One of the things that people often take for granted when playing a tabletop game is the fact that their game mechanics are almost always engineered for a series of specific goals, and this will greatly determine the ability of a Game Master to actually create a table environment that is conducive to the sort of play that is desired. The core reason for a lot of this is simply mechanical; like a video game, if the rules don’t match the desired outcomes, the game will fail.
Let’s back up that last statement with a clarification and an example, shall we? Pong is a game about playing a highly abstracted game of a tennis analogue. Were we to try to play something that felt more like soccer, we’d be slightly disappointed because the paddles are locked in a vertical direction, but the game would mostly work because we could still hit the ball back and forth down the field. If we were, however, to play a game of (American) football using the rules of Pong, we’d be patently dissatisfied, since very few of the game’s mechanics are conducive to anything that replicates the feel of a player rushing down the field with a ball or trying to kick a ball through the goalposts. We would have to modify the game to get even a similar experience, and we’d probably be just as well off ditching it and either finding an entirely different game focused on what we want or creating our own from scratch.
Note that I used an example from video games to make the concept more easily generalized, but this rule is true just as much in Dungeons And Dragons; it’s a combat simulation game and does not, for the most part, burden itself with other skills more than a simple “use it to do this” single-roll system. This, obviously, can be made up for by the means of a GM and the fact that the application of modifiers makes things a little more interesting, but it still looks at the concept of, say, social interactions largely as a “succeed” or “fail” proposition. Fortunately, the means of play allow this to be brought into a narrative context, but the core act of rolling to pick a lock will not be as engaging as, for instance, fighting a goblin, which is actually more intellectually stimulating because of all the various factors involved as opposed to the highly predictable and linear handling of skill tests in D&D.
When finding a system that works, there are several things to consider. I’ll go into more detail in this in my upcoming e-class “So You Want To Play”, but the core things that really matter are the game’s reproducibility or randomness, its complexity, and the degree of abstraction it uses.
Most games trend toward reproducibility, because otherwise some narrative elements don’t make sense; few games want to have protagonists who are exactly identical in terms of their ability to impact the world, because the game would lack the ability to make informed choices outside the challenges being faced, and essentially amount to gambling instead of allowing players to make conscious decisions to craft a character. On the other hand, randomness is somewhat important as well; it’s generally agreed that it’s not constructive to allow players a monopoly on play just because their character is better crafted according to the system’s criteria, and it can reduce the need for each individual character to take on a distinct role, allowing rules to focus on core gameplay elements rather than having to introduce new, crucial, complications to occupy the players and GM. As far as randomness versus reproducibility goes, it’s truly a judgment call that depends on the desires of the group, and whether the GM chooses to religiously follow their designs for the narrative, the design of the game, or compromise and make judgment calls on the fly.
Complexity is also another major factor in whether or not a game is appropriate. A game that is highly complex tends to be better suited for a GM or player group that prefers to have a setting to work around, easily definable outcomes, and a limited set of options, but the downside of this is that there is more of an investment and very rarely do more rules make games more flexible. On the other hand, games that have only a minimal ruleset, such as Savage Worlds, tend to be too narrative for players who like to let the dice determine what happens, and require more input from GM’s and players during play to help determine the details of what happens. To a certain degree, it’s also important to consider the degree of thought that went into the complexity of mechanics; something that focuses on extraneous elements or lacks focus will be worthless, while a very simple system that does exactly what it intends to do with each mechanic can feel more “crunchy” than a system with many more rules.
The degree of abstraction is another thing to consider; rules can be as complex or simple as the designers want if they don’t have any connection to narrative or other game mechanics. Dungeons and Dragons gets a lot of flak for having “goblin dice”, where players’ abilities either work and a goblin dies or fail and the goblin lives. It’s not an abstract system in the slightest, especially when compared to something like Cortex, but it’s a valid case; a game with very complex rules will only have a crunch to it if the rules mean something more than just reading dice to get a number; Shadowrun, for instance, has variable hits, which are modified both by the attacker’s success and the defender’s ability to resist damage, making its damage system less abstract because it allows for the possibility of glancing hits or especially good hits on a weak spot.