The word "meta" gets thrown around a lot in competitive spaces, but it's rarely defined. Players use it to mean everything from "the best choices" to "what strategies everyone is employing" to "the current state of the game." Recently, the backronym "Most Effective Tactics Available" has become widely adopted, which hampers the ability to think strategically about metagaming.
Here, I lay out a framework that identifies four distinct layers of competitive strategy. Each layer operates on a different object and asks different questions.
This is the base reality. The physics of the game. What you're allowed to do.
L0 is defined by the game's code, full stop. It is not strategic, it simply is. The question L0 answers is: What moves exist?
Every layer above L0 is emergent from it. Change the mechanism, and everything above shifts. This is why developers making even minor balance changes has such far-reaching effects on the competitive community.
The first true metagame layer. This is what most players mean when they say "the meta."
L1 asks: What works? More precisely: given that I could face literally anyone, what choices are most effective knowing that all of my opponents are thinking the same thing?
The key insight is that L1 operates behind what Rawls called the [[Veil of Ignorance]]. You don't know who you'll face or who will be on your side. You don't know their habits or their choices. You don't know if your teammates will follow up on the openings you make or leave you hanging. You optimize against the aggregate: the population distribution of strategies you expect to encounter. Notably, this means that different skill levels often have different L1 optimal strategies. Difficult-to-master strategies often are ineffective at lower skill tiers.
This is where tier lists and "meta reports" live. When someone says "Pencil is S-tier," they're making an L1 claim: against the average opponent playing the average strategy, Pencil can reliably return a high amount of value.
L1 rewards theorycrafting, understanding the game's possibility space, and correctly modeling the population. It is fundamentally analytical at its core.
Competitive communities often moralize L1 choices. "What a camper!" "Play something difficult for once!" "That character is cheap!"
Gamers often call this "dishonorable", but this is confused. L1 isn't about ethics or honor, it's about effectiveness. Still, the source of the moralization is worth examining. Below is a non-exhaustive list of sources of moralization: