When you're designing a game, you want to make sure that you have an understanding of what exactly you are doing–games are to a certain extent learning opportunities, but they are also decision machines.

What that means is that the player will always have to make a "decision", and you need to be thinking about what decisions you are encouraging with your design.

Poor design can lead to a game where there are either false choices (e.g. only one winning option) or too many confusing choices, making the game frustrating and difficult (or too easy) to play.

There are a few things to consider when designing a game that I rarely see game designers talk about, but they're important.

What is Economics, and how does it shape games?

Google defines economics as "the branch of knowledge concerned with the production, consumption, and transfer of wealth" but that's not necessarily a perfect definition.

Economics is the science of making rational decisions with resources.

When we talk about game mechanics (as opposed to game narratives or game psychology) the focus is on looking at how the various numbers or systems behind the scenes of a game encourage people to make certain decisions.

For instance, let's take an example of a game where you can invest points in four distinct attributes to improve a character. Even if each has a specific role in the game (like making the player more powerful, letting them use special abilities more often, or increasing their reward payouts for success), they can still translate to an increased quality of outcome for players' characters.

Or to use another example, there could be a game where players have the opportunity to make a large "combo" for lots
of points but with the risk of losing everything, or gain points in small clusters.

Even in a narrative form game, there are still economics: any decision that the player makes will remove the opportunity to pursue other paths (unless the game has no interactivity in its plot).

Whatever the desired goal of the player is will shape the path that they take while pursuing these goals, and the ideal objective of a game designer working with mechanics is to create a system where those decisions can be made meaningfully but without confusion.

Opportunity Cost

One thing to note about games in general, and role-playing games with heavy numbers-driven gameplay in general, is that any system where players are forced to allocate from a set pool of resources (even if this pool can expand over time as players improve their characters) makes every choice have an opportunity cost. For example, the classic video game Deus Ex has a swimming skill–something that works well in, say, and Elder Scrolls game where characters advance while they are doing things, but not so well in a game where you spend points that could be used on combat abilities or the ability to bypass obstacles on swimming. This is especially painful since alternatives exist to most uses of swimming, and novices to the game in particular tend to gain little from the ability to swim and find secret areas because they aren't familiar with those places.

That's not to say that every decision has to be equal, but rather it should be considered to be part of a floating spectrum. D&D requires characters to be built using classes, and in a typical game a character can't have more than 20 levels of classes. The most powerful abilities usually come at high class levels, and a character who switches classes after taking a few levels to pursue some particular objective needs to make sure that while doing so they are not making decisions that are going to permanently cripple themselves. The video game Dark Souls also has this; characters are likely to gain a relatively finite amount of levels, and many pieces of gear and the magic systems require characters to be very deliberate with how they spend the attribute points when they level up to be fully useful.

In both of these cases, there are typically few poor decisions, so long as players have a goal in mind and are building their characters to pursue those goals. However, they are giving up other things; they cannot do everything.
In social games, like a MMORPG and most tabletop roleplaying games, or games where there is a large amount of flexibility in how characters interact with the world (roguelikes and tabletop roleplaying games spring to mind) you want to have these opportunity costs because they keep the gameplay interesting.

Nobody can do everything perfectly, so they need to tackle challenges with the skills that they're good at, or find someone else who can.

However, if your game is going to be played by one person and they need to have an opportunity to experience a full range of gameplay, you don't want to lock out mechanics to them unless you have a good reason. Some reasons could be things like encouraging players to replay your game, which should be a decision you make based on narrative paths rather than strictly mechanical ones, or games in which the total run-time is short (which also means that players are expected to replay the game).

One potential antidote to this is to allow players to have a broad reach in defining the scope of their interaction with the game. This is difficult in a digital medium, but for tabletop roleplaying games like the ones we work on, it can work very well. In Hammercalled we do this by allowing skills to be written as IAAT statements instead of being chosen from a list–it means the players can think of things we can't, and that characters can be custom-built for settings (and vice versa).

Risk and Reward

One thing to consider as well is that games should have risk and reward built into them. Playing a game is a process of experimentation, and players should be aware of the changes they are making in the game's mechanical state (as well as narrative changes that occur through play, if the game has a narrative attached).

Having a good knowledge of what is possible and what is likely is key for this. If I know that I have a very small chance of surviving an encounter in D&D, I am likely to veer away from it.

However, that also ties into another quick point: whether or not an action is interesting is often tied to risk. If there is no risk to an action, it becomes pointless. Even "casual" puzzle games have some constraint on the player's actions, and in the example above the risk I would be taking would be character death, which is at the least a burden for me as making a new character, but could also mean that I'm not able to achieve whatever I set out to achieve with that character.

Encouraging players to really make decisions and have those decisions be meaningful is at the core of quality game mechanics. It is one thing to choose to be effective (beware of opportunity costs that leave only one good option), but it is entirely another to choose to be effective when there is some chance of failure involved.

Card games are a great example of this: the relatively simple game of Blackjack is set up as a risk and reward experiment. From that a whole school of theory and practice has sprung up to calculate exact chances and figure out what winning plays are, and in some cases it can be the only game of chance that is likely to yield a profit at a casino (though this requires some mistakes on the behalf of the casino and some effort on behalf of the player).

Wrapping Up

Games are decision machines. They allow us to experience "play", the way in which we explore our environment, with a relative amount of disinterest.

The job of a game mechanic is to push players toward making decisions based on the environments that surround them. They should be embedded or integrated into narrative elements, when possible, but even when they are not they serve as a means of directing behavior.

A good designer uses these in a savvy way to get every last drop out of them.