Simulation vs Gaming

A few weeks ago, I ran into another variant of the perennial “Great Wargamers Debate” – also known as the “Realism vs Playability”, the “Historical vs Random”, or – in my preferred title: “Simulation vs Gaming” debate. Bruce Geryk partly touched upon the subject in the excellent Geryk Analysis: Wargaming a few years back.

Despite some differences, these are all variants on the same discussion. The problem with the discussion – of course – is that the terms are so frequently poorly defined. For instance, what does the sadly overused term “realistic game” mean? I’ve yet to find a person whose definition amounted to more than “whatever feels right to me”. Taken down to its essentials, however, the various discussions almost always seem to boil down to the dichotomy of the “Unplayable Simulation” and the “Fun Fantasy”.

This is a myth.

Jake said, “Simulation and realism are the two most important aspects of gaming.” Dan nodded his head in agreement as he arranged his books, charts, and other paraphernalia on the edge of the table. “And in order to have the most accurate simulation one must have the most accurate set of rules. Fortunately, today we will use Zed Paduca’s Vive La Napoleonic Trivia, which will provide us with that accuracy,” Dan said as he tapped the cover of one of the four massive three-ring binders arrayed in front of him.
–Stephen J. Walker, “The Wargame – A Satire”, The Courier, #91, August 2004

There is a basic assumption, shared by many designers (and gamers), that greater complexity somehow makes a game more realistic, historical, and –insert favorite strategy game design goal here-.

This is a myth.

In fact, this is probably the core myth of strategy gaming: If reality is complex (which it is), then the more complex a simulation is, the better the simulation becomes. If reality contains hundred of facts, then the more facts we cram into the simulation, the more accurate the simulation must become.

From that myth is born the conclusion that leads to the never-ending debate: if “realism” and “accuracy” are achieved by loads of data, then it follows that anything that reduces the complexity and amount of data is also less realistic. Since immense complexity will inevitably lead to lower playability, it therefore follows that you can’t have a “realistic simulation” and a “playable game” at the same time.

Bill Haggart wrote an extremely interesting piece on the subject in Fire and Movement #139 (Fall 2005) called “What is a Simulation?”, in which he traced the development of the myth through the history of board wargaming. He stopped short of discussing how this myth has propagated into the computer gaming.

It is not hard to pick up that thread – particularly not for strategy games. As Geryk points out in his piece, since computer permit a game to deal with even more facts than was possible in the old days, game designers end up feeding ever more facts into their games, in the mistaken belief that this will somehow magically increase the realism. And people accept without question that knowing that the 7th Battalion of 6th Brigade is armed with Brown Bess muskets anno 1742 is somehow significantly more realistic than having a strength 5 unit.

The facts are, of course, that a simulation is – by its nature – not an attempt to reflect the entirety of reality. Rather, it is a way to gain insight into how a system works, by mimicking a few specific dynamics of a system’s operation. In short, the quantity of detail/data poured into a simulation is pretty irrelevant in terms of accurate that simulation is; the important factor is the quality of the rules and data used, in terms of reflecting the specific thing that is being simulated. To quote one of the points in Haggart’s article.

There is no Simulation vs Game dichotomy, only game mechanics that do or don’t mimic portions of reality.

Unfortunately, too many gamers (and I have been on this bandwagon too), will clamor for more data and detail because a game is not “realistic enough”. Not entirely unlike the way First Person Shooter fans clamored for more and more “photo-perfect” graphics a few years back, in search of the ultimate “realism” in shooter games. The fact is, of course, that neither of these two things make for either a better or a more realistic game. Prettier graphics in a Shooter simply makes for a prettier shooter; more details in a strategy game simply makes for a strategy game with a lot more facts.

Overloading a game with more complex rules doesn’t improve the accuracy of the simulation either. A simulation of World War 2 is not going to be improved simply because subs are are given the ability to evade surface ships, if the original rules served to simulate the aspects of World War 2 that the designer had intended well enough to begin with. Of course, one of the problems in many game designs is that it is often very unclear what the game intends to “simulate”. Is the player the King? The Finance Minister? The General? The City Mayor? When a game lacks focus, it is of course much easier to point out flaws – and the problem is then often exacerbated by the tendency of fans demanding changes to fix those flaws – usually by adding additional layers of complexity, thus muddying up the focus even more.

The model should be complex enough to answer the questions raised [the goals of the simulation], but not too complex.
— Jerry Banks, “Handbook of Simulation: Principles, Methodology, Advances, Applications, and Practice

Complexity in a simulation does not make for a better simulation – it merely makes a more complex simulation. In fact – as in many other aspects of engineering – more complexity usually makes for a worse simulation. Take the discussion of whether a tactical combat system is more “realistic” than one that simply rolls a dice. In reality, the simple system is more easily evaluated using hard facts (e.g, if a 3-1 advantage historically resulted in victory for the stronger side 80% of the time, it is easy to model this in the simple system), whereas the complex system inevitably adds in a host of complicated data and rules – thereby increasing the likelihood of the existence of”simulation bugs” that reduce the quality of the simulation.

On the same subject, there is not “one true way” to do a simulation. Any situation can be modeled from many different angles, and with many different mechanics and still make good simulations (apropos that other perennial favorite about whether real-time or turn-based games are most “realistic”).

That should not really be a surprise of course. There is not one way to do a good simulation – just as there is not one way to do a fun game. What is fun for you, may not be fun for me. Strangely, few people will contest the latter statement, whereas the former tends to have gamers up in arms.

Some people like complex strategy games. Some people like simple strategy games. Both should be able to enjoy good simulations.