Thinking About Dice Systems

Some public domain dice

When I’m coming up with a game, it’s easiest if I start with a dice mechanic that I already understand. It’s a big advantage – there is less to playtest, fewer decisions need to be made, and I can get right to designing the fun parts of the game. This is why I love hacking games so much – the dice system is already there, I just need to improve how its used. I thought I would share some of my reflections on various dice systems. Maybe it’ll save you some time in designing your own game. Some familiarity with the games listed will help – I’m not going to explain them all in any detail.

And, of course, you can always design your own new genius dice system. But in case unique dice aren’t important to you…

Fate Dice

I love Fate. Fate dice are great because the scale of success is the same as the scale of ability. That is, you know +3 means your character is Good at something, and you know that +3 is a Good result. This is very simple and direct, compared to a target number system like D&D for example. Is a +3 good in 5E? Not sure. Kind of? Not if you’re trying to roll against a DC of 20. I watch DMs struggle to interpret middling results, partly for this reason. The player rolls an 11 to do something. It feels too high for a failure, but definitely isn’t a great result. Fate dice simplify and clarify this problem completely. I have a pile of hacks for Fate Core for a reason.

Kids On Bikes Dice

Kids On Bikes uses a single die against a moving target number. Your die might be a d4 up to a d20. This is simple, and the die clearly shows how good you are at what you are doing, but also introduces some problems. For one, with a target number of 5+, a character with a d4 cannot succeed. With moving target numbers, it is hard to hold in your mind how likely you are to succeed. The probability isn’t intuitive. So you get a very simple system, one die at a time, but you need to think through how a moving target number affects those rolls.

Call of Cthulhu Dice

Call of Cthulhu is a venerable and popular system. One reason, I think, is that it uses a percentile system. A huge benefit to a percentile system is that the chance of success is always obvious. If you have a 50% chance or a 75% chance, there you go. Over time CoC has added degrees of success, which is a big benefit for a percentile system that is usual just a binary succeed/fail result. These degrees of success require a little bit of math, figuring out what half and one fifth of 73 is, but I think it’s worthwhile in-game. You could also just eyeball it – if you make the roll by a wider margin, you get more of a result, and the opposite for failing by a larger margin. A percentile system is also great for incremental improvement. Going from a +3 to +4 is a 33% increase, but you can increase percentile skills bit by bit and still feel like you’re making progress while not unbalancing the system.

Blades in the Dark Dice

I have yet to play Blades in the Dark, but having read through the book it seems that the dice system is designed to function in the 1-4 d6s range. I like the dice system, but it doesn’t have much room for variety – similar to PbtA’s -2 to +3 range. In terms of probability, though, there is more difference between 2 and 3 BitD dice than between +2 and +3 in PbtA, I think. Anyway, BitD is a good option if you don’t want much granularity. If you have a bad/good/better/best type system for ranking abilities, this dice system would work well.

Genesys Dice

I love the idea behind Genesys dice (or Edge of the Empire dice). In practice, interpreting a dice roll slows down game play drastically, and the system ends up presenting information you don’t necessarily need or that is hard to interpret in the moment. On the other hand, the system provides a lot that other dice systems don’t, and I like that. I like that your roll result can be a combination of failure but advantage, or a combination of success but growing threat. In my opinion, though, the system requires an online dice roller, or the app from Fantasy Flight, or else a lot of table time is spent interpreting your dice-roll, and that’s time spent without anything happening in the fiction. That being said, you get much more from a single dice-roll in Genesys than you do from any of these other dice systems I mention.

D&D Dice

OK grandpa, here we go. D20 + bonus vs target number is what like 95% of tabletop RPG players are doing when they roll dice. But wait, there’s more! They also roll single dice, and pools of dice, for things like damage and spell effects and recovery. They roll percentile dice for random tables and 4d6 for character creation and various dice for hit points. Part of the fun of D&D is that you get to break out and use all of your dice. The downside is that everything is an exception. You can only sometimes guess how a mechanic will work – compared to, say, Fate or PbtA. Tools like DND Beyond help a lot, but the only reason that designers are trying to hack 5E for other games and genres is its ubiquity. Designing new things for D20 style games is a nightmare compared to other systems.

Old School/OSR Dice

Is that you great-grandpa? D&D’s profusion of different dice mechanics has deep roots, and the oldest editions of D&D sometimes read like different people designed each dice mechanic without talking to each other. This was exacerbated by not always referring to dice in the core texts – you’d read something like “3-17 appearing” and have to think for a moment to realize that must be 2d8+1 and not 3d6-1. Some abilities are a percentile, some are based on a d6, and some use a d20 to roll high while others use a d20 to roll low.

World of Darkness Dice

It’s known that the World of Darkness took its dice system largely from Shadowrun, but they improved it by removing exploding dice. Dice pool + exploding dice with d6s is just…it’s madness.There is a lot to like about the world of Darkness dice system. You get to roll big handfuls of dice when you are good at something – its rewarding in a tactile way. There are also two sliding scales – number of dice, and target number you want on each die. The problem there is that probability quickly gets away from you. Is it better to roll 7 dice at difficulty 7, or 8 dice at difficulty 8? In the moment, who the heck knows? Later editions of the World of Darkness addressed this by keeping a fixed target number, but I still like the sliding target number as an additional variable. You just need to be careful about what kinds of things can change the target number compared to what adds or subtracts dice.

Powered by the Apocalypse Dice

Even I have tried my hand at a few PbtA hacks – the system is very easy to design for because everything is an exception. If you want a mechanic for something, write a move. If not, then just describe what happens. The dice system is very simple and straightforward. There is no real mechanic for making something more or less difficult (without writing multiple moves for the same thing at varying difficulty) but the system is tuned to aim toward failing slightly less than 1/2 the time and most often getting a marginal success. The key here is to make sure that failure is interesting, since it will come up often, especially with a -1 or -2 ability. Games like Dungeon World do this by letting the player mark experience when they fail, which is often enough to at least dull the sting. Most PbtA games also let the GM escalate the situation on a ‘miss’ of 6 or less, a kind of “no, and” style failure that is more interesting than a simple “no.”

Savage Worlds Dice

Savage Worlds never uses all of the usual dice except the D20, and the target number is always the same – a 4. The addition of a coin-flip “wild die” d6 to every PC roll also helps make success much more likely, which fits the fast-and-fun pulp style of SW. (“coin flip” meaning you have a 50% chance of success on a d6)

Mouse Guard/Torchbearer Dice

I love both Mouse Guard and Torchbearer, but the dice for both are basically a coin toss. You roll d6s against a target number of 4 or higher and count successes. Occasionally you can actually spend a resource to re-roll the dice that come up 6s, so there’s a little bit of the fiddling around that I enjoy, but I’d rather a little more variety than a 50/50 chance on every die. That being said, it’s simple, easy to remember, uses dice you already have, and I’ve had fun playing both games.

Parsec Dice

I know, probably none of you have heard of Parsec. It’s the game I published in 2012 through Jolly Roger Games. It uses d6s where the number of dice rolled are based on your attribute plus/minus modifiers and the target number for those dice is based on your skill. The advantage is that what you roll is set and clear without having to ask the GM each time. If you roll 5 dice with a target number of 3 or higher, then you can shorthand that as 5@3. Not world-shattering, but I’m still proud of the game, especially as my first published game.

Summary

When choosing a dice system, considerations include:

• How long does it take to interpret what the dice say?
• How often is success likely, and does that fit with your genre and expectations for play?
• Is it clear to the player how likely success is with a given roll?
• What is the tactile feeling of rolling? Are you rolling too many dice? Too few? Like ‘mouth feel’ for wine, what is the intended ‘table feel’ for your dice system? Everything hinges on a single die? Or heaps of dice skittering everywhere, to be collected and counted up?
• Do you need special or unusual dice, or can you just grab what’s in a board game box and play?
• Is there a way to get the same information with fewer, or simpler, dice?
• What does the handful of dice mean? Is it the character’s total ability (World of Darkness)? Is part of it the difficulty they face (Genesys)? Is it all of the advantages they have (D&D advantage, bless, bardic inspiration)?
• Are the dice clear without asking the GM? For example,
• have to ask for a DC for your d20 roll, but rolling damage dice is obvious as you’re just adding them up.

What are your favorite dice? What is your favorite dice system?

CounterSocial

D&D Without Attack Rolls or Saving Throws

D&D sucks when your turn comes around and nothing happens, either because you whiff on your attack roll or you cast a spell and miss your attack or your target makes their saving throw. Any game sucks when your turn amounts to nothing, but especially a game like D&D where you might have waited 30 minutes for your turn to come back around again in a complex combat encounter.

It’s also tough when you build a spellcaster like an enchanter, or someone who has a lot of spells that allow saving throws, as sometimes you’ll use your magic and nothing will happen. Unlike an attack roll, you’re already using a limited resource – spell slots – and getting nothing for it.

I was also thinking about how to adapt some of what I was enjoying so much about Breath of the Wild to tabletop, and one thing about Breath of the Wild, and most video games in general, is there are no saving throws that negate your abilities, and normally instead of a character skill roll to attack you are relying on player skill. The player skill is harder to attach to a tabletop RPG, and depending on how you do it you get accessibility issues. But that’s a line of thinking for another time.

As a rule, there shouldn’t be a point in a story when a main character takes action and nothing changes. That’s true for PCs in RPGs – when they act, something should change. What prevents that change in D&D are attack rolls and saving throws, so I wondered, how much would I have to change D&D to get rid of those two things?

Turns out, a lot, but not as much as I’d feared.

What follows are my notes so far. Feel free to use them as a starting-point. I need to playtest this idea, but I do think it’s workable. As with the other things I’ve uploaded here in the past, I’ll update this document as I improve on my notes. Enjoy!

dd-5e-without-attack-rolls-and-saves-0.5Download

Saga of Recluse RPG Ideas

I’ve been testing some of the books that I enjoyed as an adolescent, and I’ve gotten to re-reading some of L.E. Modesitt’s Saga of Recluse. I read a few forum comments by the author himself, though from 17 years ago, about how he has approached more than one RPG publishing company about a Recluse game, but that they gave up at some point because of the complexity and difficulty. Now, I can see this happening, talking to an author about their 19-novel series that spans 1900 years of history, but I don’t think an encyclopedic style game would make sense for recluse. I think a Recluse game would have to mirror the kinds of stories that the novels are – first-person narratives from the point of view of a nascent order-master or, sometimes, chaos-master, and then more broadly about how the repercussions of actions ripple through time.

So, what would a Recluse RPG need?

• A system for chaos and order magic, including the interactions between the two, the balance in a given place or person, etc.
  • Also how chaos and order magic can be used to augment or even replicate skills
• Tools for running single-player RPG sessions. There aren’t a ton of those out there, but I think there is a lot of potential here. Maybe you and your spouse want to game. Maybe you only have one friend who wants to play a RPG. I ran single-player games for my best friend for a few years in Middle and High School.
• A system for playing out the repercussions of the character’s actions over time, affecting other stories
  • Maybe even taking Modesitt’s approach and placing the stories at different points along the timeline, with later events asking the question – how did things get this way. Something Microscope-ish
• This could also be a GM-less/GM-full game, with one protagonist and then the surrounding characters in each scene played by the other players at the table. Maybe one handles order-chaos interactions, and one handles history and culture and call-backs to other events, one handles opposition, etc. Your standard GM-less/GM-full fare

There, I Fixed It: The Wish Spell

Updated and re-released for your reading pleasure!

Something that the System Mastery guys love to harp on, all the way back to their very first episode: Dungeons & Dragons’ wish spell (and similar spells in the wish tradition from other RPGs as well). As written, wish spells, or wishes in general in TRPGs, are almost always explicitly ways to disrupt players’ expectations and, in a word, screw them. GMs and DMs are often encouraged to find any possible loophole, any interpretation in the player-character’s wish that might justify screwing with them.

In 5th Edition and 3.5 as well, other than that, a wish spell is for the most part just a catch-all for replicating an 8th level spell. There is otherwise a list of possible effects that are clearly defined and limited in scope. Part of the problem is that wishes in the folkloric sense should not be spells – the simple solution here is to excise wish from the list of arcane spells entirely. But if you want to keep it, or if your game is going to feature a significant number of genies, then there must be something better than punishing players with it. (If you want to punish a character, hand them a Deck of Many Things and stand back).

The potential problems with wishes should be obvious, and there are plenty of folkloric stories about well-intentioned wishes going wrong, or at the very least not having the effect that the wisher intended. On the other hand, these problems are usually ways of moving the story forward so that the protagonist can learn something or change in some way. All too often in TRPGs, wishes are simply opportunities for the DM to punish a player for trying to be creative, when it’s the DM’s decision whether to allow wishes in the first place. For those DMs whose players are not masochists, I have some other thoughts.

The first is that a wish should be fun. Here I’m thinking of Aladdin’s first (official) wish in the Disney animated adaptation of his story regarding a certain lamp. He basically gets what he wishes for, and if anything, Genie goes overboard (as Robin Williams invariably did) in embellishing the whole scene. Rather than being a stingy saboteur, one pictures Aladdin’s DM just throwing cool things at the player-character until the player’s head spins. There are complication, of course, as “Prince Ali” draws the attention of a sinister visier and is suddenly plunged into court life having been a fruit-stealing street kid not long ago, but the story moves forward with the wish fulfilled at face value, plus interest.

Wishes should be fun. D&D should be fun. It should never be a DM power trip, or about ‘punishing’ players.

Second, a wish should indeed have a cost or an unforeseen complication, but this cost or complication should be something that is part of the story moving forward and continuing to be fun. The street rat suddenly lifted to Princedom has no actual idea how to be a Prince. No history, no family, no connections, no homeland, nothing. And as mentioned, he draws the attention of the sinister vizier. I would even recommend discussing possible complications with the player who is making the wish. I know this is not everyone’s play style, but in my experience this doesn’t diminish the fun – you kind of trade surprise for a higher guarantee that you’ll all enjoy the twist.

Third, a wish should take context into account. I still think that DMs should just eliminate wish from all spell lists where it might appear, and keep wishes as a story element. Obvious options are powerful fey or genies whom the PCs have worked to befriend. Maybe the goal of a whole campaign could be to earn a wish from a powerful entity, and then to use that wish to restore the kingdom, or end a curse, or cure a plague. But remember that the wish is interpreted in context. If a PC makes a wish granted by the genie, that genie will interpret the wish, and a wish granted by an ifrit will be very different from one granted by a marid, or a djinni. Rather than a chance to punish players, this is a chance for a DM to show off her creativity. To use this example again, a wish granted by a genie voiced by Robin Williams will be one thing – one granted by a stingy cantankerous fey quite another.

Remember that a wish’s fulfillment does not need to be immediate (unless maybe the PC adds that to the request – in which case, it could rain gold pieces or cause other upheaval). Feel free to take a moment in game when the wish is finally made (which again should be a huge story moment) to go think through what it will look like when it is fulfilled.

Discourage players from gaming the wish. A player might be tempted to go off and write out a page-long run-on sentence as her wish, full of legalese and dependent clauses. Depict the wish-granter getting bored and starting to wander off. Understandably, players will anticipate the DM trying to twist their wish against them, and will try to avoid that eventuality. Maybe reassure them, if necessary, that this is a big story moment and you’re not going to sabotage it.

Possible house rule: total the words in the wish, and that number becomes a percentage chance of failure for the wish. So if you say “Make me a prince!” Then there is a 4% chance of failure, but if you write out a mini contract rife with legalese then it could easily become a coin-toss.

So, to summarize the wish spell – don’t make it a spell at all. Make it a story element. Make it fun. Have a cost or unforeseen complication, but make it one that moves the story forward in an interesting way. Take the context of the wish, and the wish-granter, into account. And push the players not to lawyer the wish, even if you just have to reassure them.

Another Solution

After I posted this, I came up with another interesting idea for wish-fulfillment. In this version, you make your wish, and then genie is bound to do whatever they can to fulfill that wish, using their own abilities. So if a dao gives you your wish, and you wish for a million gold pieces, then that dao has to do all it can to get you a million gold pieces. They don’t have the power to just wave their hand and fulfill the wish, but they will interpret it according to their alignment and their capabilities. Probably, in this case, disguise themselves, break into a vault, and abscond with their gold pieces.

The way that a genie fulfills your wish will vary from genie to genie. A marid, being chaotic neutral, will be quite different from the dao mentioned above, who is neutral evil.

This situation could be handled as if the genie was under the effect of a powerful geas.  They would have 30 days to accomplish your wish by any means they could, or to work toward it as far as they can by the end of the time limit. As an added bit of interest, and the end of the time-limit, the charm effect ends and the genie might have a chip on their shoulder.

Personally, I like this idea, and want to try it for my world of Alaam.

Do you have any stories of wishes going well, or poorly, in your campaigns? If so, share in the comments. 

RPGs = Six Situations

I was thinking about the practice and experience of playing a TRPG consisting of about six situations, and how you could look at the challenge of designing a game as having something interesting for those six situations. I’m using tropes from fantasy RPGs here, but I think it would be easy to reskin these situations to include different tropes.

In Town

Town, or the city, or whatever your home base is. Time spent in town is time finding a way to rest and recuperate. If you are going on a shopping spree, it’ll happen here.

I like the way that Torchbearer makes resting a challenge. It is not easy, nor assumed, that you’ll find a safe place to rest. You might end up hiding in a stable or sleeping in an alleyway. I also like the way that The One Ring and Adventures in Middle-Earth require you to open a Sanctuary before you can fully rest and recover there, often demanding a quest, or at least a successful audience with the ruler of that Sanctuary.

Socializing

It’s always hard to add mechanical teeth to socializing. There’s this idea that you should speak and interact in character, and as you do so it’s hard to know when to roll the dice and when not to. What if you make a great argument, or come up with a killer lie, but then botch the roll? Or what if you make an absurd ask and then critically succeed? This kind of silliness can just be the fun of using a randomizer, but I watch groups struggle and disagree on where to draw the line here. Surely it’s because socializing is something we literally act out at the table, in contrast to exploration or combat. We never ask for anyone to test their weapon skills, but we do ask them to test their social skills from time to time.

Some groups, of course, don’t socialize much at all. You get the mission briefing, and then head to the entrance to the dungeon and kick in the door.

Traveling

A lot of time in classic fantasy and sci-fi stories is often spent traveling. There is the canard of The Lord of the Rings being mostly just people walking and looking at trees, but even in something like Star Trek you spend a lot of time figuring out what you do while watching stars zoom past.

One option is always to just hand-wave the travel and get to the next interesting thing. As a lifelong road trip connoisseur, however, the journey really is about more than just the destination. Again, I think of The One Ring, and to a slightly lesser degree Adventures in Middle-Earth, as well as Mouse Guard as games that focus on the journey itself and provide mechanics to make it an interesting challenge.

In terms of verisimilitude, when I think about traveling hundreds of miles through a fantasy landscape on foot, that would absolutely be a noteworthy life experience. Lots of challenges would arise and lots of interesting things would happen, not even counting the monsters and random encounters. I would like to have mechanics to support this.

In Camp

Camp differs from being in town, because it is a time with particular traditions like setting watches and rolling for night encounters. I was thinking of having a camp checklist, and the more things you can check off on the list, the more comfortable you are and the better able you are to recover.

• Clean water
• Dry/Shelter
• Fire
• Food

A simple example might be that for each checkmark in D&D, you can roll up to 25% of your hit dice. So with clean water and shelter but no food or fire,  you can only roll 50% of your hit dice to recover. That’s not prefect, but is a decent example. Maybe you just recover 25% of your hit points and other expendables per check-mark when you camp.

Investigating Danger

Searching a crime scene, checking for traps, or exploring an ancient tomb all count, and have been central to TRPGs from the beginning. Some OSR folks make the case that original versions of D&D were more about exploration than combat. Some games do this part really well, like Gumshoe. Like with socializing, groups have a chance to choose whether they want to handle investigation with rolls or with players describing what their characters are looking for. I have a whole blog post about how you shouldn’t roll perception that you can check out if you want. But whether you are playing Mothership or Pathfinder, investigating dangerous areas and situations is a big part of what is fun about many TRPGs.

Fighting

Most RPGs are mostly about fighting. If you read the rulebook, most of those rule are about combat – usually physically combat, sometimes social conflicts as well. But social conflicts go under this heading as well as fistfights.

I don’t feel like I have to put time into making the case that RPGs focus on fighting, honestly, but see below. They don’t have to.

Rules Modules

So, if we look at each of these six categories of systems in turn, we can also imagine a group preferring to ignore some of them. Maybe your group wants to hand-wave their way to their destination, or maybe they want to just camp and rest and not worry about safety and comfort. Maybe they will handle time in town between games, just buying things from a price list and getting straight to the adventure when they get together to play. There could even be a group that wants to skip the combats (blasphemy!). If each of these systems is built like a module that can be used or ignored, I like that idea. You can socket in what interests you and get on with playing only the things that interest you.