"...a person can go through a lengthy trial and error process and eventually arrive at something decent. Or, if they understand basic principles of theory behind the work they are engaged in, can save themselves a lot of time and compose/design much better works a lot faster than if they were ignorant of those principles."
That premise captures why I am comparing game design to music as I find the same types of dynamics in the creative processes of both fields. In music, one sketches out a theme and then creates a textured structure to help support that theme with many interworking lines and parts. The various concepts of music theory help a person do this a lot faster than if they were relying purely on trial and error. In game design, a person can do the same thing and save themselves a lot of unnecessary trial and error in the design process. There will of course still be trial and error - just not unnecessary trial and error. (ex. I don't have to waste a lot of time trying to figure out why my musical composition isn't working if I'm already aware of how voice-crossing and parallel fifths tend to create problems and how parallel thirds tend to work much more smoothly.)
So, the "Tri-Level Resource Exchange Model" was, in essence, a game analysis pattern that I arrived at after studying a number of specific games - looking for their common traits (just like how a person begins to understand how similar construction patterns occur in many types of musical pieces). Now, I want to look at a different model, the "Tri-Stage Resource Transition Model":
For a larger view of this image, click here and then zoom in.
The three main phases are Phase 1 or "Dormant", Phase 2 or "Ready" and Phase 3 or "Active".
The idea behind this model is that your choices don't involve how to convert one or more Level 1 resources into some sort of other types of Level 2 resources which are then converted again into Level 3 resources. Instead, your choices depend on what phase needs a resource transitioned into it from a previous phase more than any other phase at that particular point in the game. Here is an example:
In the card game "Loot", your cards are your resources. However, you aren't trading in different types of cards together in various combinations to receive secondary types of resources like in Settlers. Instead, you are making a choice on your turn of whether or not you want to move a card from the "Dormant" Phase 1 (which would be the draw pile) into the "Ready" Phase 2 (which would be your hand) or whether you would prefer to spend your turn moving a card from the "Ready" Phase 2 (your hand) into the "Active" Phase 3 (by playing it on the table).
The choice of moving a resource from Phase 1 to Phase 2 I have labeled as option "A" and the choice of moving a resource from Phase 2 to Phase 3 I have labeled as option "B". The greyed out boxes represent resources that were already in that phase before a player's turn.
Now, before going further, I want to point out that most traditional card games actually don't fit into this "resource transition" model I'm describing here. This is because, in most traditional card games, you are doing both Option A (drawing) and Option B (playing) equally - all within the same turn and usually in that order (not always - but usually). There is no choice involved in whether or not you want to take Option A over Option B. Instead, they are simply compulsory "steps" in your turn.
Whereas the tension is a Tri-Level Resource Exchange Model game tends to come from the choices of having to figure out "apples and oranges" types of situations in the conversion processes of which resources should be exchanged for which others, the tension in a Tri-Phase Resource Transition Model game tends to come from the fact that you can only do so much on your turn and you have to figure out which transition or transitions are more necessary at that point in the game.
Before going on, an important component in this Tri-Phase transition model is that Option B involves two or more options for "where" or "how" you could make a resource "active". For example, in Loot, if you want to transition a card from "Ready" to "Active", you then have to determine whether or not you want to make active a "ship" card (by playing it in front of you) or a "pirate" card (where you play it on another player's ship).
Let's look at another example:
In Tikal, you can perform both Option A and Option B in the same turn. However, unlike traditional card games, these options aren't required steps on your turn. You could spend your whole turn on Option A by transitioning a bunch of guys from their "dormant" phase into a "ready" phase on one or more of the camp sites if you wanted to. At the same time, you could also spent your whole turn transitioning guys from their "ready" phase at the various camp sites into an "active" phase by moving them to various areas up from grabs on the board. However, how much or to what extent you want to invest in Option A and Option B on your turn is up to you.
So, in essence, this type of model is not about trying to deal with "apples and oranges" types of situations. Instead, it boils down to the dilemma of how often or at which points in the game you deem it more necessary to get resources ready to go into action versus activating resources on the board to try and secure an advantage. To borrow a cliche, with this model you are essentially trying to decide whether to "fish or cut bait".
My "Heavens of Olympus" game utilizes this model. In the game, you are either moving planets from their "dormant" phase into a "ready" phase by paying to have them made (i.e. forged) for your turn, or you are moving them from their "ready" phase onto the board to receive payment. The trick is that doing the same thing as another player carries an added cost if you both choose to do it on the same turn. There are other aspects to the game, but it fits strongly into this model I'm describing.
In these three articles about "music" and game design, I've touched briefly on some generalities or "models" of design that I've noticed in various games. Though there are more than just these two main models I've discussed, I point these two out because they have distinct differences from one another while the games that utilize them share some interesting commonalities. I highly suspect that the game examples I used were not designed while necessarily thinking about these models I'm describing. Instead, my suspicion is that the designers were simply trying to design good games and hit on some game play dynamics that worked. The thing is, by studying these models, if one is aware of the generalities that have proven true in already successful game designs, one can then begin to pick up on potential weak areas in one's own designs more quickly and one can articulate exactly what the problems are more clearly.
Just like how studying the common trends in various Mozart compositions can help a person learn about fundamental principles of good music composition, studying the common trends in various Knizia or Kramer designs can help a person learn about good game design.