In my instructional technology and design classes, I've been introduced to several theoretical models of instructional design. As I was eating breakfast one day, I thought about how these might apply to RPG design. I'm familiar with several models, including ARC, ADDIE, and ASSURE. They are all fairly similar, and all might prove useful, but in the interest of space and focus, I'll approach this using the Heinch, Molenda, Russell and Smaldino's ASSURE model (first detailed in 1999 in Instructional Media and Technologies for Learning). In my opinion, ASSURE fits better and will be easier to explain well (though ADDIE would have lent itself to a catchier column title).

ASSURE is an acronym, which stands for Assess learners, State objectives, Select media, methods, and materials, Utilize media and materials, Require learner participation, and Evaluate and revise.

Assess learners

In an educational setting, this would literally mean to know the demographics, technology familiarity, level of content knowledge, level of interest in the content, etc. of the student you'll be teaching (or for whom you're preparing the lesson materials). In the case of an RPG, you don't have learners, but you do have a target audience, whether it's your gaming group, or a segment of the overall RPG-buying public. Don't be egotistical and think you'll write the one game to rule them all, or the last RPG anyone will ever need. The hobby is too diverse for that to happen, so you'll need to know just who you're writing this for.

There are a lot of ways to categorize gamers, from rec.games.frp.advocacy's Threefold to Ron Edwards' various Forgite theories. Whether you find RPG theory useful or not, you should at least be able to mentally picture the kinds of games people will be playing with your games. Will it be World of Warcraft-esque “grinding,” as D&D's fourth edition sometimes encourages, Original D&D-style tomb raiding, in which the players are tested as much as their characters are, the kind of political games you often see in Vampire: the Masquerade, or something entirely different?

Are your target gamers interested in combat, deep in-character experiences, theatrics, politics, or what? There is no right or wrong answer, but it's important, when designing a game, to know the answer.

State Objectives

In an educational context, this is essentially the lesson, skill, or knowledge you want to convey. In a game-design context, this overlaps somewhat with your target audience. I've talked about clear goals since the first column, when I called it “intentionality.” The point it, successful games, like successful educational products, have a focus, a goal. Original D&D was a game through and through, and it tested its players wits more than their characters' abilities. The Hero system began its life as Champions, the first point-balanced Supers game (and that's still where it shines, IMHO). Sorcerer was Ron Edwards's stab at a writing pure narrative game without any of the gamist and simulationist baggage he saw in White Wolf's original World of Darkness.

Games can have hybrid goals, combining aspects of other games or media the designer admires or finds intriguing, but those goals really need to be clear not only to the designer, but to the players. I don't think that it's ever said, explicitly, that Shadowrun was a cross between D&D and 1980's Gibsonian Cyberpunk, but it's clear as day in the earlier editions. Even the Shadowrun itself is the futuristic re-imagining of an old-school dungeon crawl. Players who were familiar with D&D could dive right into Shadowrun, and those who weren't could easily pick up the style from the examples, flavor text, and sample adventures.

Select Media, Methods, and Materials

When designing an RPG (prior to formatting a publishable book form), this “stage” means selecting the kinds of mechanics and setting elements that will best support your goal. For example, if you want a gritty, realistic feel, selecting a combat system with abstract hit points (like D&D) or high-flying combat maneuvers (like Feng Shui) would be a bad idea, as would undervaluing the lethality of common weapons, so that characters could easily shrug off bullets or sword wounds. If you want a tactical game that challenges the players, don't make the mechanics so abstract or closed-ended as to base everything on character skill. Leave room for the players to affect things. Likewise, if you want a fast, furious, stunt-driven style (like you might see in anime, kung fu films, or Errol Flynn movies), a detailed tactical system with huge numbers of stunts and maneuvers which interact in different ways with each other and with a separate set of powers could be a bad idea. If combat resolution is slow and awkward, the combat itself will feel slow and awkward, even if the maneuver you spent the last thirty minutes resolving would look awesome on screen.

Setting elements should also support your goal. D&D's “points of light” default setting does its job well, because it separates towns, which are usually relatively safe and civilized, from the dangerous and lawless outlands where PCs can fight monsters, raid tombs, and generally act as a law unto themselves (and where they can expect no rescue or backup). A well-ordered Renaissance setting, with established domains and courts, would have only gotten in the way of the default D&D play style (though it would have been much better for a game of politics and intrigue).

Utilize Media and Materials and Require Learner Participation

The U and R in ASSURE are fairly elementary in RPGs. Utilizing the game you created basically comes down to putting it before your players in a form they can understand, whether it's a nicely formatted PDF of just an understandable verbal explanation. Requiring learner participation is sort of meaningless in an RPG setting. In the educational world, it means make sure your students are involved in the learning process in a meaningful way, and that you're not just lecturing or telling them to answer questions out of textbook. But anyone who's playing an RPG is, by definition, participating. So these two sort of happen automatically, if the first three are handled.

Evaluate and Revise

Evaluating and revising is the last step, and arguably the most important one. I typically write games for the sheer joy of creating something (and I suspect many other amateur, and perhaps professional, game designers do the same). But to make a game really work, you have to revise and edit it. Look at what worked and what didn't work in the playtest. Talk with the players; they can tell you what they thought worked and didn't. Make changes, and then playtest it again.

Playing a campaign, even a short one of half a dozen sessions will tell you a great deal about how the game actually works, because the players will be truly invested in their characters, and will try things that may not ever come up in an intentional playtest.

In conclusion, there's nothing magical about ASSURE, ADDIE, or any other design model, but having a series of steps to keep in mind can help you understand and remember ways to make your game better. The sad fact is, there are probably more barely-playable fantasy heartbreakers (that is “it's like D&D, but better because [you can play as cat people] [armor reduces damage] [you don't memorize spells] [etc]”) floating around the net than there are original, playable RPGs. Aside from reading and playing a wide variety of RPGs (so you don't have D&D as your only point of reference), the most important thing you can do to make your game better is to have a plan, and think things through.