. Escape abilities (flight) are one obvious way, as are weather manipulation abilities. Even precognitive abilities can buy time or grant opportunities to change course.

    Finally, there's no reason such a disaster needs to be instantly fatal. In the swords & sorcery genre, a successful saving throw can be liberally interpreted: A PC ends up clinging to a large chunk of floating debris that somehow manages to avoid the worst of the crashing storm. Plus it's reasonable to treat this kind of hazard as (periodic?) damage to the PC's hit points; if the storm ends and they're not out of hit points, then they survived.

  17. I agree with most of the previous commenters. If the megadungeon is the focus of the game, I wouldn't include a TPK in the rolls in normal circumstances. However, rolling for weather conditions and danger brings up some other options:

    Since the 'safe sailing season' is quite short, and thus the time the PCs have to make money in the dungeon is finite, all sorts of options open up. Are they willing to risk sailing home later, when the seas are rougher, if they can do a few more delves and come home with better treasure? That new magic sword won't do shit for you if the ship goes down with all hands. And this is down purely to PC greed and risk-taking, the equivalent to staying in the dungeon when you're low on HP, spells and healing and should really pull out to rest.

    Similarly, you might come up with some "minor weather mishaps" rolls at the beginning of an adventuring season. If the PCs can't lay out enough $$$ for a good captain, navigator, crew etc, then maybe their ship gets damaged and they need to make that money back and repair it before they can go home at the end of the season. These are some possibilities for fucking over your players in a fun way which involves decision-making and promotes further adventuring, instead of just rolling to create new characters.

    Now this may involve suspending realism a bit(?) if the high seas are really an all-or-nothing proposition. But I wouldn't let that get in the way.

  18. I have to chime in with other posters: A shipwreck with fatal consequences is not an 0/1 event. Even if a longboat hits an iceberg and is damaged it won't go under in a second. Let the PCs bail for their lives, repair the ship with spells etc.. And then they can still die...

    Clinging on wreckage in artic waters would'nt work, though, longer than a few rounds (depends which length your rounds have - 1 minute AD&D rounds? Maybe one...)

    Instead of having a %-chance of death, I would use an encounter table with hazards, that can be acted against (or not) by influencing the course, the captain, the speed, etc.. and a weather table. Storms and fogs could prevent the ship from avoiding that easily spotted iceberg in sunny weather...

  19. There's something disturbing about air travel spam showing up in a thread about total party kills during voyages.

  20. That's pretty funny stuff, Tom. I was going to delete the spam, but in light of your comment, I'll leave it up for the irony.

    Air crashes are no laughing matter, but I'm reminded of the expression, "When your time is up, your time is up - but what you need to be worried about is whether you're flying on the day the pilot's time is up".

    When I develop systems for making it safely to the island, I'll include some element of rune casting and divination: fate and wyrd is such a part of Norse folklore.

    Don't sail with a captain whose "time is up".

  21. My first question would be, why are you making the setting work like the real world in the dark ages? Isn't that a simulationist assumption to begin with? That is not required in order to put PCs and NPCs on an equal footing. The trek could still be dangerous; just make the dangers pirates and sea monsters (these are challenges that will feel much more fair to players).

    A few other ideas:

    - Mythic Reality: have a "shipwreck" be represented by a fight with sea demons (maybe that's what all shipwrecks really are).

    - Go the "Eyes of the Overworld" route and have a wizard send them there via demon courier (or whatever). Leiber uses similar devices in the Lankhmar stories sometimes to place the protagonists in a situation. (Perhaps The Black City is too low-magic for this; I don't remember the setting details right now.)

    - It's supposed to be a game of player skill, right? One could make a seagoing minigame similar to the desert in King's Quest V. This could even be a "dungeon" in the sense of a graph with encounter area nodes and edges leading between areas. Maybe the PCs have found a "treasure map" which will lead them through (using standard hexcrawl rules for getting lost). Or maybe they can get clues about which routes to take from the one-armed man in the tavern.

    - In Fourth Edition, this would probably be handled as a skill challenge (even if you're not using skills, you could use a similar mechanic based on ability scores, saves, combat, or something else). I find skill challenges to be a bit too abstract and unimmersive (new word!) for my tastes, but it might still be interesting to revisit them, as this is the kind of problem they were designed to solve.

    Personally, if the point of the game is the dungeon, I wouldn't do any of those. As Pat wrote: You put all this work into your megadungeon right? So making it impossible to get to makes no sense.

    But if the dungeon is only one of several locations of equal interest (say, at least 5) and you are using a "just in time" prep approach, then I would seriously consider one of those other challenge systems. I think the "mythic reality" one would be the most memorable.

    I also really like HDA's idea of a "safe sailing season", as that brings in time management issues.

    I would in no circumstances ever just make a die roll for the ship sinking, even if the probabilities had been communicated to the players.

  22. When the adventure requires a hazardous journey just to arrive at the adventuring site, should the characters have a chance at failing to make it?

    Sinking the ship doesn't end the adventure; it just changes it from a 'sailing exploration' adventure into an 'arctic survival' adv.

    The adventure happens where it wants to; not where the players want it to. Just ask Ernest Shackleton. ;)

  23. A TPK on a single die roll, a roll over which the players have no influence? No, thank you. I wouldn't even to that to players in Paranoia.

    Well, maybe in Paranoia.

    Partly because it's such a frustrating experience. Even without any particular emotional investment in the PCs, we've still spent time rolling dice, buying gear, preparing spells, and so on. And, if we're any good at this, we've spent some time maximizing our effectiveness as a team against the dangers we expect.

    Dice clatter. "Sorry, guys, your ship sank. Start over."

    Here's the dull part: Creating a new party in this situation is essentially the same as the last time we did it. The dead guys didn't come back to tell us what killed them (probably), so we have no new info. Chances are, we're pretty much re-creating the same party (insofar as dice rolls allow), so we're going over the same ground.

    Again and again. Until the dice are kind to us.

    So, no, no TPK roll. Now, an unexpected encounter with sea monsters or pirates? Great! Even a wandering damage table that tells us a certain number of characters didn't make it, or a percentage of our supplies were ruined, would be better than a TPK.

    Or, hell, just start us out at Trade Town. I'll forgive a little bit of rail, in this case.

  24. I know I'm late to the party, but this is too good a discussion to just pass by.

    How would the ship sink? Why? What happened? This is the difference between an arbitrary game-stopper and a game. The ship sinking, all hands, doesn't need to be the end, as so many have already said, but you could certainly use the ship sinking to still let them achieve their goal--getting to your megadungeon. The mechanism for that travel doesn't disappear, it changes. It may make for an extra play session before they delve Inside, but as long as you are introducing a complication instead of an arbitrary TPK, they shouldn't mind. Especially if their characters are already on the sheet.

    Now, off to read your follow-up post. ;)

  25. Old post, not sure if I get the central ideia (English not my native language) but here's my five cents about the subject.

    RPG is like a jazz song. There's a phrase, and there's improvisitation. Let the player dictate the rithym of the game.

    While in the ship ask them what they want to do... If things became funny, let the ship sink and dare them to get out of this alive. Improvisate...

    By the other hand, dont't hesitate to do a "Deus Ex Machina" to rescue them. The funniest part of a RPG is that is fantasy. A Pirate ship is up to rescue them, why not?

    And don't forget the Chekov's Gun. Maybe one of the pirates has the key to open a secret room to that one dungeon... or has a map... or whatever.

    Be smart. ;)