Describing combat is to convey the emergent result of many game complexities. Though mechanics, variation, design space, and balance are all gears in the gameplay engine, how they fit together to shape the emergent whole is a separate more complicated matter. And though discussing balance does consider emergent gameplay, it does not actually convey it.
To help describe combat we'll need two new terms/concepts:
Phases of Combat
This is an indispensable concept that is commonly overlooked when describing combat. A phase of combat is defined as the set of conditions greatly shaping potential strategies and playstyles that all parties are affected by equally. Phases are usually long (in video game standards) taking 30 seconds to minutes at a time. They also generally do not include when one party changes the conditions unevenly for one or more parties (including itself). For example, knocking down the opponent in Street Fighter is not a phase. Pressing the advantage like running a contain strategy in Star Craft (keeping the opponent stuck inside their base) is not a phase either. Phases are also mainly used to describe changes in global battle conditions. After all, all combat occurs in at least 1 phase (set of conditions). The concept is best learned through examples:
Interplay Barriers
Have you ever studied a gameplay strategy that never worked "on the field of battle?" Have you ever practiced a powerful combo or any specific sequence of moves and never completely pulled it off in a match? One reality of combat is that you rarely get to just play "your game" against an opponent of equal skill or higher. With game systems deeper than Rock Paper Scissors, interplay (counters) is combat. The beauty of deep gameplay systems is that when one player pushes, the other usually is forced to respond in some way (pull). The disadvantaged player will try to minimize the damage or turn the tables. When one player continually pushes, he/she sets the tempo for the match forcing the opponent to respond and keep up.
For a simple example consider a Street Fighter match between Zangief and Sagat. Zangief is extremely dangerous up close. Sagat is pretty great at keeping opponents back and he's good up close. Zangief cannot do what he does best while Sagat attacks from across the screen with fireballs. Until Zengief finds a way to work his way into close quarters, he's very limited. So a good Sagat can play a smart fireball game keeping Zangief in a limited counter situation as long as possible. Notice how many times the Sagat decides to back away from Zangief.
So we can define an interplay barrier as a level of play or a counter strategy that one player can force another to use or suffer a significant disadvantage. Interplay barriers are similar to skill/learning barriers. After all, forcing a player to use a strategy or to step up their game is another way of forcing a player to prove their DKART skills.
Here are some examples...
You can think of interplay barriers as branching, hierarchical tests. When sizing up opponents interplay barriers are often used in a step by step progression. If you pass the first test, I'll move on to the second. If you pass the second, there's always a third. The tester in this case is the person who asks the questions or sets the tempo. Sometimes, the way an opponent responds with their counter strategy turns the tables on the tempo of a match. So the tester or the tempo setter gets no free passes.
In part 3, I'm looking closely at how human players complicate gameplay in the best way possible.