This is part two of the post giving an overview of my new approach to my RPG project, A Wilder World. This post will deal with magic.

I’m going to recap how magic was conceived in the previous version of the game, since not a lot has changed but a lot of different people are reading my blog. The big points are this:

While a given working of magic might be thought of as a spell, A Wilder World does not have a spell system as such. Magic is treated as a force that wizards and other magic-users can control rather than a list of menu selections to choose from. Being a pyromancer (for instance) is like having another set of muscles which, when flexed, make fire do things.

Magic comes in three sizes: trivial, ordinary, and extraordinary.

Something that could very easily be accomplished without magic or that has little lasting impact on the world is trivial magic. Stirring your drink with a lazy wave of your hand, fetching a book off a shelf that a person could have gone and picked up, making a harmless shower of sparks to announce “Why yes, I am a wizard.”… these things are trivial magic. Merely having magic allows you to do these things, which are similar in scope and intent to D&D’s “cantrips”. They are the special effects and roleplaying flourishes of being a wizard. No roll necessary, no cost normally.

Anything that would have required real effort and a die roll to accomplish by other means is ordinary magic. Fetching a key from a peg on the wall outside the locked cell you’re in, using a book to hurl itself at someone’s head or using a shower sparks to injure, blind, or distract would all be ordinary magic. Ordinary magic always requires a roll, a roll that is noticeably more difficult than it would have been to do the same thing by mundane means.

“The same thing by mundane means” does not refer to the summoning the key or creating a shower of sparks, but the end result of the magic. If you try to circumvent a locked door, whether by magically causing the lock to open or disintegrating the door or blasting it to cinders or teleporting just past it or turning into mist and pouring through the keyhole, the net result is still that you have defeated the obstacle posed by the door.

Extraordinary magic is the stuff that is clearly impossible to achieve without magic: traversing miles in the blink of an eye, opening or closing a portal to another realm, et cetera. Extraordinary magic is bounded by laws not subject to mortal knowledge or approval; e.g., it’s either built into the story or approved in a desperate moment by Storyteller fiat. The daring wizard who manages a long-range teleportation feat in AWW has not added a new trick to their repertoire, but has more witnessed a once-in-a-lifetime occurrence that few mortals who traffic with magic will even see.

The categories of trivial and ordinary magic can be expanded by having the right sets of abilities; e.g., talking to the dead should by definition be extraordinary magic, but it may be trivial for necromancers, and there may be wizards or magical beings out there who specialize in site-to-site transportation. These abilities do sketch out the limits of what is trivially and ordinarily possible for those “flavors” of magic.

In the previous version of AWW, the use of magic was governed by an attribute called Magic that would basically “sub in” for any check attempted through magic. You just had to explain what you were doing and how. The “flavors of magic” abilities would define specific bonuses for magic uses that fell within their purview, with instructions that Storytellers be flexible and generous, but reasonably firm about things like “I make a key out of fire and use it to pick the lock” or “I make a fiery hand reach out to pick up the book.” It was also possible to make a character who could only use magic when that fell within their designated “flavors”.

In addition to the checks being harder, I originally intended for Magic to cost more than other attributes (because it can stand in for any of them) and using magic to accomplish something was assumed to be slow, risky, and obvious (loud/bright) compared to the mundane approach.

So here’s what has changed:

First, there is no Magic attribute equivalent among the abilities you can pick. Abilities that allow you to use magic in the fashion described above fall under the heading of “Wizardry” (to distinguish this kind of ad-hoc spellcasting from other, more specific abilities that might also be magical), and each flavor of Wizardry is its own thing: Necromancy, Illusion, Green (plant) Magic, et cetera.

Now, you might be thinking this means there’s no non-specialist wizards in the game, especially if you weren’t following the previous development cycle. Some flavors, though, are more about the method of casting spells than the way the magic manifests: ritual, song, et cetera.

Each level of a Wizardry form you have gives you one die to roll when making checks using it. You don’t start out with one die as you do with normal checks, because an ordinary person has no recourse to magic at all. It’s all extra.

You can get more dice by combining a method of Wizardry (Circle Magic, Ritual Magic, Song Magic) with an aspect of Wizardry (Illusion, Necromancy, Pyromancy) if you decide when you take them that they are bound together, meaning you always have to use them together. You can have multiple such bindings affecting the same ability; only one needs to be satisfied (and can be taken advantage of) for a given casting. So a bard seeking power and flexiblity could have multiple aspects of Wizardry bound to Song Magic; the level(s) of Song Magic would increase the power of whatever the bard does, at the cost of their magic always having a conspicuous musical component.

Under the system of archetypes-as-ability-packages that the previous version used, each flavor of magic came with a set of side abilities that in some cases where what players really wanted; e.g., people looking to play a Batman-style adventurer/gadgeteer who always has the right item tucked away taking Conjuration for the ability to get new gear on the fly. These side abilities are now separate abilities, so they can be taken without being tied to the Wizardry system and properly defined as non-magical if they’re meant to be a non-supernatural innate ability or a “meta” thing (in the sense of storytelling, rather than superpowers).

Now, the other change is more of a refinement. Magic before was risky in the sense that it had an immediate failure rule that didn’t apply to most other checks (as AWW allows you to do a preliminary check as a sort of “cautious consideration” before you try things like jumping over a canyon) and limited in the sense that failure could wound you and/or accumulate Magic Burn, which would impair future magic use. It effectively “burned out” a portion of the character’s magical ability.

But the Magic Burn mechanism, if it were a meaningful limitation, would create the sort of curve for magic where it would tend to get less useful as it was used over the course of adventure, meaning that wizards would either tend to hoard their magic use until the end, or they would be more dramatically useful at the outset, when things are easier anyway.

And what we might call the “committed risk factor” would sharply limit the willingness of people to try wizardry in some of the cases where it would be most exciting and high-flying, the things that maybe nobody else in the party was prepared to deal with.

So I’ve re-imagined the limiting factor of magic into a less one-size-fits-all form.

Like the rest of Wizardry—and indeed the game as a whole—this can require some good-faith negotiation between the player and the Storyteller, but the basic idea is that old saying that magic always has a price. But not all magic is the same, and so the price isn’t always the same.

When you make a character who has Wizardry or similar magical abilities, you basically choose a payment plan. Available choices include (but may not be limited to):

• Control: Choosing this as your limit means that your magic always works, it just doesn’t always do what you intended. If you fail a check using Wizardry, the Storyteller rolls a die and decides what happens, ranging from random happy coincidence on a 6 (not what you wanted, but something beneficial) to something fairly benign on a 4 or 5, and then increasing levels of bad turn for 3, 2, or 1. As usual, the dice and rules don’t tell the story, just provide a general result. A Pyromancer with control issues is likely to have wild gouts of flame, where a summoner of some sort is likely to have the spirits/creatures show up and simply not respond to their commands. As an additional limiting factor, there is a small but cumulative chance each time you succeed on a magic check that there will also be a side effect.
• Balance: If you choose this as your limit, then the universe or some force within it exacts a price in the form of bad luck every time you use your magic. It might be something so small you don’t notice it, it might happen right away or somewhere down the line, but every successful magic use results in a roll similar to the failure cost of Control. If a drawback or disaster is indicated, this is not always a direct, immediate side effect of the magic use (unless the Storyteller sees an opportunity for some creative irony), and in particular it’s not directly related to the magical energy/effect.
• Favor: Favor means that your character, rather than controlling magic directly, is a thaumaturge; you are in contact with some higher (or lower, or other) being or beings who will do magic for you. Making this your cost, though, effectively makes each use of magic a round of negotiation between you and the force you work with. It might be an easy, routine negotiation if you’re doing the being’s bidding, but you might have to pay an actual agreed-upon price or agree to a contract or code of conduct to be able to use magic for your own purposes.
• Self: Each time you use Wizardry, you risk taking on an increasingly more inhuman (or inelven, or whatever) aspect, usually in a way defined when you create your character (like becoming increasingly demonic), though it can also be related to the form your magic takes with each specific application. This not only affects how people react to you, but how you react to people and things, again in a way defined on a per-character basis; a pyromancer might become increasingly obsessed with creating fire for its own sake, a diabolist might become increasingly venal and selfish, a druid-like character might become increasingly animalistic or withdrawn from the world. Skilled roleplayers can play out this limitation all on their own, but there is a mechanical aspect for those who need more guidance; essentially, once you reach a certain tipping point, the Storyteller can roll a die when you try to do something 1) helpful and 2) not involving more of your magic use and tell you, “You can’t bring yourself to do that.” Additionally, as you accumulate lack of self, you start to run into a version of the Control issue described above, except it only strikes when your magic works… so you can succeed on a check, but fail to do what you set out to do because you were too distracted by how awesome your power is, or how pretty the fire is, or whatever. Lost “self” is regained over time.
• Blood: If you want to do magic, first someone has to bleed (or the equivalent for their biological makeup). It can be you, a willing ally, or an enemy (provided you have the ability to hurt them). This is one of the most predictable and easiest to control prices, but also one of the grimmest and most implacable. The ones above it basically give you the chance to get off for free, where the blood price is always paid. But you also know what it is.
• Life: Life works similarly to blood, but it’s purely internal. You are always the one who pays the price, but on the plus side, you don’t have to be able to cut someone.
• Effort: Your magic is “free”, but it takes intensive effort (mental or physical, your choice) to pull off. You are never considered to be at rest when doing magic. This also translates to magic being more time-consuming; any kind of instant magic will be mentally or physically exhausting. This is common for ritual mages who are already not counting on quick turnaround for mystical labor.

Any of the costs can be mitigated somewhat by spending more time on an individual magical working: the more time you put into a spell, the less likely it is to run wild or take a serious toll on you; the more time you spend flattering or praising your otherworldly contact, the more likely they are to dispense a supernatural gift; the longer you spend drawing out the blood, the more mystical energy you coax from it with the least pain, et cetera.

Trivial magic is still trivial, and so its price; a mere pinprick, a short prayer or word of flattery, random butterfly fluttering level occurrences, et cetera. You can also take levels of an ability called Arcane Reserves to give you “freebie” uses of ordinary magic in a day without the price/risk.

Some magical abilities that aren’t Wizardry also fall under the cost mechanic. Using these abilities and using ordinary magic are treated as equivalent for purposes of incremental effects.

Note that the assumption is that all your magic comes from one source and thus has the same price, but you can define a character in a more complicated fashion. A character might have most of their magic bound up in Favor but one particular “gift” that comes with a cost of self, for instance.

…

So, the short version is that there are no discrete “spells” to be learned in A Wilder World, but you instead define your character’s magical abilities using a combination of archetypal methods and aspects, and the price you pay when you use your magic. The goal here is to create a system that is flexible but limited, with magic-users whose abilities more closely map the way these things play out in stories and folklore than in typical “gamey” systems.

This is Part I of a three part post delving into my thoughts on a new iteration of A Wilder World. Parts II and III will deal with magic and combat. This part deals with the basics of the die resolution system and character creation.

First, the attribute system is basically just… out. Having twenty different axes of numerical customization right alongside the “take two or three concepts and smush them together” character creation thing, it was like having two pretty strong but unrelated bases of character generation in the same game.

While I liked the idea that you could make your Acrobat/Assassin more or less acrobatic than someone else with the same build and then choose how to spend whatever points were left over, it also meant you could make an Acrobat/Assassin who utterly failed as an Acrobat/Assassin. You can’t stop players from making borked characters if they want to, but in a “plug and play” character creation system you shouldn’t need a guide to tell you how to spend your points.

Without attributes, all we care about is notable abilities, notable differences from the norm. To use an example from D&D spin-off novels, we’ll take the Icewind Dale characters by R.A. Salvatore. Wulfgar the human barbarian has what we might call legendary strength, especially when fueled by his rage. At one point he arm-wrestled a balor (which is a balrog with the serial numbers filed off), and at another juncture he took the edges of a dimensional portal and bent them shut. I mean, it was a magical device with an actual edge so he wasn’t literally grabbing space and time and folding them, but still.

So, legendary strength.

His adoptive father, Bruenor, is a dwarven warrior, so we’ll say he has notable strength. We would expect him to win a tug-of-war with the average human more often than not, but lose to Wulfgar.

Bruenor’s other adoptive child, Catti-Brie, is a human swordswoman and archer. As a reasonably fit hero, maybe she’s a bit stronger than the average adult human even in a time and place when the average adult human is a laborer, but her physical strength is not the thing that people remember her for. If she’s the one trying to shift a heavy rock, it’s only because neither Bruenor nor Wulfgar is around and her magical piledriver bow and magical stonecutting sword are both missing… all of which is to say that it’s not terribly important for us to know exactly how strong she is because she doesn’t solve problems with brute strength. Her solutions are often Gordian in nature, but she doesn’t need as much raw strength as either her foster brother or her father to back them up.

Their friend Regis, the pampered halfling confidence artist, deals with heavy rocks by persuading very strong people to move them for him. In a team of what is basically fantasy superheroes, the fact that he’s there at all is an indication of the fact that his superpower is charm: he is friends with all the most powerful people. It is probably worth noting that strength is even more affirmatively not his area than Catti-Brie, but having noted that, it is not necessary to quantify in exacting detail exactly how not strong he is. We just know he’s not strong.

Now, we could use words to describe these levels of ability, and back around maybe January or so I was toying with a color-coded system, where average would be green, yellow and red would be one and two levels below, and then we’d use the shiny metal colors for higher levels of ability: like copper, silver, gold, platinum, mithril.

That’s a lot of levels for a system that by design doesn’t need or want to be very granular, and it also has the disadvantage of not being visually intuitive. There would be a period in the learning curve where you’d still have to look up what “copper” means, and there’s no reason to suspect right off the bat that “copper” beats “green”.

So let’s stick with numbers. We’ll call where Catti-Brie is 0. She’s probably a bit stronger than average, but not so much that the game cares to mark it, and it’s not her particular area of expertise. In most situations, she’s going to have better things to do than apply elbow grease, and allies who are better suited to those problems where only raw muscle will suffice.

Bruenor we’ll say is +1. Wulfgar we’ll call +2, with another +1 when he’s enraged. Regis is -1.

This gives us a way of referring to differing levels of ability in a way that the difference is obvious at a glance, but we also need a way of meaningfully translating it into game terms. There isn’t really a combination of die rolls where straight up adding +2 for Wulfgar’s Strength as compared to -1 for Regis’s would really capture the difference between a relatively sedentary halfling and a legendary human warrior.

So let’s borrow a page from D&D 5E’s rather elegant replacement for a lot of “stack ’em to the heavens” modifiers, their concept of advantage/disadvantage: rolling extra dice and then taking the better roll or, in the case of disadvantage, the worse.

We’ll stick with the conceit of only using 6 sided dice for A Wilder World, for reasons of simplicity, predictability, and accessibility. If we define checks or rolls by a Difficulty number that is the minimum number needed to succeed, we can have tasks defined from Difficulty 2 to 6, with odds of success ranging from 87.5% to 12.5% for the average character.

If you have a score of -1, you have to roll twice and use the worse result. If your score is positive, you can roll extra dice and use the highest one. Each additional die halves your chances of failure, without ever dropping the odds to 0.

So say that a non-raging Wulfgar is trying to kick in a wooden door. We’ll define this as a Difficulty 5 task; the average person would have a 1/3 chance of managing it. He rolls two extra dice, and if even one of them is 5 or higher he succeeds. That’s pretty close to a 70% success rate, compared to Catti-Brie’s 33% success rate. Note that I’m doing the math here, but it’s not at all necessary for players to do the math, just understand that more pluses = dice = more chances to succeed. You can roll your three dice and glance at them and if anything is 4 or higher, you know you made it.

Meanwhile, Regis has an 11% success rate. Which might seem high, and some people are doing the math in their head to figure out how many tries it would take him before his success was basically guaranteed… but the thing is, AWW goes with the philosophy that a check is not “Does it work this time?” but “Does this thing I’m trying to do work?” The random factor abstracts away all the different innumerable variables that would be in play in real life. So it’s not that Regis can try to knock down the door nine times and he’s bound to succeed once. It’s that 11% of the time, when Regis encounters a door like this, he’s able to take it down.

It still might seem odd that any door that Wulfgar can take down is theoretically within Regis’s reach, which is why there’s another wrinkle to difficulty levels. For now, let’s call it magnitude. It’s the “you must be at least this awesome to ride” filter on doing stuff. Magnitude is noted as a minus after the difficulty. Imagine a reinforced security door that is Difficulty 5-1.

What’s the minus do? It shifts your effective score, and if you’re left with lower than -1, you can’t even try.

So when faced with the reinforced door, non-raging Wulfgar has Strength +1, Bruenor has Strength 0, Catti-Brie has -1, and Regis has -2, meaning he’s out of the running.

There is a thing called Heroic Effort where you get to use your whole dice pool for check you’re awesome enough to attempt, but it’s a resource-tracking thing. It’s not too limited, because part of the idea here is that each level of ability is a whole order of awesomeness above the next plus down, but it’s also not automatic.

Now, I’ve been talking about Strength as a concept here, which seems like a stand-in for an attribute even though I’m saying that the system is attribute-less. This is why I also used Rage as an example of something that could conditionally stack with the Strength bonus, to ease into the next concept I’m bringing up.

As much as I like the smushing-together-two-archetypes system… it really contributed a lot to the problems I described in my last post. It’s a great idea, for something that functions more like D&D, but with a focus on gonzo character customization—kind of like D&D 4E’s Gamma World spin-off, or D&D 4E’s hybrid system—and with D&D 5E style “grow into your complexity” system.

It’s not great for a system like A Wilder World where part of the idea is that characters can be relatively “feature complete” at level one, or where the idea is to keep conflict resolution/risk management on a quick and easy basis.

So here’s my new basic concept for character creation: you make a list of stuff your character has going for them. Each of the things you list are considered Abilities. Most abilities are 1 Character Point, though some might wind up being 2 or 3, and some (many) of them will be available at different levels; Strength +1 is one point, Strength +2 is two points. The Storyteller can cap how high an ability can be at character generation depending on the intended power level of the campaign, and also sets how many points are available.

Me, with my taste for high-flying action-adventure at level one, a sort of cinematic fantasy feel, I’d go with something like 10 or 12 points, but it’s possible to play with a pool of 4 or 5 points.

A secondary list of the same length follows the first. These are things called Perks, which are smaller than Abilities. Knowing a language is a Perk, while being a skilled linguist who can find a basis for communication in any language is an Ability. Possessing a musical instrument and knowing how to play it is a Perk, while having an alchemist’s kit and knowing how to use it is an Ability. Having a small side weapon or light armor is a Perk, while having a full-sized weapon or metal armor is an Ability.

While Abilities and Perks can be roughly sub-divided into different categories like gear, followers, skills, advantages (advantages specifically describing things most people can do naturally but you can do better, i.e., things that would otherwise be attributes), there are only these two steps in character creation: spend points, pick perks.

The Abilities would be a lot more roughly sketched than any special abilities listed in the previous iteration. Many of them—particularly the pseudo-attribute Advantages—will basically just consist of describing the areas of endeavor they cover, as well as some rough guidelines for figuring out the fringe benefits (like being strong = can carry heavier burdens for longer). Some of them will have specified mechanical effects, including bonus perks. Points in Linguist = extra languages, for instance.

The archetypes I spent so much time lovingly devising will be included as basically packages of suggested abilities: these are good things for merchants, these are good things for nobles, these are good things for assassins, et cetera. While this makes the plug and play element a little more involved, I feel it makes character creation as a whole simpler, and also more flexible and consistent.

Under the previous version, multiple archetypes essentially had “knows special contacts” as a special ability but the handling for each one was different. Same thing with archetypes that had flexible item allotments, or the ability to summon helpers.

Of course, with attributes and archetypes gone, magic is going to work a little bit differently, though the basic structure of it will be familiar to those who read the old system. The basic ideas of “magic comes in different flavors, magic can do anything that could be done without magic plus a few extra things depending on the flavor, and magic pays for its flexibility by being costly, difficult, slow, and loud compared to the mundane way” are all there, though the whole cost of magic thing is being refined in a way that I think will be more interesting.

I don’t know if green tea with mint is my personal miracle elixir for creativity, or if the taste and smell of it is just a memory trigger for certain times in my life, or I am just feeling the euphoric effects of being well-rested after a couple of weeks of being anything, or what, but yesterday afternoon, and then again last night, and then when I woke up this morning, I was having a lot of thoughts about A Wilder World, the roleplaying game system I’ve managed to bring almost to a point of completion multiple times over the years.

I haven’t talked about it in public much lately, but I have had a few private conversations about it, particularly with Shweta Narayan, that helped crystallize what wasn’t gelling in the recent, most close to complete iterations.

As I see it, there are three basic problems with the system I devised.

I: It basically starts with high level play.

The idea of hybrid characters who are “complete” at level 1 instead of having to multiclass/dumpster dive over a series of levels to get the exact combination of abilities you want is important to me. If you want to be a rogue with an animal companion or a wizard who fights with two swords, you shouldn’t have to spend half your adventuring career assembling your schtick and winding up less competent overall than your peers. That’s key. That’s the whole kernel of the idea.

The problem is that my solution gave you characters with something like 12 to 16 distinct special abilities at level one. For a player like me who is experienced with high level play in other systems, this is not a big deal. I’m used to the juggling of multiple resource pools and keeping track of multiple moving parts. I thrive on that. But not all players do, and newer players can be turned away or overwhelmed by that kind of complexity.

Even a lot of the people who were excited by AWW were more excited about it as a character generation system than an actual game to play. Which, you know, I get that. But it should be a game people can actually play.

5th Edition D&D manages to both impress me on a design level and frustrate me as a player with its approach to this, in that most classes at level 1 aren’t much more complicated than a level 1 character in some flavors of OD&D and then grow into customization and complexity over the course of the next two levels, blossoming into a character that falls somewhere between 3rd and 4th edition. I’m particularly impressed with the fact that nearly every character class can have spellcasting abilities by level 3, if they desire that path, without any feats or multiclassing.

In terms of learning curve it really seems well thought out, and is similar to what I had intended in my more D&D-flavored side project Adventure Song. But that kind of solution doesn’t work for A Wilder World, where the idea is that if you’re a magic-using thief or warrior you can have that at level one.

II: Because nearly every ability is “exception based” and abilities come in packages, thing stack weirdly…

…a problem that is compounded by an overly-specific attribute system that allowed you to achieve things like being a skilled warrior or master attribute either by stat points or by picking appropriate archetypes.

So if you wanted a character who was scary, you could put all your points in Coercion, or you could take an archetype like Intimidating or Brute, or you could buy gear that makes you look imposing, or you could put all your points in Coercion and take archetypes like Intimidating and Brute and buy gear that makes you look imposing. All of these things would make your character scary, sometimes in different yet overlapping ways.

Some balance might be retained by dint of there being a point of diminishing returns and the saying about “putting all of one’s eggs in a basket”, but when there are so many different complementary roads going up the same peak it’s hard from a design standpoint to figure out what all the ramifications of combining them are, to say nothing of doing it from a player’s standpoint.

This is part of why I kept running into problems with the math. The result of the system was that edge cases weren’t actually edge cases.

III: The numbers were never quite as transparent or intuitive as I had hoped/thought/intended.

My goal was always to keep A Wilder World fairly light on math so that it would be approachable and fast paced.

Now, math is not my particular forte, by which I mean I’m a lot better with words than numbers. But I’m good enough with words that there’s a pretty wide territory for my number skill to dwell in, and so I’ve learned over the course of my game design hobby that I tend to underestimate how my abilities to do arithmetic quickly in my head stacks up against the average person’s. I’m also better at remembering something I myself wrote than other people are, because I wrote it.

These two facts add up to me routinely failing to realize how complicated the number schemes I was using could get. Even when the rules were clear, the results were less intuitive than I’d believed.

When I was talking with Shweta earlier this year about shifting to a system that is less dependent on math to resolve conflicts and more reliant on the idea that each character has certain areas of expertise/ability, the system I came up with for codifying that grew pretty complicated pretty quickly.

These are all serious problems that become apparent whenever I step back and look at what I amassed in my aborted test draft of A Wilder World, to the point where I don’t think it is viable as anything except an experiment in mash-up character design. And I do think it is a good example of that. But what I want is something that is both playable and as easy to pick up and fast-paced as I intended it to be.

And I’ve hit on a new approach for that. I was going to say that I think I’ve hit on the right approach, but the thing is, I’ve thought that before. And been wrong.

And the other thing is, wrong or not, it was still worthwhile.

I mean, this is one of the big advantages of game design approached as a hobby rather than a multi-million dollar industry. I can spin my wheels and try things out and learn from the experience.

I’ll be blogging about the new approach in more detail, but just as I started my blogging about A Wilder World by laying out the basic principles I was aiming for, I wanted to start this new round by talking about what I need to avoid.