Simulation vs. Playability: "Gamist" vs. "Simulationist" is an Axis
The Background Discussion for a Blog Series
It’s All Simulation, but What Kind of Simulation?
In last Friday’s Weekly Geekly Rundown, I linked
’s recent post where he rekindles discussion of the Gamist/Narrativist/Simulationist theory that was very popular on Ron Edwards’ Story Games boards of the early 2000s at indie-games. While the theory predated Ron Edwards’ systematic presentation of the GNS Design rubric, he was its most passionate advocate and it gained real traction in his community.
I gave a cursory discussion of GNS Theory and compared it to Jeffrey A. Johnson’s Fourfold Way of FRP from Different Worlds Magazine #11 (1980) in my piece on the Swedish Super Hero role playing game Supercrew about a month ago. I’ll be expanding on that discussion in the future, but Ryan’s piece covered one of the main points that I was going to make in my follow up piece, namely that “Gamist” and “Simulationist” approaches don’t stand on their own because all non-abstract games are a kind of simulation.
That’s an easy point to make because saying “all non-abstract games are a simulation” is a tautology and it is one that Jeffrey A. Johnson presented more clearly than design theorists who wrote twenty years after his cornerstone piece in Different Worlds. I say cornerstone piece because game designers like Aaron Allston and Robin Laws were very influenced by Johnson’s piece and the subsequent discussion that happened in Different Worlds. What you will notice if you compare Johnson’s Four Fold Way to the GNS triangle is that “Simulation” and “Pure Fantasy” are on the same axis in Johnson’s model. That axis is the “realism” axis where a games simulation goals are argued to be on a continuum between pure fantasy and simulation.
The use of simulation as an axis descriptor here is a little different from what Ryan means in his piece, in that the entire axis represents what a game is simulating rather than whether it is simulating. That’s because this builds upon the classic “designing for effect” versus “designing for incident” debate touched upon by John Hill in his Designer Notes for Squad Leader in Issue #9 of Fire & Movement Magazine.
It is the difference, in some ways, between Dungeons & Dragons combat and the combat system of Runequest. Both systems have a form of “hit points,” but they represent different things and the combat systems are at different levels of abstraction. In D&D, a character is perfectly healthy until all hit points are removed and then the character is either unconscious or dead depending upon the version of the game. The combat system is abstracted so that the “hit roll” was initially designed to be a reflection of the “one opportunity” to effectively hit an opponent within a time window (either 1 minute or 10 seconds depending on the edition). Parrying and dodging are merely modifiers to a hit roll and there is no actual blow by blow representation of play. D&D is “design for effect” in its origin, though has become more "design by incident” due to how players enact the system.
Runequest on the other hand is a blow by blow representation of combat where a character rolls to hit, the opponent then decides whether they parry or dodge, and if they do so they roll to see if they successfully dodge. While a person’s “body” has a number of “hit points,” so too does each limb and if a location suffers more damage than it has the capability to absorb it becomes useless in some fashion. The Runequest system, which was based on Steve Perrin’s experiences fighting in SCA tournaments, simulates individual attacks. While players of D&D might often think of their attack as their only action that round, in older editions this was not the case as it was the only “effective” attack and the only statuses in D&D were alive and dead whereas various limbs could be wounded in Runequest. Both games are simulations of combat, but one is more “Simulationist” in the Johnson sense.
The key here is that one is a simulation of the end effects of combat while the other is a simulation of the individual incidents that occur during combat. The very old RPG En Garde! and the Avalon Hill game Gladiator are even more incident oriented in their designs.
Further Expansion of the Concepts
To expand on this idea a little, both Chess and Men of Iron are to one degree or another simulations of medieval warfare. Men of Iron has what Elias, Garfield, and Gutschera (2012)1 would describe as higher "intensity" of the medieval warfare conceit than Chess, but both do share that central conceit. On the Elias, Garfield, and Gutschera "Scale of Intensity for Conceits," Chess is rated a 3 (Very Light Conceit) while Men of Iron would likely be rated around an 8 (Simulation, but with many sacrifices to gameplay). As a further illustration Tic-Tac-Toe rates a 1 (Purely Abstract) and Squad Leader ranks a 10 (Full-on Simulation). This might make one wonder where Advanced Squad Leader would rank, but I digress.
By the discussion above, you would see that John Hill would disagree that Squad Leader is a 10 on the Intensity for Conceits scale, but that is a discussion for another time. What is important here is that Elias, Garfield, and Gutshera have created a scale that rates a game based on how much the design is dedicated to simulating “incident” level occurrences versus those that are simulating at an abstract level where the game and the simulated concept are only loosely related.
There are two games that I think best demonstrate a pure “Gamist” perspective. These are the World of Warcraft game by Fantasy Flight Games, not to be confused with their World of Warcraft Adventure Game, where the combat system feels more like shifting tokens around endlessly more than it feels like combat. The moving of tokens feels more like it is the game than what it is simulating and it’s not a particularly fun game. It is, however, very fiddly.
The second is the Marvel Universe Role Playing Game that was published by Marvel Comics in 2003. That game used no dice and players shifted stones around on their character sheets in order to perform actions. Once again, this was a game that had almost no narrative connection to the thing being simulated.
I understand that there are those who may disagree with the initial premise that "all non-abstract games are a simulation" and dismiss it because it is a mere tautology. There will be others who completely disagree with the premise as an a priori. I believe it will prove useful to have a series of discussions about role playing games as simulations of the various subjects they address and how they simulate those concepts.
One area where I disagree with Ryan, and seek to clarify, is that I don’t believe the framing of games as simulations is exactly what Ron Edwards' GNS (Gamist, Narrativist, and Simulationist) system of game play analysis was getting at when describing games as Simulationist. I tend to interpret GNS Theory as a theory of play that can inform design and not a theory of system deconstruction and design. As is suggested by my discussion above, I equate Edwards’ Simulationist classification with “incident” level simulation and “gamist” with a focus on abstract mechanics over incident modeling. As Ron states in the above linked essay, "These terms, or modes, describe three distinct types of people's decisions and goals during play." These are player goals toward which games may be designed, but in my opinion a "Simulation" is not the same as a "Simulationist" game.
To illustrate, the excellent Narrativist game The Extraordinary Adventures of Baron Munchausen is a simulation of storytelling as the good Baron himself might engage in it. It is not a simulation of the Baron's adventures. James Wallis had considered making a game where players could enact those adventures, but decided against it in favor of a simulation of storytelling in a particular style. Another example of a game that is a simulation of storytelling is Tales of the Arabian Nights. Both Baron and Arabian Nights simulate the activity of storytelling within their conceits differently, but in the end the game play of both are best described by the stories created within the rules of the game. There is a reason that Wallis calls these kinds of games "Story-Making" games. Another example of a game of this type is The Committee for the Exploration of Mysteries and the sole purpose of play is to tell stories about things that have happened and not to experience emergent stories by simulating the adventures. Thus it's possible to have simulations of storytelling that results in story-making which in the end results in storytelling when the results with game play are shared.
Okay, enough of the metaphysics of games being simulations. Let's move forward please -- ed.
Games are also about fun, and to be fun games must be playable. This is as true of role playing games as it is for any other kind of game. As Robin Laws says in his masterwork Robin's Laws of Good Game Mastering, "Roleplaying games are entertainment; your goal as GM is to make your games as entertaining as possible for all participants."
The key part of that statement is "as possible for all participants." Because players come to a gaming table with different "ideas of fun" as highlighted in Edwards' GNS theory, roleplaying game design must make decisions between the level of depth of simulation and the playability of the system. Some players find granular verisimilitude and accuracy of representation entertaining. For these players reading Chapter H of the Advanced Squad Leader rulebook is as much fun as actual game play. Other players might enjoy quick systems or systems that foster the creation of narratives.
Historically, the conflict that informed John Hill’s design was the tension between "realism" and "playability." In Issue 8 of Moves magazine (1973), Victor Madeja argued that "Commercial wargames fail to accurately represent modern war. Although no game will ever recreate the confusion, horror and destruction of war, we should at least expect a wargame to partly simulate the decision-making process involved in actual battle. Instead we have chess-like caricatures of reality. What semblance of realism we were led to expect is sacrificed on the altar of playability" (Emphasis mine). For Victor, there was a clear distinction between realism and playability and he thought that games at the time leaned too much toward playability and not enough toward actual simulation .
By Issue 14 of Moves (1974) John Hill, the eventual designer of Squad Leader, addresses the conflict by stating, "One of the hardest problems facing any war game designer is the careful balancing between playability and realism. Actually, any reasonably competent wargamer could probably design a realistic 'simulation,' but to design a good game is something else. As an example, 1914 was an excellent simulation of corps level fighting of that era, but as a game it was worthless -- it couldn't be played." John Hill would eventually go on to become an advocate of what he called "abstraction." This was a controversial game design philosophy in which the designer cared less about "what actually happened" and more concerned with the "effects" of what happened and how to model those effects. So, for Hill the fact that gunfire affected morale was more important than modelling the specific physical effects of bullet trajectories. Examples of "abstraction" designs in role playing games include D&D's "hit points" and the "effects based design" of Champions.
The tension between simulation and realism is one that has been discussed in role playing games since the origin of the hobby. In the Advanced Dungeons and Dragons Dungeon Masters Guide, Gary Gygax writes, "Of the two approaches to hobby games today, one is best defined as the realism-simulation school and the other as the game school." -- You can see in this discussion the origins of Edwards' GNS theory. -- "AD&D is assuredly an adherent of the latter school. It does not stress any realism (in the author's opinion an absurd effort at best considering the topic!). It does little to attempt to simulate anything either. Advanced Dungeons & Dragons is first and foremost a game for the fun and enjoyment of those who seek to use imagination and creativity...As a realistic simulation of things from the realm of make-believe, or even as a reflection of medieval or ancient warfare or culture or society, it can be deemed only a dismal failure...Those who...generally believe games should be fun, not work, will hopefully find this system to their taste.”
In his paragraph on design intent, excerpted above, Gygax clearly puts himself in the "abstraction" design camp. His discussion of Hit Points in the DMG also makes this clear, whereas those who criticize hit points or how armor "makes it harder to hit and doesn't stop damage" fall more into the Simulationist camp. I would like to say that I disagree with Gygax that his game "does little to attempt to simulate anything." I would argue that it is simulating heroic fantasy, but it is doing so from an abstractionist position. It's a small distinction, but not an unnecessary one.
As an aside, most gamers or designers are a combination of abstractionist and simulationist. Ken St. Andre, the designer of one of the most abstractionist rpgs I have ever played, doesn't like armor class systems because they don't simulate what he wants. This is the case even though his Tunnels & Trolls and Monsters! Monsters! combat system sacrifices specificity for playability and speed of play.
When it comes to the tension between "simulation" and "playability" there is not a procedural definition of what is right or wrong. What is right or wrong doesn't even depend on what is being simulated. What determines whether it is better to favor simulation or playability is how that decision works within the rules set and the goals of the game itself. Sometimes it is important that a game be a good simulation of what it is trying to represent. Champions is very much an "effects based design" system in character creation, but its combat system simulates the panel to panel flow of comic books extremely well. Villains & Vigilantes has a random character creation system that favors simulation -- though it also includes GM "rulings over rules" -- over abstraction as it defines specifically what Flame Powers and Ice Powers do and how they work rather than define effects and have you decide what matches what. Both are good games.
In the coming weeks, I'll be looking at some games and how they address the Simulation/Realism vs. Playability/Abstraction conflict. I'll be starting with Villains & Vigilantes and how it emulates Force Fields and Telekinesis in its simulation of super heroic conflict. While I think that the V&V system overall is quite good, I believe that the designs of these two powers demonstrate good "simulation" on the one hand and "awkward" simulation on the other.
I'd like to leave this conversation with two quotes for Will Hindmarch and Jeff Tidball from their Things we Think About Games.
1) Theme and gameplay are two different things.
2) Balance is not the same thing as fun.
Elias, George Skaff. Garfield, Richard, and Gutschera, K. Robert (2012), Characteristics of Games. Massachusetts Institute of Technology Press.
Just like sports, it's about having fun.