Mechanic for finding someone?

So in D&D, one player made a bounty hunter character and I gave him
the list of thieves and wanted persons that I had already placed in
prep. Now, in between sessions, I made a couple of more colorful
criminals, their crimes, appearance, stats and motivations.. I wonder
if you have any tips for mechanics to determine 1. Where they are, and
2. How he can find them? Get clues etc?

For the starting city, but not for the wilderness outside (which is
also relevant, he is Druid/Outlander), we have a detailed map (down to
the house/canal/boat level).

He can talk to the dinosaurs via a spell so that's how he's found the
ones he's already met. Including another PC (PvP for the win), but I
don't want to use only that trope and I also want something
more general because I can imagine this will come up in a
city/wilderness milieu.

It's not so much a problem in the dungeon, but it is on the city map
and hex map.

I almost started making a AW-like move "When you wish to find someone,
roll X..." but "wishing" something isn't leading from the fiction,
that's just intent.


  • What's your goal with this mechanic?

    After all, there is a world of difference between "How do I make finding people into an exciting adventure" and "How can I handle this as quickly as possible, so we can move on to other things?"

    If you make an AW-style move, yeah, my preference is to use a clear fiction-based trigger. e.g. "When you spend money in the city markets and make it known you're looking for someone..." ("take -1 if want to try to do discreetly", perhaps).
  • I'm cool with it being an exciting adventure that takes up time. It's not just a background mechanic.

    But... what I concretely want is to for the mechanics (as opposed to my whim, which is how we're doing it now) out where the person is, geographically, and how the PC could find the person.

    Then we could do it quickly, or "Oh, he's way over there... then never mind, I won't pursue him", or do it as a detailed exciting adventure, or whatever we felt like in the moment.
  • The way I'd do that would be to have the character spend time doing information gathering in places that are likely for the given target to have frequented. I'd model the information gathering in the abstract as suitable skill/ability checks, depending on the mechanical framework of the campaign - could be search checks in 0th ed. D&D, or Ability checks in Mentzer-like, or skill checks if there's something akin to 3rd edition skill system in place. Difficulty, expense and time spent would all depend on where they're searching, with what methods, and how much of a mark the target left in that area.

    On a failed check I'd describe how they learn stuff about the area they're canvassing, probably throw in some colorful detail, side hooks, whatever comes to mind. On a successful check they get those as well as an information source, which will then provide them with the kind of information likely for the connection the target has with the area being searched. If the system has degrees of success, a low degree of success would gain likely kinds of information, while a high degree would indicate a stroke of luck - the target has made a mistake, leaving a clue behind, or somebody wants to betray them, or whatever. If the system I'm using doesn't evaluate success scope, I'll use a random table to figure out the nature of the clue a success in information gathering gains.

    Note that if the PC is originally searching in an area that actually does not have any witnesses or clues or whatever about the target, the only thing they get with success is "you're pretty sure there isn't anything to be found about them here". Of course, a smart villain just might have been extra careful about hiding their tracks, so it's up to the player to decide whether their success was good enough (or they've garnered enough successes, if not evaluating degrees of success) for them to really believe that they're barking up the wrong tree.

    Basically, the above is exactly the same answer I'd give when adventurers are looking for secret doors in a dungeon. Note how the process has the nature of a dialogue, as the player and GM discuss the strategy and tactics of information gathering, which influence the expense, time and difficulty of making successful information gathering checks. Meanwhile the GM has the secret layout: what are the habits and personal history of the perp, where they have their friends, where they go shopping, and so on, that they can use to interpret what the success in information gathering means. So we have two facets to the issue that interact to determine whether it is possible at all to find someone: the effectiveness of the information gathering, and the effectiveness of the information security of the target.

    For example, if the PC was searching for just some guy who'd been outlawed for a passionate murder, and who was now living "in the woods somewhere", the PC would logically speaking need to find out the following bits of information, roughly in order:
    1) Where they used to live and work and love before being outlawed, who knew them, that sort of stuff.
    2) Which area of wilderness they're hiding in.
    3) Where, exactly, is the hideout, or the person in question.
    That first bit should be simple, but you never know with players - I've seen people go on red herring hunts for several sessions at a time because it never occurred to them to you know ask the bailiff for such basic information. The second bit can be found out by haranguing their friends and family, simply on the premise that if the guy is acting like a normal outlaw, they will meet with their parents and spouses and whatnot on occasion, to get some food and new clothes if for no other reason. (In case you had an image of some sort of survivalist commando living in absolute seclusion when I said "outlaw", do discard that - it's not realistic either historically or today, that's just not how people act.) The third bit requires either staking out a regular contact to catch the perp when they sneak into the village, or getting somebody in the know to betray them and lead you there (unlikely - surely they haven't shown their specific hiding place to anybody except the most trusted), or doing some field work with various sorts of ranger skills to outright map the wilderness area and maybe track them to their hideout.

    The above sort of process would come about not because the GM has prepared it as a mechanical framework, but rather as a dynamic outcome of the dialogue between the player and the GM: the player tells you what they want to find out, you brainstorm some ideas about how they might go about finding it out, do some rolls, get some results, and then repeat until the perp is found. In some cases it'll be just one roll and do the initiative, in others it'll be lots of dead ends, just like in real life bounty hunting.

    I dislike the AW-style move for this style of game, obviously, for the reason Sandra explains: it determines whether you get information not based on the fictional situation, but on the dice roll itself. This can of course be fixed by changing the outcomes of the roll to indicate not success in finding someone, but merely success in finding clues, or ascertaining that there aren't any, in which case we're back to the above suggestion.

    And of course, as Paul notes, if you don't actually want to make a game out of it, then I just recommend a single "bounty hunting check" on whatever mechanism, and award reward money (no xp of course) per month of bounty hunting work as the PC works as a bounty hunter during downtime. Maybe award an injury, or even death, if they fail the monthly check - I'm cool with getting fucked up during downtime activities if your downtime is particularly dangerous in nature. Maybe assume 1-3 caught perps per month, or one per degree of success, if that matters. Not that I would be very likely to skip this sort of adventuring premise if it fell to my lap - doesn't get much more adventurous than bounty-hunting insofar as "real life" occupations go, frankly. Sort of like a player told me that their PC wants to go looking for the Noah's ark in their roal as a mystery archeologist, and I'd just roll some dice and let that be it.
  • Ah, I noticed that you also asked about prep: as you can see, my method would rely on the GM actually knowing the particulars of who, what, where the perps are, so of course you'd need to prep that, or be good at improvising.

    I would suggest writing a short life history, line by line, for each perp, indicating where and with whom and doing what they've been. Just a half dozen details about each. This should give you everything you need later when the player tells you that they'll start asking around in the neighbourhood where the crime happened: you already know that the guy was actually living in the wharfs of the city for the last three years, despite the crime happening at the market, which would justify hiking up the difficulty in information gathering a bit until the PC figured out (from the mission background info or by meeting somebody who'd met the guy while looking at the market, whatever) that the guy was a sailor and therefore the temporary housing and slums around the wharfs might be a better place to be looking.

    Another thing you might want to write down would be their motivations and personality, because those would impact the trail they leave: if the crime was one of stupidity and the person wasn't a career criminal, they would be likelier to leave all sorts of clues and witnesses, while the opposite would be true for e.g. an ex-bounty hunter who already knows how to think about tracking, and what to look out for - they would be using false identities, being generally tight-lipped, not seek company while on the road, not tell anybody where they're going, and so on, all to increase the difficulty of trailing them.

    The data you basically need is whatever it is that you as a GM need later on, when you need to gauge things such as "did this guy come to this bar often enough over a long enough period of time for anybody currently present to recognize him from a drawn sketch". For me this would be the aforementioned: a rough life history, plus an understanding of their nature and motivations. Presumably a good bounty hunter would attempt to basically do a writeup of the same exact things to predict and understand their quarry, so you get a process where the player is slowly compiling the same exact facts the GM already has.

    One more thing: to me as a GM it would be obvious that I'd put all this together in terms of "enhanced realism", meaning that the difficulty of the bounty hunting cases would be all over the map, with some cases trivially easy and others essentially impossible, leaving the players to decide for themselves how much effort they want to put into each one. If you're operating under a different paradigm, be sure to vet your bounty cases for difficulty, so they're as challenging as you need them to be for meaningful play.
  • Burning Wheel's Circles was made for this. Now, it is a skip-to-the-exciting-bits technique, though the exciting bits are tied up in finding the person and what happens when you do. Also, as written it may be a bit just-in-time/no-myth for your sensibilities, though it doesn't strictly need to be.

    Looking at it more broadly, what level do you want this mechanic to operate on? I can see several options:

    - Just figure out where the character is, and drop hints to the player. You as the GM know the answer, and you collaborate with the player to figure it out from the clues: "Oh, you're looking for so-and-so? She usually hangs out near the docks, but I heard that she goes to the market on Tuesdays."

    - Skip the details of the character's routines and focus on how the player finds out, the process of the search. "So, how do you plan to find her? Asking people at the docks if they've seen her? Sounds reasonable that someone might have, roll your Charisma to see if they tell you enough information to find her."

    - Focus on the result of the search. Based on the player's intent and the character's actions, describe the result. "So you're chatting with random people at the docks, seeing who saw her last? She's here, but she's upset that you've been so loud in looking for her--she's afraid the city guard might have overheard you."

    There are probably other options. You might also want to consider if they're trying to track down a specific person ("Natasha, the smuggler we met last week"), or just a general kind of person ("Any smuggler who knows their way past the city watch").

    Even more broadly, finding someone is basically a mini-mystery. I'm not sure if any mystery-solving techniques would work (Gumshoe? I still need to play it) but that might be another source of ideas.
  • I guess I should've named the thread "Mechanic for placing someone" because while I made the list of their crimes, appearance, motivation, monetary reward and inner soul struggles based pretty much from a mix of random tables and whim, I had no idea where they would be at, physically or socially except by resorting solely on prep-time (or, worse, on-the-fly) whim.

    Eero, your method is very "naturalistic" and I do like it. That's gonna be my fallback if I don't find something simpler. I'll just have to print another copy of the map to mark their locations on, and also make like a "chance that the person you're asking knows the place"-mechanic.

    Now that I think of it, Polychrome presents a sort of semi-mythful technique where it goes like this:

    1. The players ask for any fact they need to get what they want. Anything
    2. The fact itself the DM is given license to make up on the spot (!!! I do not like that). In this case, where the person currently is.
    3. But the way that they find out the facts is highly proceduralized and partly based on random tables.

    Kinda the opposite of the "Zendo"-like inductive/abductive prep of, like Eero suggests, hard-determining the facts and then improvising the clues.
  • Wouldn't most locations of bountyhunted individuals be readily deductible for the GM by the virtue of knowing everything about the crime and the criminal? I was sort of assuming above that once we know the who, why and what, we could easily jump to some conclusion of natural intuition. "Well of course they would hide out at their brother's hunting cabin." "Well of course the thieves' guild will put them into one of their safe houses." "They have a ship, so that's where I would be heading first of all, and it's either France or Denmark after that, naturally."

    Thinking about it, this obviously depends on what role, if any, you accord to "natural intuition" in your GMing. For myself, I recognize that no prep can cover all cases, but I don't like value-tainted GM decisions either (those would be the ones where the GM chooses one way or another because of the meaning the choice will have on the game), which is why I train myself to provide "neutral" decisions off the cuff through natural intuition: whatever it is that I see in my imagination to be the case is what the players get, no censorship. This is actually a pretty crucial part of value-neutral refereeing, it seems to me, given the amount of choosing a GM has to do in real-time in D&D.

    Natural intuition seems to me like it's the right tool for the job in this case, as you already have so much data on the NPC, so it's easy to let the imagination run on what they would be doing next. If I suspected my neutrality for some reason (particularly high stakes or a player suggests the exact thing I was going for a second before I make the conscious choice or whatever), I'd just dice for it to introduce a bit of randomness. For example, my go-to randomization with natural intuition is to roll 50/50 on the idea I have, and discard it at the say-so of the dice, forcing me to come up with another idea; this ensures that while I will choose a naturally-occurring idea most of the time, there are no predictive guarantees about it.

    I guess that if I really had no clue whatsoever about where the perp could be hiding, I'd just take a map of the area, randomize a location and then invent (by natural intuition) the reason and story of how they came to be wherever they happen to be. That last bit is important, as unless they're a stochastically trained wizard, they're not likely to simply teleport to some random spot to hide; we need to know what attracted them there and how they traveled and from where, because otherwise we cannot know what to tell a player who starts interviewing wagoners and road wardens about the fugitive on this or that roadway.
  • In D&D, you don't need a mechanic for placing an objective (even a character), that's your realm as a DM. You place the character for whatever reason you like, wherever you like: for thematic reasons, they're holed up with their girlfriend because in this adventurous fantasy world love is everyone's motivation for doing everything. Or, for challenge reasons, they're holed up in a tower full of traps and trickery because that's a cool challenge the players will like. Or, for simulation reasons, they could travel X miles with their Y supplies before running into Z opposition and having to obtain shelter at map coordinates A, B. Whatever your aesthetic is for laying out the world - and you get to decide what that is because D&D is a game that benefits greatly from the freedom given to the creative director known as the DM - apply it to this.
  • That's right. But some things I want to decide myself and some things I want to leave to mechanics.
  • edited July 2015
    If you don't feel comfortable choosing between the options JD suggests, you could always roll between them.

    Roll d6 for target's location and status:
    1. Holed up with romantic partner in affluent city
    2. Manning defensible tower on border of hostile factions
    3. Prisoner of evil wizard in their trap-filled fortress in no-man's land
    4. Sheltering from the snows somewhere in the mountain passes
    5. Next door drinking in the inn
    6. Having recently achieved political office in a corrupt city

    Then you just look around your map for suitable affluent cities, corrupt cities, contested borders, mountains and no-man's lands.
  • That's pretty good
  • Or more generic:

    1. Gone to ground in a very undesirable place (e.g. drunk, passed somewhere in the slums)
    2. In an adventure location (e.g. dungeon) [Roll again: 1-3, by choice; 4-6, against their will]
    3. Very surprisingly nearby. [Roll again: 1-2, a bit of travel will suffice; 3-4, near and accessible; 5-6, far closer to you than you might have liked! (e.g. staying with the person you're currently talking to, standing right behind you, one of your hirelings, etc.)]

    As a GM, you can use your creative abilities to find a suitable location - if you roll a 2, and there are three dungeons nearby, you can pick the most suitable (maybe this NPC loves gems, and one of the dungeons is rumoured to hold a gemstone hoard, so, duh, there she goes) or roll a die to pick one.
  • It would then be very D&D-esque to roll to see if you find out pretty much where they are (full success), or just some hints, possibly misleading (partial success) - particularly if the person didn't want to be found.
  • That's right. But some things I want to decide myself and some things I want to leave to mechanics.
    Well, you are the mechanic in D&D! The players, in engaging with the world of D&D, aren't engaging with just random tables and percentiles, they're engaging with you and your vision for what a cool adventurous fantasy world is.

  • edited July 2015
    If you're open to useen randomness in a free associating words kinda way and not opposed to interpreting vague results, you can use some "idea generators" like the RoRy story cubes or even the action/subject word combination of systems like Mythic GME.

    Using the event generator at this gme implementation (, I rolled Waste/Adversities as a combination. From that, I could determine that the fugitive could be in dire straits at a desert or artic waste, depending on what's in the could roll more combinations to flesh that out as needed.

    I find the story cubes to be very suited for little sequences of events (or bare plots) as you arrange the results in a chronological line. In combination with the above, they could inspire a nanotechnology (stupid auto-correct) nano-story indicating how the fugitive ended up where s/he is.

    There's also another nice tool for this named BOLD. I'll find a link for it. I'm pretty sure it's PWYW.
  • edited July 2015
    Jason, don't worry.
    The world is plenty cool. Last night was the best session of my life I think.

    One player is a dinosaur hunter and former criminal. The other is a dinosaur loving druid and current bounty hunter.
    First they were searching for information about the criminal gang in a city on the waterfront, jumping boats and climbing ropes. The dinosaur hunter secretly helped the criminals flee from the bounty hunter.

    They found a tree that unlike the normal trees next to it didn't flutter in the wind, but the sun and shadows played on the trunk extremely fast. When they shot arrows into the tree red rings floated along the surface down to the ground where they dissappeared. The bounty-hunting druid turned himself into a velociraptor and attacked the tree but the tree struck him and they retreated.

    Then they went to an opening to the glitch where there were normal trees outside. In the treetop were four thin clawed beings that shot arrows at them. The velociraptor climbed the tree while the dinosaur hunter hid behind another tree to shoot them. They hurt the velociraptor with their claws so he had to cast healing word on himself.

    They went into the glitch where they found a blinking green box with the symbols "TIXƎ" in white text over a black-and-white checkered tile floor.
    The next room was a bath room where a grey completely woolly, hairy man who splashed water on himself to turn into a giant monster. The monster was immune to the dinosaur hunter's arrows but died from the bounty hunter's thorns, but not before he got a bite in that infected the bounty hunter with the glitch. But he had a spell to save himself.

    I like writing dungeons (and even there I use some die rolling to get a good mix of empty rooms, full rooms, treasure etc).
    But I do not yet know how to manage a list of wanted criminals and their hideouts and the clues to those hideouts well.
    I'll keep thinking about it, this thread has had good suggestions.
  • Dreamer, oh, I almost missed your post. I have some story cubes so maybe I'll bring those out. I have five sets. The dinosaur themed one will be good for this campaign.
  • edited July 2015
    I know you like Fudge dice. One simple oracle I've used in The Bureau (my multiple-die-sizes AW hack) is to roll 2dF:

    A '0' means that you add a detail which is pretty much what you'd expect.
    A '+' means you say, "The good news is..."
    A '-' means you say, "The bad news is..."

    This gives a nice spread.

    Each die leads you to say something (don't add them together or anything).


    "We're looking for Mr. Gladly. We've heard he comes through town on drinking binges sometimes."
    (roll: a '0' and a '-'.)
    "Looks like that's exactly where he is: at the tavern. (0) The bad news is, he's mad drunk and he's getting arrested at the moment. (-)"
  • Oh, that's great!
  • Jeebus I just realized that auto correct mangled some of my words. Nanotechnology should read as nano story. ☺
  • Also, don't overlook Eero's 50/50 technique (it got buried in his long text there somewhere, but it's REALLY good). It's a very effective way to do this kind of thing:

    You come up with the first thing that comes to mind - your natural intuition on the subject, as Eero puts it. (I think of the improv injunction, "Be obvious!" Whatever comes to mind first, that's what you use.)

    Now you check if it's true: flip a coin or roll a die. On a high roll, it's true. On a low roll, it's not.

    If it's not, you have to come up with a second idea. Roll or flip again, same thing (50/50).

    As you can see, half the time you'll get your "natural intuition", but, less and less often, you'll get much less likely stuff.

    "Hmmmm. Ok, I guess he's not at the docks. Where else might he have gone? Maybe he got rounded up by the police... ..."

    I like to think of it as a quasi-discovery, and the process naturally leads from "obvious" ideas to (more and more rarely) the bizarre and unexpected. A nice balance.

    It's not that different, in theory, from writing out a custom random table ("1-3, he's at the docks, 5-6, he's been rounded up by the police, 6, he's actually run off with the Sheriff's wife..."), except much more time-efficient and elegant.
  • You know I've had such problems by running from the top of my head. I kinda want to have something like Colley's Rule in place. Finding the second good idea. Or the "but” principle. ”He's a drunkard… but on Sundays he's always in church praying his knees off to the Raven Queen for help sobering up"
    But I'm more into "not deciding” at all. I'll keep working on this today
  • I think the thing to do is first decide where that someone is and then determine what skill or skills would be most appropriate for finding them. In my game I have a Survival skill that you can take for any kind of environment (including Urban). So I often use that for trying to find someone in a crowd at a city alongside a successful Detect Roll. If it is more complicated than the players are looking for someone they pretty much know is in the streets somewhere (like they have to track the person down somehow), that I would play out with the players going around asking people in the city, finding different leads and maybe eventually finding the person. If the person doesn't want to be found, then it is going to be more of a challenge than just scanning the crowd.
  • Another, perhaps odd approach, would be to determine first how difficult the person in question is to find (roll a die if you don't want to decide), and then use that as a chance of discovery.

    So, say the DC for finding an NPC is 25. The players go about looking for them, even though you, as the GM, don't know where they are. You can still, of course, apply bonuses and penalties for the locations (finding someone in a swamp? hard) and their approach to looking.

    When the PCs hit that target number... they find the NPC. As simple as that.

    (Alternatively, you can just a roll a d6 every time they search, and, on a '6', that's where the NPC is. Whether they *find* him or not is up to their tactics or rolls, however.)

    That takes your hand "off the tiller" completely.

    But this all also comes back to what I was saying in the other thread about improvisation and prep: I think that so long as you have a consistent and solid approach to making decisions as a GM in the game/campaign, you can relax about any particular choice being "good" or "bad". After all, someone coming up with methods like these in the first place is clearly not playing with a railroading agenda in mind: you can trust yourself in that sense.

    Once you have a principled process, you can simply trust that it works, and go with it. (For example, Eero's 50/50 rule allows you not to "feel bad" about throwing out whatever idea comes to mind first, because half of the time you'll find you were "wrong". You could, of course, change the process to prefer later ideas: for instance, you roll one die, and on a 5 or 6, your first idea is true. But if it's not, add another die for your roll for the second idea, then three dice for your third idea, and so on. This way you'll mostly end up going with your second or third idea.)
  • edited July 2015
    "What would the system say?"

    I would look for similarly-shaped dynamics in the existing game rules - perhaps there's a rule for finding an item in a market, or some other needle-in-a-haystack situation. Then I would extrapolate from that mechanic to create a new one for Person-Finding. The idea is to preserve the system's own "feel" for certain dynamics.

    Of course that assumes I have a little prep time available. If I'm on the fly, I use the 50-50-descending-likelihood mechanic, which I sometimes call a Binary Tree.

  • (Binary tree - a great name for it!)
  • It's one branch on a binary tree :p
  • I see you have a problem with the term; it is used in the same way by programmers when describing a set of nested booleans or paired factors within an array structure, so you probably have a problem with those terms too. One never traverses the entire tree, that's true. But if you do it properly, you propose an alternate at each step, thus there is an unselected dangling branch at each level, suggesting Platonic treeness. Besides, not all trees are symmetrical.

  • edited July 2015
    I know :p
    My nick name at my old job was "the algorithm queen". I'm kicking it over at Euler project also.
    It was fun. People would walk over to my desk and say "Oh algorithm queen, what's the best way to solve-this-or-that". I'd drop what I was working on and give them an answer. Like "No, this has to be the denominator, see? And if you do this part recursively, you'd have to memoize this part or you'll run out of stack space." And most usually they even had no idea where to start so I'd do that too. Then they'd do all that messy implementation and testing and (brr..) actual deployment on the servers. What a paradise...
    I'm way more into pie in the sky coding and systems architecture than I am with things with real responsibility :/
    I guess my limitations are more emotional than they are intellectual :(
  • I'm trying to figure out your DM style from this thread to see if this idea would be useful, and I'm not sure so I'll just leave it here:

    Set a target number for different effects:
    x - Player gets a clue about known associates
    y - Player gets a clue about whereabouts
    z - Player finds the perp

    Each time they hit target number x (whatever it is) they get +1 on future rolls for that perp. Whenever they hit target number y, they get +2 on future rolls for that perp. Z is much higher, so you need to collect clues to stack bonuses. The clues can be dead ends, but the mechanical bonus sticks. So you might interview a known associate and that goes nowhere, but the +1 bonus still gets counted in order to represent narrowing down leads, etc.

    Now here's the part that I'm not sure you (or your players) will like: Whenever the player hits an x or y, the bounty hunter gets to decide the clue that is found. It seems like this mechanic is kind of side-questy (if it was a main plot element, you'd know where these people are) so it's mostly a side thing for the bounty hunter to do. Since that's the case, why not let them have a little spotlight and get to decide the details of how they track this person down? For myself as a player, I'd really value that freedom and creative input into the fiction.
  • What's "a main plot element"? I do not know what that is
  • Hmmm, good question, I used the phrase pretty thoughtlessly. I suppose (for an NPC) it would be one that you knew all the details about because they are important in your campaign world and you expect the players to run into them.

    I think the most important point I was trying to make by using that phrase is that it seems like these bounties are things that it would be okay for the player to make up details about because they don't affect other aspects of the game/fiction in super meaningful ways that might nullify or complicate your prep work.
  • edited July 2015
    Thanks for the clarification Ludo. I'll weigh your suggestion in with the other suggestions. It's pretty close to what I was leaning towards.
  • Last time in my game I tried something that gave me an instant adventure without using a random table: Roll 2d6 over the map, let players choose one (or let them choose a dice color and then roll the dice). Check the location closer to both dice; players start the adventure at the dice they choose and the number indicates how dangerous is that place, the other dice is the destination and again, the number tells how dangerous is the place.

    Compare both numbers and places. Why would PCs go from one point to the other, considering who they are and what they do? They can be escaping a dangerous place or going into one to kill something/get something. And if both dices are low, it means the danger is in the road to the place, so it can be an escort mission.

    Applying this to the where to place the criminal, when a player starts searching for someone you can roll one die on the map for each clue they get on his whereabouts. Take note of the location closer to where each die landed. The number indicates how close the criminal is to that place in days/hours or how probable is for the criminal to be there.

  • Thanks! Got this idea from Tony Dowler's How to host a dungeon so, thanks to him actually!
  • So many cool ideas in this thread, all very different from each other. How to choose between them? Curious to see what you go with, Sandra...

    I'll add one more option to the list: rather than establishing the existence of a person of interest and then rolling to see where they are, sometimes it's more efficient to include "are they nearby and/or accessible?" in the process of establishing who's noteworthy in the first place. Like, if the party needs a blacksmith who can work adamantite, and they start digging for info on one, "there is only one and he might be far away" needn't be a particularly likely result. "There are twenty and here's the nearest one" might be more useful.
  • edited July 2015
    This is what I'm doing now:
    I created a copy of our city map (which is pretty detailed, it's from Schley with the text filed off) and dotted it with the numbers one to twenty in pretty arbitrary locations.

    Then for each of the bounties, I rolled a d20 so I knew where they were. I really like this solution. It's something I can note very concisely on the list of bounties, or on any given NPC or glitchcult or inn.

    I also created a table, 1d30, filled with options from the "socialite challenges" in Silent Legions. I cherry-picked the like twenty-five or so that I liked best out of the "The NPC knows, but they want something" and "The NPC knows, but they fear something" tables. I also added six or so "They don't know" and one "They don't know, but they pretend to know, roll on 'want something'-subtable".

    I figured that like 75% or so of "I don't know" would be "realistic" in one sense -- that out of anyone in the city, it'd be like 1/4 chance of them knowing. (Those "socialite challenges"-tables aren't meant for random passersby, they're meant for specific NPCs that are definitely hep to the mystery of Silent Legions.)
    But I didn't want so many duds. I wanted the "I don't know" to be possible but rare. My players have a responsibility to only ask people that are pretty likely to know something. That are connected to what they're asking for. If they start abusing it, I can still wrap the table in a 1d4 "chance of rolling on the real table" thing. So far, I want it to be a side adventure, not just boring logistics / random canvassing.

    In play, this has worked very well. The first bounty they found, they got soft, and helped her instead of turning her over to the guards. Very interesting. I love my group, I don't try to predict them.

    Me knowing exactly where they were was a big load of my mind and a big stress relief.
    That helped me visualize things when they asked the innkeeper and I rolled on the table that the innkeeper was helping the bounty with food, but was afraid to say that, because the bounty had a connection to the glitchworld. We used petitioner/granter (thanks Robin Laws) to resolve the scene with the innkeeper and they got a lead on the bounty.
    She was hiding out in an abandoned marble temple in an overgrown garden (thanks to Schley's detailed map).

    They heard her sobbing in a linen closet, sitting high up on a big pile of rolled up cloth, pillows and blankets. They confronted her but found out she wanted out of the glitch cult. (This came out of a table in Red Tide, thanks Kevin Crawford for that. But that was pre-noted in the notes I had on her.) She was wanted by the city guard, capital punishment, for taking part in a ritual.

    She gave them directions to the glitch cult's hide out where the players had an adventure with both sneaking, social, exploratory and combat encounters. I had prepared a couple of the rooms (and all of the NPCs) and improvised other rooms, like corridors and other pass-through areas. Here also our detailed city map helped me a lot.

    This AP report is also relevant to the myth/no myth discussion. I love having lots of prep (as long as I can get an overview of it and find my way around the prep and not miss anything -- unlike Princes where I've made lots of mistakes). But there is going to be holes and then I'm going to the tables. And if there are still holes after looking at the tables I've got, I've got to improvise. Which I can do, having run no myth for a decade.

    Is it as fun to walk around in my improvised areas as it is to interact with a fun old school "toy dungeon" module? No, it's not. Not in my opinion. But it's what I had.
  • Excellent post, Sandra. That's exactly how a good "semi-prepped" approach in a sandbox can work, I'd say, and it gives good results.
    We used petitioner/granter (thanks Robin Laws) to resolve the scene with the innkeeper and they got a lead on the bounty.
    How did that work, exactly? Sounds interesting.

  • I created a copy of our city map . . . and dotted it with the numbers one to twenty
    for each of the bounties, I rolled a d20 so I knew where they were
    I also created a table, 1d30, filled with options
    - The NPC knows, but they want something
    - The NPC knows, but they fear something
    - They don't know
    - They don't know, but they pretend to know, roll on 'want something'-subtable
    My players have a responsibility to only ask people that are pretty likely to know something.
    If they start abusing it, I can still wrap the table in a 1d4 "chance of rolling on the real table"
    Love it! Great system.
  • David, thanks for the summary!

    The players wanted something from the innkeeper. Whether or not he knew where to find this girl, Tiliva. [Implicitly, but not stated at the table: the innkeeper is the granter, the players are petitioners.]
    We went into the mode where you say what your character says; I was portraying this innkeeper.
    I was like, shuddering, and going "No, I don't know her! Who told you?" [Being a little transparent to set up the scene -- but by now, I can be pretty subtle and the players still pick it up.]
    And they were like "We think you do know something". [Implicitly, tactic: laying the cards on the table and asking for the same]
    And I was like, fearfully "No. I'm not involved". [Implicitly, tactic: flat out denial]
    And they were like "Who're you afraid of?" [Implicitly, tactic: good cop]
    And I was like "I just don't want there to be any misunderstanding. I don't want you to think I'm caught up with that glitch stuff. You know I hate it." [Implicitly, tactic: being clear, setting boundaries, identity politics]
    And they said "GLITCH! WTF ARE YOU IN WITH THE GLITCHERS YOU SCUM!" [tactic: bad cop]
    And I: "No, it's not like that at all, that was what I was trying to say. I was helping her, but..."
    They: "WERE YOU HELPING THE GLITCH? That's so wrong!!!"
    I: "This is what was afraid of. You guys jumping the gun. Look... I can't explain it but... I'll tell you where she is if you promise not to turn her into the city guard." [The innkeeper diegetically thinking: they'll understand once they see her. Tactic: bargaining.]
    They: "Sure, sure, we promise, now where is she?" [Tactic: false promise, deception]

    Robin Laws' idea is simply to be mindful of motivations/desires and of "tactics".
    Guys as you know I'm pretty challenged when it comes to social skills.
    I was doing allright in my no-myth days but I had a lot of trouble portraying random shopkeepers in those first few sessions of The Lost Mine of Phandelver.
    I picked up Unframed and read Laws' chapter and everything became much crisper and better and more dramatic in the dialogue scenes, both the players and me did better just by me seeing the "petitioner / granter" structure and tactics exchanges in my own head, not even mentioning it to the players, just talking the scenes out old school style.

    But after a new player joined he was really trigger happy with "I ROLL PERSUADE" "I ROLL DECEIVE" etc, and I was struggling a bit with it, the other players approached me and they were like "It's kinda frustrating that you need social skills to resolve social situations".

    And I was like. OK. We need to whip out Hillfolk proper. I want to try it.
    And we tried it.
    And we've gone back to D&D but informed by the idea of "dramatic scenes", even though they've nixed the drama tokens (for now). So far (two sessions in, since our DramaSystem game) it has worked very well.
    Heaven knows that I didn't have any social skills. But this way of thinking has really helped me and now I think it's helped the players, too.

    I know that Laws had fiction in mind, it grew out of Hamlet's Hit Points and analyzing cable TV shows.
    But it's also similar to the teachings of behavioral therapy on social skills.
  • Whoa. If you are down to make that post its own thread, I'd love to follow up on it. This sort of useful awareness during dialogue roleplay is really interesting to me. How do you cultivate it? What problems does it and doesn't it solve? Stuff like that.
  • (Me too.)
  • That's an interesting description of how you break down the seemingly freeform dialogue in tactical terms. I have a similar analytical viewpoint to that stuff as well. I haven't seen that Laws treatment, but sounds to me like it's just the thing for this sort of game.
  • I'd be ripping of Robin Laws! :) C.f. Unframed, it doesn't overlap completely with Hillfolk (but Hillfolk also has useful info on the topic). I was so down on Hamlet's Hit Points and gave it such a negative review... but if I had known that he was building up to DramaSystem I would never have! I misinterpreted it as a big ol' railroading game.

    How to cultivate it? Same as any mindfulness excercise. It's about remembering to use it -- remembering to identify petitioner/granter and motivations/personality (which leads into tactics, whether you think about it or not).

    Whenever two people in the game talk, think "Who wants what from whom, and why can't they get it?" (while keeping in mind that "no reason, they can get just it immediately" is sometimes a good answer).

    It solves the feeling of vague, meandering, tame dialogue scenes and creates crisp, to-the-point, clear, satisfying and unambiguous dialogue scenes while still (paradoxically enough -- maybe I'm just seeing this apophenically) allowing or even affording great subtlety and nuance.

    It also very well manages the length and framing of such scenes (and empowers players to set the pace of dialogues).

    What doesn't it solve?
    It doesn't feel like SotC/Diaspora (pre-Fate Core) social "combat", nor like Duel of Wits from BW/BE.
    It's more like being an actor.
    If you want highly mechanized social situations, this isn't the system for it. For example, reaction tables, charisma rolls etc, they lose a lot of their meaning with this paradigm.

    The traits from 5e gain a lot of meaning in a system like this, though, and they're more my speed anyway.
  • No new thread? If I pursue this and reply, it's going to both kill the "find someone" discussion and miss any newcomers who might be interested in dialogue logic but not the thread title.
  • I didn't know how to move over the posts. Plus the first page in the new thread ought to be Laws' text and I don't know if that's even legal. It's only like three pages. How do I solve both those problems?
  • edited May 2016
    I made a new post David!

    Edit, one year later: Here's the link. Back when we made these posts, both threads were current and thus easy to find.
  • edited August 2015
    Glad of that, because I thought of a new mechanic for lost people today and I didn't know where else to post it. Basically it's a combination of the binary roll with a distance indicator.

    Roll 2d6. Consider high rolls to be "the obvious direction" and low rolls to be "the unobvious direction". The difference between the roll and the midpoint (7) represents the distance they traveled. So this one roll represents "how far and how obvious" their trail is.

    Optional rule for more complex/random trails: If you roll doubles, the trail ends there; the quarry is hunkered down somewhere in that location. Otherwise this is just one trail segment, and you roll again for the next one, repeating this process until you've gone the maximum distance the quarry could have moved (or rolled doubles, whichever comes first). Pursuers may need to make a new roll (depending on the game system) to pick up the trail whenever one segment ends and another begins.

  • I got surprising mileage from both parts of this mechanic tonight.

    One of the monsters I threw in on a whim in a dungeon (not even the most primary dungeon for the city, but they went there after a random rumor) and forgot about was a creature called Intellect Devourer. I only chose it from the CR list in the DMG, I didn't even look it up. It was a deadly encounter when I built the dungeon for two level three players, but one more player joined the game so it was a just hard encounter.

    In my campaign, I've refluffed every monster and I described this creature as a metal, British-style round cookie tin on sharp insect legs.


    It killed the healer and WENT INTO HIS BRAIN!!! I took the character sheet and took over the role while that player rolled up a new character.

    We played out a tense "short rest" out in the jungle, like a stand off, as they tried to figure out what was going on. It was classic author stand / actor stand stuff since they knew that the DM was playing their character. We RP:d it out, DramaSystem style! It ended with the "cookie-tin" leaving them "I'm just gonna take a leak in the jungle" but started heading for the city.
    They hid from it and I described it (the statblock says it can detect people via psychic means) that it looked straight at them, opened its mouth and revealed the cookie tin.

    Then they ambushed it and killed their own friends body, and the cookie tin then took over one of them! (Everything by the RAW.)

    The last surviving player, with six hit points left, rushed to the city (eight hours + exhaustion level 1) and went to the inn and took a long rest.

    Eight ours of rest later, the player woke up and found two crooks (the newly rolled PCs) and hired them as mercenaries to find this monster.

    And I used the "where are they" numbered map to roll where this cookie tin in hobbit body would go first in the city. It was the very same inn (a one in twenty chance!!!). It took over the innkeeper (their trusted NPC since three sessions back)!!!

    And then I could get good use of the "what do they know" tables as they canvassed the city, putting up lookouts, asking small time crooks for where this hobbit had went. And they found a witness that had seen him enter the inn.

    They questioned the innkeeper but I tried my best to have STONE POKER FACE. But they searched for the hobbit's body and found it so they killed the innkeeper. Then the cookie-tin managed to kill the last remaining PC and jumped into his own body, and then we played that out DramaSystem style (with me in the role of the PC, and the other two players with their new characters, the hired crooks). It ended with the former PC going off into the city. They shadowed him (they don't know that he can read their thoughts) and he went to another random location -- it happened to be something that makes perfect sense because he can report to his liutenant that I've placed there.

    The dice help the city really "come alive" like an ant farm and I can just sit back and enjoy this!

    What a wonderful session when it comes to technical agenda.
    Really really happy with it.
Sign In or Register to comment.