This chapter of Tools for Dreaming is an attempt at pushing a philosophy I call “conscious design,” in contrast to what I see as the problem of unthinking repetition of RPG design cliches. This is so, so not me lording over other people. There are so many design cliches that I struggle with all the time. The influence of Apocalypse World has helped me get better at realizing my RPG ideas, but I’m well aware that that comes with a set of deign cliches too.

Conscious Design

“Engage in conscious design.” That’s my most important bit of design advice, not only for RPGs, but for anything. I put it on a T-shirt even.[1]

When you design a game, you can do just about anything, but you need to do it consciously, and be aware of the effects things have on play. That’s a lot harder than it sounds, but it’s something that RPGs need to get better at. There are patterns that RPG design falls into, and while there’s merit in using methods you and your audience are familiar with, the unthinking repetition of well-worn conventions may have stunted the medium’s growth. That’s to be expected when such a large portion of this hobby is dedicated to relatively minor variations of D&D, but unless Wizards of the Coast or Paizo are seriously looking at your resume, you shouldn’t be beholden to Gary Gygax’s highly specific vision.

It’s not inherently bad to use time-tested methods, but just because something has stuck around for a long time doesn’t mean it’s the right choice, especially if you’re transplanting it to a different context. D&D-isms are often a problem in RPGs not because they’re bad mechanics, but because they’re a poor fit for a game that’s not primarily for running high-risk/high-reward dungeon crawls where survivability is something characters have to earn. In some of the later chapters of this book I’m going to try to break down some of the most basic elements that make up an RPG. There are a lot of things that you likely already know, but hopefully I can give you a new perspective on those things, so that you can see just how many choices there really are.

Designing consciously means always striving to deepen your understanding of the medium, and this book is basically an attempt to help make progress with that, for myself as well as you the reader. You need to be able to step back from old habits and clichés and analyze what you have in front of you, which requires the ability to break things down in ways that most people don’t get out of casual play. It’s all too easy to fall into unthinking design even when you don’t have 40 years of D&D breathing down your neck. Whenever I’ve had a game design process take a long time, it’s usually because I’ve let some flawed ideas calcify, and the solution is often to cut out or radically simplify rules that I’ve left in through multiple drafts.

“Conscious design” doesn’t mean that you can’t throw stuff at the wall to see what sticks. While you should strive for conscious design, it’s not in the nature of games, especially RPGs, to be something you can fully anticipate. What you do need to make the process of refining your game into a finished product one of reaching that understanding. Epidiah Ravachol (who is a brilliant game designer) does a thing he calls “playstorming,” where his gaming group will essentially make up a game as they go along, designing and trying things out on the fly. It’s a fun activity for its own sake, but it’s also a starting point. Playstorming won’t produce a finished product, but it can create a solid starting point for developing a game and achieving conscious design.

Conscious design is especially important when it comes to understanding the perspectives of the people who play your game. If you make a list of 12 powers and tell players to choose two, you have a perspective of needing to write a list of 12 things, the player has to worry about picking the right two that will define their play experience, and the GM has to think about what powers each of the 2-6 players’ characters will have and what they’ll do to the game.

To illustrate the concept of conscious design, I’m going to explore some common examples of unconscious design in detail. These show how tradition and symmetry can lead designers astray.

The Perception Pitfall

It’s a cliché in D&D games that the DM asks the group to do perception checks to see if the PCs notice something important. If everyone fails, the party may have more difficult going ahead, or they could even miss something they need to make progress in the adventure at all.[2] If one or two PCs succeed, you get the kind of awkward thing where the DM tells them and you have to see how they relay that information to the other players.[3] The whole thing is inelegant and shows a certain lack of thought about what effect it has at the table. Earlier versions of D&D didn’t actually have perception checks. The players would describe what their characters were doing to check for traps and such, and the DM (taking the PCs’ ability scores into consideration) would tell them what they find. This has a danger of running into “pixelbitching” (to borrow a term from graphical adventure games) when players end up spending an inordinate amount of time trying every little thing to find the right answer to an obscure puzzle, but to some extent it’s an unavoidable consequence of making puzzles an aspect of a game. While the old-school approach lacks the appeal of a (seemingly) clear-cut rule, it has the advantage that it’s based on the players’ choices rather than a die roll.

Skills became commonplace in other RPGs pretty early on, giving characters numerical representations of how good they are at a variety of tasks and fields. Among other things this was a way to have a game without classes that could still provide meaningful mechanical distinctions between different characters, as well as distinguishing between characters of the same class in games that do have them. AD&D 2^nd Edition (and a few supplements for 1st Edition) had optional rules for “Non-Weapon Proficiencies,” but when Wizards of the Coast made D&D 3^rd Edition, they sought to update the game to more current RPG design methods, which included adding a full-on skill system. “Spot” was a natural thing for the designers to think to include in the list of skills, but they made it a skill with the same mechanics as every other skill. While it’s reasonable to limit the proliferation of disparate sub-systems that characterized AD&D, perception is a very clear case where it required a more careful examination of how the mechanic fits into the overall structure of gameplay. It’s entirely too common a cliché that failed perception checks can make a game grind to a halt, and while a well-written GM advice chapter will warn against using skill checks that way, it’s not particularly unreasonable for someone sitting down and looking at the skill rules to conclude that that’s how Spot checks should work.

D&D 4^th Edition introduced the concept of “Passive Perception,” where each character has a value (10 plus their Perception skill bonus) that represents what kinds of things the character will notice without making an active search. A DM armed with a list of the PCs’ Passive Perception scores can just refer to it and announce when one of the characters has spotted something, rather than taking the time to have everyone make rolls, total up their results, announce them, and see who rolled high enough to hit the DC. It depends a lot on the DM having the mental bandwidth to use it, but it neatly cuts out a lot of the busywork that perception checks involve.

Robin Laws’ GUMSHOE System meanwhile is designed around playing out mystery stories. While you make dice rolls for things like fighting, when you investigate, your investigative skills are actually pools of points you can spend on trying to find clues to move things forward. You have to figure out the right place and method to search, but you won’t fail to find a clue because of bad luck with the dice. Jared Sorensen’s InSpectres goes a radical step further, leaving the true nature of whatever mystery is going on undefined, and letting the players simply create what their characters discover on the fly. It provides a very different kind of experience, but it does what it does very well.

Ultimately, the real question is what the purpose of introducing uncertainty into PCs’ ability to notice certain things actually is, and what the particular method used actually achieves. GUMSHOE shifts the challenge to one of applying your investigative resources in the right places, while InSpectres transforms the whole concept into a part of its collaborative storytelling process. Skill checks in general require a lot of mental bandwidth and snap decisions from the GM, so that although they outwardly feel like hard mechanics, in practice they’re hugely dependent on the players’ human interactions with the GM. That isn’t a bad thing, but it’s certainly something to be conscious of.

Disadvantages of Disadvantages

It’s less common nowadays, but there are still a fair number of RPGs that let players take “disadvantages” (Flaws, Hindrances, etc.) in exchange for more points to spend on positive things. It’s a decent idea, but it has certain pitfalls. GURPS is probably the clearest example, and also one of the RPGs with the most extensive selection of disadvantages. In GURPS you make a character from a single pool of points: 100 points for a kind of heroic but average person, up to 300 or 400 for an out and out superhero. A lot of the game’s mechanics are based on the idea that spending 10 points on one thing is equivalent to spending 10 points on another thing. In practice there are a lot of “incomparables” though, without any direct equivalent in value elsewhere. Is 10 points worth of extra running speed actually worth the same as 10 points worth of extra hit points? That’s difficult to say. The particular campaign also can greatly affect how much practical value things have. Spending 10 points on science skills may be a wise investment in one campaign, but in another, putting those points towards being able to shoot a gun could be far better. Even if you’re going to have a game where PCs will be both shooting and sciencing in roughly equal measure, it’s still hard to assign objective values to them. To some degree that’s an inherent issue with making game characters in general, but when it comes to disadvantages, GURPS and its brethren have some particular issues you should be aware of.

GURPS is a well-made RPG overall, but it has the issue that it incentivizes players to try to get as many points worth of disadvantages with as little actual detriment to their character as possible. You could get a lot of points by making your character blind and paraplegic, but chances are you’re not playing in a campaign where a character like that could properly participate. A lot of disadvantages are contextual though. If your character is deathly afraid of cats (ailurophobia), it’s something that only matters insofar as the GM makes cats a part of the game. That means the GM has the job of manually regulating things so that this disadvantage comes up with appropriate frequency. If there are never any cats, then the player has gotten some free points, but if there are All Cats All The Time, the GM is being kind of a jerk. In contrast, disadvantages that affect purely mechanical things, like giving your character a slower running speed or fewer hit points, are things that the player can’t really ignore in play. In a typical gaming group, the GM could easily wind up needing to keep track of and keep relevant a couple dozen Advantages and Disadvantages across the different PCs, all while trying to do so in a fair and equitable way.[4]

Given that GURPS has been around since 1986 and is currently in its 4^th edition,[5] that issue obviously isn’t a total deal-breaker, but it’s something that RPG designers have tried to fix in a variety of ways. A lot of it comes down to making these negative traits be something that players can’t ignore, or that would be to their detriment to ignore. In Golden Sky Stories the Weaknesses you can take are tied to specific Additional Powers (making them match both thematically and in terms of game balance), and for the most part they’re consequential, so that the GM (“Narrator” as GSS calls it) doesn’t have to go out of their way to make them matter. In Fate, a character’s Aspects can do double-duty as both advantages and disadvantages, and while you don’t actually have to write your aspects in such a way that they express disadvantages, it’s better to at least have some of them be double-edged, because that’s how you gain Fate Points, which in turn are critical to effectively playing the game. In 7^th Sea meanwhile, you pay points for Backgrounds, and you get a bonus when the story element in a Background comes into play.

I think the issues with Disadvantages ultimately come down to the need to interrogate what effect a given mechanic actually has at the table. GURPS appeals to the idea that you can quantify things about characters and write them down on your character sheet as more than just flavor text. In D&D you can declare your character is a pacifist and try to play that way,[6] whereas in GURPS you can take Pacifism as a –5 to –30 Point disadvantage (depending on which version of pacifism), and it has a detailed description. GURPS has a lot of disadvantages (the Disadvantages chapter of the current edition’s Base Set is 48 pages long), and a ton of character traits in general (the first 257 pages of the 336-page Basic Set: Characters book is pretty much all character creation, spanning hundreds of different traits). While that’s partly because GURPS is a universal RPG that allows for just about anything (hence the disadvantages include things like Decreased Time Rate, Invertebrate, and Unusual Biochemistry), it also comes from a general bookkeeping approach to character creation. The game simply doesn’t leave a lot of room for intangibles that the rules don’t touch on in some way. It (usually) doesn’t care about the color of a character’s skin or hair, but their personality, down to small quirks (which are a special category of –1 point disadvantages) is fair game.

That extends to just about everything in character creation, and while there’s a certain appeal to an RPG rulebook that will tell you what biochemistry is, there’s a good chance it simply falls outside the scope of anything that matters in your typical campaign. That’s where it becomes helpful to be willing to simply let some traits fall outside of the scope of the game mechanics. While in GURPS being an accountant is a skill (IQ/Hard), in an awful lot of games it would amount to flavor text or a custom distinction. A character who is an accountant can potentially be interesting, but very few gamers are going to be doing the kind of campaign where a player has to make an Accounting check.

Simulation vs. Abstraction

There’s a thing in fiction writing that something that happened to you in real life won’t necessarily make a good addition to a story. You might be presenting it wrong, it might just not be a good fit for the story, or any number of other things. Similarly, how close a game rule matches reality is orthogonal to how good it is for making a quality game. I’m not advocating against simulation—though admittedly it’s not a big part of my own style of game design—but rather pointing out that simulation is not in itself a progression towards quality.

In 1986, Leading Edge Games put out Phoenix Command, an RPG with incredibly detailed rules for modern combat. Its combat system will resolve how far a given type of gun with penetrate through various types of armor, tell you which specific finger gets injured, and so on through a set of equations and tables. The game is an impressive achievement in many ways, but it’s also an intensely time-consuming way to simulate a gunfight. I certainly don’t begrudge anyone who enjoys Phoenix Command, but there’s a reason the game has become an interesting obscurity and collector’s item rather than a game that many people play. It’s neat that such a game exists, but a lot of people want to have fights be quicker and less messy.

Although D&D has some elements that veer into simulation territory, some of its most central rules are highly abstract. When Dave Arneson’s Blackmoor players complained that their characters shouldn’t go down from a single hit (in the way wargame units typically do), he added the concept of hit points to the game, apparently inspired by a naval wargame. Where in Phoenix Command you accumulate and track injuries to specific parts of the human body which have specific effects, D&D characters simply have a pool of hit points, and when you run out you’re dead or dying. How many HP a character has left only tells you how close they are to going down, and by itself that total doesn’t suggest any specific injuries or cause any mechanical disadvantages. There are plenty of options between the two extremes of Phoenix Command and D&D, but hit points have the advantage of being substantially faster and easier to handle. “You have this many points, if you get hit you lose some, if you get healed you regain some, and if you lose them all you’re out of the fight.”

HP are excellent as a game mechanic, but a poor simulation of how people get injured. If you take enough of a beating that you’re halfway to death, chances are you’re not going to be able to run or swing a sword nearly as well. It also doesn’t make sense that a high-level character has more meat to cut into the way a dragon does. Originally, Arneson and Gygax just didn’t give it much thought, because they were focused on having a fun game. Mike Mornard (a.k.a. “Old Geezer”), who played a lot with Gygax, likes to say, “Hit points simulate hit points.” In the AD&D 1^st Edition rules, Gygax explained HP as an abstraction of not only the capacity for injury, but a character’s level of stamina, their ability to keep fighting. He envisioned two warriors swashbuckling, where a “hit” in terms of the D&D rules would take off HP, but represent a small nick or even a miss that took a lot of effort to avoid. Thus, a more experienced warrior was better at enduring through battle, and had more HP. The level 1 character with a handful of HP doesn’t have less meat than the level 10 guy, but it takes a lot less to land the blow that will take him down. This was a post hoc rationalization for an abstract, game-y rule, but that rule is sufficiently intuitive and functional that most people get at least the general idea without too much explanation needed.

When I was younger, I and the friend who introduced me to RPGs had a weird running joke about the perfect real-life RPG, which would be staggeringly complex, and allow a virtually perfect simulation of how things worked in the real world.  I’ve since started writing a short story based on that premise, where a guy finds among the game collection he inherited from his uncle a strange game in the form of a few hundred hardback books. (I should finish it one of these days.) It’s hard to say just how big an RPG that accurately simulates everything would be, but it would be massive. Tabletop RPGs have to be able to more or less fit into human brains (with the help of reference materials and such), so even if you’re planning to make a highly complex RPG, you still have to pick and choose what elements to actually include.

This was an issue with RPGs of the 1990s, when designers would routinely include detailed rules for simulating a huge variety of different physical situations (drowning, falling,[7] medical treatment, etc.), regardless of whether they were at all likely to come up in actual play. This made more sense for GURPS, which is meant to handle virtually any genre, but it’s also something that showed up in more specialized games. Palladium has a strange habit of including the same rules for alignments and insanity in every single game they made, even though those things don’t especially fit with things like Robotech. While there were undoubtedly some people who appreciated having those rules when they came up, I think the better option is to provide simple, broadly applicable tools that make it easy to handle different situations as they come up. If the system lets the GM have players make checks and assign wounds, it’s easy for them to go “Make a swimming check; if you fail you start drowning and take a wound.” Moreover, delving into the minutiae of how to handle specific situations can distract from providing tools that the game needs to cover its chosen subject matter well. Robotech doesn’t especially need insanity rules (at least not the ones copied and pasted from other Palladium games), but it certainly could use rules to handle putting on concerts and swaying people’s emotions, considering that those were central to the premise of the entire first story arc of the series.

There’s also the fact that while there definitely are games that benefit from realism, a lot of RPGs deal with subject matter that’s so rooted in genre that realism is counterproductive. Heroes Unlimited is Palladium Books’ superhero RPG,[8] and in the Villains Unlimited supplement, Kevin Siembieda included an essay about how in real life knocking people out is difficult and dangerous. In game terms, you’d need a natural 20 to do it, and it’s scarily easy to give someone a serious brain injury instead. The problem is, if you’re claiming to be making a superhero RPG, that’s entirely beside the point. A character discovering the hard way that you can’t just knock someone out like in the movies would be great for a Fiasco game, but if you’re selling your RPG as being about superheroes, it’s reasonable for players to expect to be able to do stuff like knock out random thugs the way Batman and Captain America do all the time. You could practically make “Knock Thugs Out” a stat in some cases, and it’s easy to imagine a player encountering Siembieda’s treatment of knockouts turning to the GM to say, “Have you read comic books? Like ever?” If Heroes Unlimited billed itself as a game for more realistic superheroes, it would be a very different and much more unique game, but since it’s not, bolting that essay onto it was frankly a pretty strange thing to do.

---

[1] You can buy one from Spreadshirt if you want.

[2] This has happened even with public demo games run by D&D staff.

[3] Also, there’s the thing where the DM will call for a perception check purely to keep the players on their toes, so that you end up rolling the dice for… nothing.

[4] While I think the Aspects in Fate are better in a lot of ways, some versions of the game just plain have too many of them. Personally I think 3 aspects per PC is about right, but in Spirit of the Century each PC started with ten.

[5] Though granted pretty much everything Steve Jackson Games puts out these days is secondary to Munchkin.

[6] Though a ton of things about the game will work against you.

[7] Jason Morningstar’s game Drowning & Falling is a parody of this phenomenon, and an RPG where literally every challenge the PCs face involves either drowning or falling in some way, even if it’s drowning in someone’s eyes or falling in love.

[8] I am not a fan of Palladium’s “Megaversal” system, as you probably guessed if you’ve compared it to basically anything I’m advocating in this book. I will give Heroes Unlimited some points for including a wonderfully gonzo range of superhero types, and for having some great old-school superhero art, including a cover by industry legend Jim Steranko.