What’s the function of a Stronghold?

Working on my Ruins of the Shattered Empire campaign, I was thinking again about Kenshi, a wonderfully weird sandbox indy game set in a desolate post-apocalyptic wasteland full of bandits, robot skeletons, ancient vaults, ninja, cultists, crashed satellites, insect men, random orbital lasers from the sky, and flesh eating giraffes. The game has no story. You just start somewhere in the desert, with nothing but the shirt on your back – if your character is one of the lucky ones – and your only goal is to survive by getting something to eat and avoid getting eaten yourself. Unless your character is one of the mentioned robot skeletons. It’s a wonderfully odd game that feels like something that would have been made in the early 80s if the technology had existed back then. People who like things like Veins of the Earth or Ultraviolet Grasslands will probably appreciate the style. There are various bare bones NPCs around the game world that ask to join your team or can be permanently hired for a one time payment. The desert is full of hungry beasts and nasty bandits, and while it certainly is possible to play the game as a lone wanderer, a very attractive option that opens up very early on is to build a small base with a wall that protects your people while mining ore to sell in a town or working on a patch of dirt to grow your own food. You still keep getting attacked by raiders who’ll easily break down your gates after a minute or two and loot your little storage shed, and so you can easily find yourself in an endless cycle of expanding your base to provide more food and income to expand your group with additional warriors, so that you can expand even bigger to add your own workshops to make your own weapons and armor instead of having to buy them. It’s often compared to Dwarf Fortress and Rimworld, with it’s own special type of weirdness and hilarity.

In addition to building materials, you also need to first research construction plans for new buildings and equipment, and for that you need books of ancient knowledge. A few of which can occasionally be found in stores for a hefty price, but the more rare ones require you to go explore ancient ruins in increasingly more dangerous parts of the massive wastelands. And setting out on an expedition to find and explore these ruins always reminds me of good old D&D wilderness adventures. The ruins themselves are all pretty small, so I wouldn’t call looting them dungeon crawling, but getting your group of scavengers to those places and hauling back your loot to your far away base is just like wilderness adventures should be.

I’m also now remembering how I always found the use of NPCs as part of PC’s “gear” in Apocalypse World a really cool approach, and how they can be used as really nice adventure hooks, but I don’t want to go onto another tangent and actually get to the point.

Base building in Kenshi is in many ways an economy sim in which you figure out how to assign your characters to different jobs, optimize workflows, manage your resources, and invest your profits into new technologies and expanding your operations. It’s a lot of fun on a computer that takes care of all the math, and you spend hours upon hours on it by yourself. It’s not something that translates to playing a roleplaying game as a group. All the stronghold building rules for RPGs I’ve come across so far fail because of this. But the aspects of defending your stronghold against raiders and having to go out into the dangerous wilderness to gain resources you need to maintain and expand your base are also endlessly exciting, and those activities are the daily bread and butter of D&D adventurers.

I am really intrigued by the idea of giving the players the tools to take over any abandoned or cleared out ruin, fixing it up and fortifying it, and using it as their main base of operations while they are exploring the surrounding wilderness. And after some pondering on the subject, I believe the best way to approach this is not to start with any mechanics for upgrading a base or price lists for various expansions, but first figuring out what kinds of functions the stronghold should play in a game that is still fundamentally about going into dungeons to find treasure. This really is just throwing around some ideas and sorting out my own thoughts on this.

What a Stronghold should be for

Safe Resting Place: This really is the primary function of a stronghold in the wilderness for adventurers. A stronghold provides a place where the party can rest and recover from their ordeals without having to make wandering monster checks. I plan to run the campaign without clerics, so healing either takes a good amount of time to recover naturally, or use up healing potions that are valuable and can not be infinitely replaced. This should make a place where the PCs don’t have to worry about monster attacks.

Treasure Vault: If the players have their stronghold guarded by mercenaries while they are out of adventures, I would consider storing their new treasures in their vault as having “returned with treasure from the wilderness”. Since they are also no longer under constant threat of being attacked, that means they have completed their adventure and can get the XP that their loot is worth.

Supply Depot: In addition to storing treasure at the stronghold, the players can also store supplies of food, water, ammunition, lamp oil, and tools. The stronghold might even have its own well or cistern to provide an endless supply of water. Using their base as a supply depot means that the players don’t have to carry as much supplies to get to the dungeon and back, and if they should be running low while in the dungeon, a resupply trip to their stronghold would be considerably shorter than returning all the way to the nearest town. Of course, they first need to get the supplies from the town to the stronghold, which can be a small side adventure in itself.

Necessity of Hirelings: I love hirelings as a game element, and really want to see wilderness adventures turning into large expeditions of a dozen people or more. Having just four or five PCs as the whole party is nice for a lot of campaigns, but I think wilderness exploration campaigns should be at a much larger scale. Wilderness exploration is more than having one outdoor combat encounter between the town and the dungeon entrance. That’s the kind of game the Expert rules are for. Having a stronghold full of supplies and treasures means the players need someone to guard all of it while they are away. And they probably don’t want to leave some mercenaries they picked up in a tavern alone with all their money for days on end on a regular basis, so they should also have some trusted retainers to leave in charge while they are gone. With a more permanent base, you also probably will want to have additional servant staff to cook and make repairs, tend to the animals, and you can see how this can escalate very quickly.

Money Drain: One thing that lots of people have been thinking about a lot for a very long time is what players should be doing with all the money they make on their adventures. Especially in a campaign where XP are gained from finding treasure to mechanically support the PCs’ endless hunger for more gold, the whole thing becomes increasingly less believable if the characters are already drowning in more gold than they know what to do with. Conan is always up for an opportunity to steal some gold because he’s constantly broke. In such stories, the heroes spend their loot on wenches and ale, but enjoyment of luxuries is not something that you can really get across through the mechanics of a game. Players saying that their characters go on a massive tavern crawl after an adventure is maybe fun once or twice, but stops feeling rewarding after that. A stronghold is a great way to drain the coffers of the PCs. Every expansion or upgrade to their base costs money, and all the guards and staff need to be constantly paid for. The wages are pretty cheap, but if you include proper tracking of time (without a meaningful campaign is impossible, as you know) then all the time that the PCs are spending in the wilderness while searching for ruins, days spend healing from injuries, weeks spend learning new spells and creating potions, and whole months stuck inside waiting for the end of winter, this all adds up.

Merchant Access: This is related to the aspect of Supply Depots above. Once the players have established their stronghold and have to make regular runs to the next town for considerable amounts of supplies (all those hirelings need to eat), they can become important enough customers for traveling merchants to make detours to sell their goods to the PCs. Maybe not with whole wagons, but at least with a handful of mules. In addition to regular supplies, such merchants can have a number of special items for sale that the players might be interested in, like potions or maps, and also provide the players with new rumors when they are away from civilization for long.

Trouble with the Neighbors: Even with solid fortifications and mercenary guards, the treasures and supplies inside a strongholds will attract all kinds of people and creatures. Some might be out to raid the place, while others might simply not appreciate newcomers in their territory. The possibilities for adventures beyond the default treasure hunting are endless, without the typical situation of sending the players to chase after prepared adventures. Pacifying the surroundings is a good way to let the players be proactive and deal with situations in whatever ways they come up with, without giving them a villain with a plan they have to stop before it is too late.

What a Stronghold should not be for

Economy Sim: As I mentioned earlier, managing your resources and working out production systems can be a lot of fun if you’re playing by yourself on a computer, but just isn’t something that works as a roleplaying game. Adding a smithy to your stronghold or constructing a wind powered water pump for your well can be fun and exciting, but I think it really shouldn’t turn into a resource management game.

Generating Income: In Kenshi, I started my first base as a small mining camp to simply mine ore, smelt it into metal plates, and sell them in the next town to make money with which I would buy anything else I need, such as food and medicine. Getting your stronghold self-sufficient and even profitable is a fun idea, but that would go directly against the overall premise of the campaign and one of the main purposes for having a base. The upkeep costs for having the stronghold is meant to provide the financial pressure to keep the PCs going into dungeon to search for more treasure. The strongold being a source of money instead of a giant money sink would work completely opposite to that. While being landowners with servants working for them can be a fun idea for some roleplaying games, it just doesn’t fit here.

Seat of Government: Related to the point above, becoming the biggest dog on a stretch of the frontier and clearing the surrounding land for settlement can be a great motivation for characters. But once you get into that kind of stuff, there’s not going to be much room or time for continuing to go dungeon crawling. You could still go into underground places to fight the enemies of your domain, but then you end up with a completely different type of gameplay from sneaking around in the dark to steal treasure without alerting the inhabitants.

This as a broad overview of where my thought are on this subject at the moment. We’ll see if I’ll get around to put further work into this and develop it into some kind of system with established mechanics and procedures.

Discovering Sites in the Wilderness

I’m a big fan of wilderness sandbox campaigns, but never been really enthusiastic about the hexcrawling approach, in the sense of “go from hex to hex until you find something”. A 6-mile hex is something like 80 km². Even a large castle might not be noticed while simply moving through such an area, and if the area is forest or mountains, you would have to run straight into it. Spending some amount of time to search a hex to see if you discover something also doesn’t seem convincing to me. If you’re a treasure hunter, you wouldn’t just pick a random spot in the wilderness and start searching it with a fine comb. That takes way too long to find anything of interest. What I believe adventurers would do is trying to make their way to sites that they already know about and that look promising for holding treasure.

Under this approach to adventuring, the players first need to have clues where to look for treasure and adventure. So here’s a couple of ways that PCs can learn of new sites to add to their own map.

Highly Visible: Castles and watchtowers are commonly build on high points where they can overlook a lot of the surrounding areas. Sites like that could be spotted by simply being in the same hex they are in. Or in particularly clear terrain, even by being in a hex next to it. Though if the site is hidden among trees or mountains, it would remain hidden even party is moving through the same hex.

Sites on Roads and Rivers: If a site is directly on an old road or a river that the party is using for navigation, the players discover it automatically when they pass that spot. In some cases, it might even make sense to road signs or something similar point the players that something worth investigating lies down a side path from the road the party is currently traveling on.

Found Maps: Players can find maps among the treasures they pick up which show some sites that are known to them, and some sites that are not. This allows them to go search for and discover the new locations by following the clues on the map. They could also buy maps from certain individuals, or be given a map as a reward from grateful NPCs.

Rumors and Quests: Locals simply tell the party about sites they know but are not on most maps. This also provides the players with some vague idea of what they might have to deal with when they get there.

Following Tracks: After a random encounter, if the players try to follow fleeing enemies or follow their trail to where defeated enemies came from, the tracks can lead them to a nearby site where the creatures have their lair. If no matching site is anywhere nearby on the GM’s map, a quick lair can be put together with a small cave or campsite plan and rolling up a lair encounter by the wilderness encounter rules.

By using all five of these methods to give players hints where they can find new sites they didn’t yet know about, it should be quite easy to make it all feel quite natural and a consequence of the players’ actions, rather than the GM deciding the party needs a new site to be send to. It’s not a big red glowing sign telling the players “the next prepared adventure is here”.

Random thoughts on alchemy and sandbox campaigns

With the top level worlduilding for the Shattered Empire having reached a pretty concrete and solid stage, I’ve started turning back to thinking about specific details for an actual sandbox campaign set somewhere in this world. And while discussing some ideas, we happened to talk about players spending the money they get from all the dungeon crawling, travelling merchants as random wilderness encounters, and how generic magic weapons are boring compared to setting up your gear to fight a specific monster. It got me thinking about how the Witcher games don’t let you drink potions or put poison on your weapon while a monster is trying to claw your eyes out. You got to do your alchemical preparations before you step into the monster’s lair, perhaps even brewing up some special brews you didn’t normally keep stocked in your inventory.

And that discussion gave me an idea: Let’s have alchemical ingredients that can be used to create potions to help fight specific monsters, but these ingredients are hard to get and can each be used for a number of different potions.

Usually in videogames with crafting systems, you have basic ingredients that you can pick up from defeated enemies or growing in the environment in practically infinite amounts, and special ingredients that are only dropped from a single monster in the whole game to make one specific item, or make you choose one of two possible items that can be made with it. For the record, I hate crafting in videogames. It’s mostly a pointless waste of time that clogs up inventory space to drown you in junk that is worse than stuff you find. And I think one of the main flaws with these mechanics is the way they treat the ingredients.

But let’s approach this differently to make it something that is narratively meaningful and doesn’t turn into a convoluted minigame. The really badic ingredient, which are dirt cheap and just grow everywhere in the environment, are not even worth tracking. They have negligible costs and require no effort to acquire. And to make the special ingredients much more interesting and meaningful, make them both limited and expensive and make them useful for multiple different things. Letting the players fight a monster and take ingredient A from it, which they can turn into item X is really just a different way for giving them item X from the monster’s treasure. But say instead that there are maybe a dozen of the monster in the whole sandbox and the players might end up with anywhere from 0 to 12 of ingredient A, depending on how they explore the world and interact with the creatures, and that each of the ingredients can be used to make item X, Y, or Z. If these items they can make have limited uses that mostly shine in quite specific situations, the players will have to make a choicd if they want to use the three ingredients they have to make one of each item, three of the same item, or two of one and one of another. Since they don’t know what situations they will be dealing in the future, the correct answer is to keep them stored and make whatever item is needed, once they know it is needed. This adds a few additional scenes to adventures in which the PCs are making specific preparations for the adventure ahead, based on the information they have of what they arw getting into. That’s always great to make a world become more alive and present.

But also, you can make these ingredients available in small quantities for high prices at merchants. If it’s travelling merchants, it’s going to be a one-time offer as well. If the objects that can be made are sufficiently strong and have appropriately limited uses before they are used up, then these ingredients might be worth quite a lot to the players. And asking to take a look at the alchemy chest of each merchant they encounter can be a highlight of each shoping trip.

Buying the ingredient isn’t the same as buying the final item. There’s still alchemical work to be done. If it’s something that takes longer, it can be useful to employ an alchemist, especially when this kind of work comes up regularly. Which also costs money. And that alchemist is going to need a lab. Don’t let him do that stuff over a campfire with a little box of alchemy supplies. A proper lab best fits into a permanent base, which can be a much bigger source of expense for the party.

I’ve not done any work yet on either ingredients that could be found on adventures or be one-time offers by travelling merchants, or the kinds of items that could be made with them. But I think it’s a really interesting direction to follow that could potentially add a lot to sandbox campaigns.

The Pencil is mightier than the Board

You know all these kids these days, with their battle maps, miniatures, and VTTs? I think it’s one of the worst aspects in which RPGs have gone into the completely wrong as they developed, following right behind the total mistake of adventure paths, campaign books, and “telling a story” (the curse of Dragonlance).

It’s gonna be this kind of post.

Now I am not that old. I actually played way more videogames than I’ve run RPGs in my life, and been doing it for much longer. Baldur’s Gate was my first introduction to D&D, and in my early days as a GM I spend way more time as one of the admins on a German Neverwinter Nights server than playing at a table with my first group. It’s not that I am opposed to any of that on principle. But in my opinion, when it comes to playing an RPG with character sheets and dice, and having a thinking GM running the show, all this stuff like highly detailed battlemaps and fancy looking tokens don’t actually work as visual aids, but instead are serious distractions from the game.

This also doesn’t have anything to do with 5th edition or how kids these days are playing the game. The whole issue goes back to at least the launch of 3rd edition over two decades ago, when the game was designed from the ground up to revolve around position markers on a square grid. (Which was a really strong incentive to entice players to give the publisher more money for miniatures, which were already a money printing machine for Warhammer at that point.) But it seems to me that even with 5th edition significantly cutting back on that as the game mechanics are concerned, all the tools available to make very impressive looking maps yourself, various brands of plastic dungeon floors and walls being around, and the ease with which people can show of images of all swag (I know, “old man”…) seem to have made all this stuff much more prominent.

But my stance is that all this superfluous stuff does not make your campaigns better, but instead detracts from them. It might very well be that this perception is impacted significantly by how ADD affects the brain processes information, but to me, highly detailed battlemaps or fancy toy soldiers don’t inspire my imagination. Instead, they lead to a greater abstraction and create distance and detachment from the scene. I am very much in favor of running adventures with basically an empty table. To me, the table is a surface to roll dice and put your papers down to write. Indoor maps can – and should – be nothing more than a simple sketch to communicate the general layout of the area when explaining it verbally woul be too cumbersome, and regional maps are best handled as props in the form of sheets on paper that look like someone in the game world drew them by hand.

Ingame map from Thief.

The less visual input you get from stuff that is on the table, the more the mind is encouraged to create an environment in your head that reflects the details in the GM’s description of what you see. At least in my own personal experience, when you shove tokens around on a square grid, you start perceiving the encounter as a logic puzzle that is to be solved by computing an optimal sequence of moves.

Something like this looks very fancy and perhaps even evocative. When you first look at it, you might even start imagining for a moment what the place would look like if you were a person stepping into it through a door. But as the encounter progresses, all of that color that hints at an environment starts to fade away, and the only thing you really keep seeing is the grid lines and the barriers that block ranged attacks and spells. Because those lines are what is relevant information if you play like that. The grind lines are the signals, while the colors that hint at a stunning and memorable environment are  only noise.

To take the most extreme case, who ever imagines the battlefield, hordes of peasant levies, mounted knights, and war elephants that are represented on a chess board? It doesn’t happen. They are simply a grid on a board and abstract tokens that can be moved within certain rules. And even when you have the action of a fantasy RPG being connected to a non-mechanical story, I see this mental process of abstraction happen all the time. In an RPG, encounters should be scenes. They should not be logic puzzles. Or as someone phrased it many years ago, “don’t look for solutions to obstacles on your character sheet”. That’s not what RPGs are supposed to be for. That’s not where their amazing unique potential lies.

Moving tokens around on a terrain map is something that goes back to wargaming before roleplaying games were ever a thing, so it’s not like there ever was a moment in which they were introduced. But wargames have a completely different intention to RPGs. Their whole point is to be an abstraction and a simulation, to play out  how battles might have turned out differently if certain tactical decisions had been made. Making things more objective was the goal for both military planners and armchair generals. Imagining yourself as an officer on a hill talking with messengers delivering reports and orders was never the goal. But that’s exactly what fantasy RPGs are about.  To imagine yourself stepping through great ancient gates, or descending down a damp stone staircase into the darkness. To see vague forms slithering in the shadows and staring into the face of a demon as it rises from a burning pit. Putting a toy demon on a square grid just doesn’t deliver that.

There are many situations where less is more. And less visual aids means a more vivid imagination. (If the GM is doing a decent job with descriptions, but that’s a different story.)

Another thing that keeps happening to me is that any time I work on a new world map, I always get fed up with whatever shiny map making program I have and keep going back to go grab a pencil and some graph paper. Not quite sure why, but the process of designing environments always goes so much better and faster by just using a pencil. I guess in some part, it might be because of the limitations of the tools. There is no choice of colors or different line strength. Putting down a line is really easy, as is adjusting one. And there is no temptation to make everything right and good looking the first time so you don’t need to redo it all later, because you know the result will be a smudgy mess and making a nice handout to show the players has to wait until all the placement decisions are locked in. I’m not even hyperbolic when I say designing maps is probably 10 times faster on paper than doing it on a computer.


The 5th edition of Dungeons & Dragons was written with the specified intention to use simple language and not to get too caught up in excessively  granular definitions. Which is generally a good idea, given how fiddly the 3rd edition was, and how dissociated the mechanics of the 4th edition were perceived. GMs are supposed to use common sense, or make a ruling about how some unusual interactions between two rules should work out in their campaign. That’s how most games other than 3rd an 4th edition work. But of course, there will be some casualties along the way. Blindight is one of them.

The definition of blindsight is terrible. “A creature with blindsight can perceive its surroundings without relying on sight, within a specific radius.”

Refuses to elaborate further.

Goes to printers.

That really doesn’t tell us anything! “Blindsight is a sense”, it is telling us, but nothing more. It really is completely useless. What blindsight actually is and what it does is something GMs have to make up themselves. I always feel that a good starting point is to go back to older versions of the rules and look at earlier definitions of a mechanic or rule. Of course, writers can always choose to use an old technical term and apply it to a completely new concept, but I think in that case they would have taken more care to explain the new mechanic and not forget to write it down “because it seemed obvious”. 3rd edition fortunately has much more detail on the effects of blindsight, which greatly influence my personal ruling on what it does in my 5th edition games.

The name blindsight implies that it is very similar to sight, but operates even if a creature has no sense of vision. Within the range of the blindsight, we treat the creature as if it could see, even when it can not. The primary effect of this is that within the range of the blindsight, the creature ignores the effects of blindness. Though we have to distinguish here between the mechanical effects of blindness as specified in the game rules, and all the other limitations of not being able to use vision.

When a creature or object is being lightly obscured or heavily obscured, it inhibits the vision of creatures, but apparently nothing else. You can still hear through sources of concealment, shot projectiles through it, or cast spells through it. As such, I would rule that to a creature with blindsight, nothing is ever obscured. (And all that lighting conditions ever do is to make things obscured.) Though I probably would also rule that to a creature whose blindsight specifically comes from its sense of hearing, environmental noise could have the same effect on their blindsight as obscurement does on vision.

Now does blindsight allow a creature to perceive things and other creatures that are behind cover? My ruling is that it does not. If a creature has an exceptional sense of hearing, then it already has advantage on Wisdom (Perception) checks that rely on sound. If a character is trying to hide behind some crates that provide full cover, from a grimlock that isn’t actively searching, I would probably ask the player for a Dexterity (Stealth) check to be extremely quiet. While no check would be needed if it was a goblin instead. Heightened hearing works around corners, but blindsight does not. It’s still called sight, after all.

However, a big question mark still remains and that is the topic of illusions. The spells blur and mirror image both specify that they don’t have any effect  on creatures that have blindsight. But no other spells do. The most important one being of course invisibility. In 3rd edition, things are very clear. The description of blindsight says that it makes invisibility irrelevant. The 5th edition description of the invisibility status say “An invisible creature is impossible to see without the aid of magic or a special sense. For the purpose of hiding, the creature is heavily obscured.” Blindsight certainly is a special sense. And when my ruling for ignoring obscurement is applied, it definitely allows the normal perception of invisible creatures and objects.

The minor illusion and silent image spells allow you to create an image. I would rule that a creature that can’t see the image with vision (normal or darkvision) would not even be aware of its presence. A creature that has both vision and blindsight would see it only with its regular vision, but not its blindsight. So I would rule that such a creature would automatically make Intelligence (Investigation) check to identify it as an illusion without having to use an action to examine it. In contrast, the major image spell creates an image that includes the right sound, smell, and temperature of the illusionary object, so I would rule that this fools creatures with blindsight just as well as any other creatures. I’d make the same distinction with the hallucinatory terrain and mirage arcane spells.

Skill Rulings

Somehow I managed to run my last D&D campaign very successfully without really paying much thought to skills. And I can’t even really remember any specific instances where I had the players make skill checks. Surely we can’t have been playing for five months without any skill checks coming up, but with the way things played out, there wasn’t much need for them. For most interactions, the outcome happened to be obvious without a die roll needed. But going forward, and aiming for adventures with more ruin exploration and investigation of strange phenomenons, I think it will be useful to properly look at the skill and come up with a general ruling on when and how they are used in my next campaign.

When to call for checks

As a general rule of thumb, I like to go with  the approach of “Assume Competence”. The PCs are adventurers, doing adventuring stuff for a living. They also are natives who have lived their entire lives in the setting of the campaign. If they encounter something that someone in their position would be likely to know, there’s no point in being shy about the information. Just tell the players what they are looking at, and the relevant context of what it means. Similarly, if an action is something that you wouldn’t think of being a problem, and the off chance of a possible failure wouldn’t carry big significant consequences, there is generally no point in calling for a check. As such, I think checks with a Difficulty Class of 10 (easy) are generally not worth rolling unless the PC in question has a really poor modifier, and DC 5 (very easy) checks only make any sense if the consequences of failure would be catastrophic. (I like making players deal with what fate has given them, and carry on forward in the face of defeat. Success in any venture is never guaranteed.)

How to roll checks

Another policy I’ve adopted in the past is that ability checks and skill checks are always called by the GM, never declared by the player (who then usually rolls a die without waiting for a response). I like to first talk through the whole situation and make sure the player has understood what’s going on, and I am clear on what the player is trying to accomplish. A die roll is almost always made to generate a randomized answer to a question. It’s important that we’re all on the same page what the question is first. And then, as the GM, it is me who decides which ability or skill applies to the attempted action. Something that I’ve not done yet, but absolutely plan to going forward, is to always declare the DC of the roll when calling for a check. This has three reasons.

The first thing is that it speeds up play a little bit. When a player shouts “I want to do X, and I’ve rolled a 16”, I need to take some time to think about what an 18 means in this situation. If a roll is a 2 or a 24, it’s obvious if that means success or failure. But a 13 or a 16? Deciding if the DC should be a 10 or a 15 retroactively after the player already announced the result is a situation I would refer to as sub-optimal. Now I’ve to make a judgement call I really don’t want to make in the first place. That puts me in a weird spot and takes time while the players wait for me to decide what to do. And with my ADD, I tend to get moments of brain lock in these situations and take even longer to sort out my thoughts what just happened and what I have to do now that the player announced a number.

The second reason is dramatic. If you declare the DC before the roll is made, all the player can stare at the dice to see how it lands, and if the player mentioned what the modifier is on the roll, they get the result immediately. The player does not have to tell me the result and I don’t need to come up with an eloquent way to describe the outcome. It’s already there for everyone to see.

And thirdly, it establishes that I as the GM am completely disinterested in the outcome of the skill check. The players propose an action, I tell them the DC. All the responsibility of what happens next lies exclusively with the players deciding to take the action and the roll of the dice. I am not influencing the outcome of their plan one way or another. (Though, of course, I still have a great degree of creative freedom of what a success or failure actually means specifically.) This is an essential component of actual open-ended sandbox play.

Specific Skills

Strength (Athletics) is very straightforward. Make a judgement call on whether a physical task is medium, hard, very hard, and call for a roll on the corresponding DC. Failure while climbing means the character makes no progress that round. Failing by 5 or more means the character falls at the halfway point of that round’s movement.

Dexterity (Stealth) is checked once per “obstacle”. Any group of guards counts as a single obstacle, regardless of how many guards are in the group. Getting to the stairs while staying out of the lamp light is one obstacle, going up the creaking stairs without making noise is another obstacle, even through they could both be done in the same round. To hide in combat, the character first needs to break line of sight to any enemy you want to hide from (movement) and then conceal yourself in a hiding space while being observed (action). To attack from hiding, against an enemy who knows you’re somewhere nearby (because he just saw you run into the room or around the corner seconds ago) you need to be able to move from your hiding spot towards the enemy and make the attack in a single turn. Sneaking up on enemies who are currently in combat is impossible, unless you’re invisible or something close to it. Sneaking can be done as a group check. If at least half the members of the group get a Dexterity (Stealth) check that beats the Passive Perception of any guards or monsters, they are guiding the other characters with hand signals on when to move and where to step to also make it undetected successfully.

Intelligence (Arcana, History, Nature, and Religion) are usually passive skills. Assuming Competence, if something the players encounter would be known to all sufficiently educated people, they get the information for free as part of the initial description without specifically having to ask for it. The GM is the eyes and ears, and also setting knowledge of the PCs. These are things the players can’t really ask for unless they already know their significance. Skill checks can still be made in situation where players have a specific question about something they’ve been thinking of themselves.

Intelligence (Investigation) is always an active skill. An Intelligence (Investigation) check always serves to provide an answer to a question the player states to the GM. Figuratively speaking, a check is made when a player puts a magnifying glass to something the GM already described. The players still have to think by themselves to select an object for further investigation and ask a specific question about the object. “Can I find traces of poison?”, “Are there any signs of tempering on the metal?”, “Is it possible to estimate the person’s height based on the footprints?” Again, DC 5 and DC 10 checks are generally not worth making a roll and players get the answer simply for thinking to ask about it. Intelligence (Investigation) checks are made when only a trained expert could get an answer. The DC for the roll may be kept hidden from the players if it seems appropriate for the situation, but the check is still made openly. In that case, failure could mean either “it’s not there” or “you can’t see it”.

“I search the room.” It is possible for players to make an Intelligence (Investigation) check to search an area or a specific object, but only one check may be made for each area and object, and only a single discovery may be made that way. If the whole party searches together, it counts as Working Together, and a single check is made by the most skilled character at advantage. It’s best for players to first search an area “manually” by describing what they are looking for and where they are checking specifically, as they will automatically find anything that is hidden in a spot where they thought to look. An Intelligence (Investigation) check at the end of the search has the chance to reveal one more hidden object that their previous searching has missed. This is an application of the paradigm “You can not roll dice to avoid playing the game!” that still keeps the Investigation skill in the game and useful.

Wisdom (Insight) is used to judge an NPCs sincerity and earnestness. Players have to announce a suspicion and make an active skill check. The roll is made in the open by the player, but the DC is always hidden. Both a failed check and a genuinely sincere NPC result in the reply “You don’t sense any duplicity.”. A high roll gives greater confidence that the NPC is actually sincere, while a low roll means a great degree of uncertainty. It is up to the players to decide what to do with that information. Again, group checks can be made by all PCs present at the interaction. More than two people searching only decreases the amount of time it takes, but does not improve the odds of discovering something.

Wisdom (Perception) is usually done passive, but players can declare that they are actively scanning their surroundings for things that stand out or could be a threat. In situations where it matters, this counts as an action for each round. (While a single character making a Wisdom (Perception) check still has the same odds as Passive Perception, groups of characters all watching actively do improve their odds of one of them spotting something well hidden.)

Charisma (Deception) checks are called any time the GM thinks an NPC has reason to be suspicious and not take the PCs at their word immediately. The checks are made against an opposed Wisdom (Insight) check by the NPC.

Charisma (Persuasion) checks are made to win over NPCs who are hesitant about a course of action. It can not convince NPCs to do things that are directly against their own interest. If trying to convince someone of the truthfulness of your claim, the DC depends on how plausible your story is. If trying to persuade someone of a course of action that is in their interest, the DC depends on how great the price for the NPC will be. The NPCs Wisdom or Intelligence bonus might be added to the die roll if it seems appropriate, as smarter NPCs would be more likely to understand the necessity for certain decisions. For tricking NPCs into assuming an action is in their interest even though it is not, Charisma (Deception) checks are made.