Metagame Meditations pt.3
Sunday, January 30, 2011 at 5:09PM
Richard Terrell (KirbyKid) in Dynamics, Interplay, Metagame, Misc Design & Theory

It's important to understand that in actual competition, the metagame play level can shift up and down. All players don't play at their highest level right off the bat. It's common to see players ease into combat and step up their game as necessary. A good reason for doing this is that playing at a high level generally requires more skill, and this can be very stressful. If you don't have to play at level D to win, then why bother? Also, keeping your best strategies a secret until needed is a great way to preserve their effectiveness. If two players are playing at level D, let's say, sometimes a player will use a strategy from level C. Though this move may be much riskier than a strategy from D, it's still up to the opponent to recognize the shift and counter it.

It's completely possible to play well at level D and not be so good at levels A, B, and C. This is common for players who enter a competitive scene of a game with a well developed metagame. Such players learn the higher levels/lessons quickly because everyone else is playing at that level. At the same time, they get less exposure to previous metagame levels. When a player plays in such a way to test the ability for an opponent to counter their strategies at a particular metagame level, as opposed to just assuming they can, this is called keeping the opponent honest. It's the equivalent of calling an opponent's bluff. 

The experts keep you honest. They remind you, "That was not a safe move. You cannot trick me with that. That will not stop my advances." ~Sirlin. Playing to Win.


Furthermore, competitive games are generally designed not to have a dominant strategy (or a small set of dominant strategies compared to the emergent scope of the game). So even games with highly developed metagames aren't always played at a constant D for example. I call the revisiting or regression of a game played at one metagame level to a previous level down grading/shifting



Though not a multiplayer game consider this hypothetical example: Small Mario = metagame level A. Super Mario = B. And Fire Mario level C. Imagine how a boss battle would play out. As fire Mario, the fire balls are deadly projectile the player can use from a distance to defeat the boss. If the player gets hit, Mario turns into Super Mario. With no projectile, the player must figure out how to take out the boss with platforming. As Super Mario players also have the option of taking a hit to land a hit. But if the player suffers another hit, everything is put on the line for Small Mario. So though there's a powerful (perhaps dominant) strategy with Fire Mario, there is the potential for the metagameplay to be down shifted to deny use of this dominant strategy. Decay dynamics and other types of interplay help avoid dominant strategies while facilitating the use of a variety of metagame levels through down grading


The previously discussed RPS and Meta-The-Game examples are extremely simplified. With multiplayer games more complex than Pong, an accurate record of the evolution of their metagames can contain branches and loops in a more web/tree like structure. Additionally, if you record "normal" strategies (non-interplay barrier strategies), the tree will be more full with data.

In the image above of the hypothetical metagame of Meta-The-Game, each interplay barrier is demarked with a capital letter. Notice how the barrier E was developed after the metagame developed down to the D branch (indicated by lettering order). Sometimes players develop new techniques or strategies that branch off from a previous level.

Lesser counters are indicated by a lowercase letter based on which metagame level they were created under. Notice how these lowercase counters become less effective and more effective at different points in the development. For example, the strategy a1 is useful at the beginning (level A), becomes unused for the most of the rest of the metagame, and only becomes useful again at level D. Also notice how after all the development in the A-> B-> E-> G branch, all of a sudden the strategies of level A become an effective counter to G. I call these large-emergent-metagame loops ouroboros


What does the metagame look like of a game with multiple goals? Recall that multiple goals are hard to compare to each other in terms of success viability because they operate on different value scales. Fortunately, the inability to compare two strategies isn't a factor when recording a metagame. What we care to note is how combat is developing. That's it. So even if we can't directly compare two different strategies aiming for two different in game goals, we can note which strategies are actually working, which are being used more, and which is being used as interplay barriers. 

Because strategies of multi-goal games are difficult to compare against each other, there is a tendencies for branches of their metagames to develop independently or irregularly. This design tends to create trends in the metagame like how interplay loops work. The difference is, these trends aren't a clean "geometric" shape relationships of consistently weighted options.  


Let's recap...


If you're the kind of reader who frequents the critical-gaming blog, then you may be wondering where all of my specific examples are. After all, video game writers need to be specific. In the case of describing metagames, there's a problem inherent to the subject. Understanding and describing a metagame of very popular and/or complicated games generally requires high level play, hundreds of hours of study, and an audience who has comparable knowledge of the subject. This is hard work even for people as dedicated to games writing as I am. 

In part 4, I'll discuss what innovations we can make to recording and understanding developing metagames better. 

Article originally appeared on Critical-Gaming Network (
See website for complete article licensing information.