Commander is the most popular format of Magic: the Gathering. It’s a social format that works with two, three or four players (sometimes more, depending on the group) politicking and chatting while playing with singleton decks built from a 27,000-card-deep pool. For how accessible the format may be, it has some problems that stem from its history as an unofficial gametype:

One of the easier solutions to all three problems is to overhaul how the banlist works. But with one of the deepest card pools of any format, “easier” looks like an overstatement. Commander needs to borrow from another semi-competitive, high-complexity game: Pokemon Showdown.

Pokemon Showdown is a competitive 6 vs 6 Pokemon battle simulator. Its competitive scene is managed by a community called Smogon University. Smogon has, over the course of about 20 years, developed a multi-tiered system balanced and maintained by volunteer veteran players. Smogon’s official tiers are: Ubers, OverUsed (OU), UnderUsed (UU), RarelyUsed (RU), NeverUsed (NU), and PU (a pun on the other tiers’ abbreviations). Other tiers have cropped up over the years. When Pokemon Alpha Sapphire & Omega Ruby came out, Smogon needed to make a tier above Ubers called Anything Goes (AG). Little Cup, National Dex, Doubles OU, and Monotype are competitive gimmicks that explore the depth of Pokemon’s systems and Pokedex. Every tier--except AG--has a banlist and a set of clauses.

The banlists are defined by two methods: usage statistics and committee decisions. Smogon has access to Showdown’s match stats, so a tier committee can see which Pokemon are used the most (this is where the tier names come from). A pokemon with about 5% usage in a particular tier is defined as part of that tier. The Pokemon has the proper moveset and statline to be competitive and effective at that level of play. It is definitionally banned in lower tiers when it reaches that usage threshold. For example, Salamence is an RU Pokemon by usage, so it is forbidden in NU and PU. OU players can put Salamence on their teams. Its usage stats are significantly lower in OU, however, because that tier has access to much stronger Pokemon than RU does. Tier councils do not have much of a say in what gets removed by usage stats, although a tier’s most popular Pokemon often get removed before then by the banning process. A tiering council can ban a Pokemon that is causing problems within its metagame by either voting within the group (usually referred to as a quick or emergency ban), or by asking skilled members of the playerbase to familiarize themselves with the issue via special ladder and then voting (they call this suspect testing). The council brings a centralizing or problematic monster forward as a suspect to be interrogated through community discussion and focused play. Players who surpass a rating cutoff within the suspect ladder get the ability to vote for or against that monster’s banishment. If a Pokemon is found guilty, it is moved to the tier’s banlist. Pokemon on a tier’s banlist are eligible for use in the next highest tier’s teams. Pokemon banned from OU go to Ubers, and the Pokemon hypothetically banned from Ubers go to AG. The same systems exist for moves, abilities and generational gimmicks (Dynamaxing was suspected and banned from OU at the start of the Sword & Shield era, which also banned it from all lower tiers). The tiers balance themselves through usage statistics and community-backed bans. If a player wants to use their favorite Pokemon, they can play the tier in which that Pokemon thrives.

The clauses are complex bans that cover more than one move, ability or strategy with a single rule. Smogon has seven general clauses that apply to the official tiers:

Other tiers have additional clauses where necessary. For example, OU has three additional clauses that affect what Pokemon are available in the tier and how they appear in the simulator. The Ubers council created a clause banning Rayquaza from mega evolving in the Alpha Sapphire & Omega Ruby metagame. This change in policy led to the creation of AG as a format where interested players could use Mega Rayquaza. AG only uses the Endless Battle Clause; otherwise, it has no restrictions on team building. Sometimes clauses are removed, such as when Moody was changed in Sword & Shield, leading the OU council to remove the Moody Clause (they ultimately banned it anyway via quickban). Metagame councils have a say in how clauses are written and enforced through policy review, a system by which intertier discussions can produce large changes to the game and how it is managed. Bans on some moves, like Baton Pass, act like clauses when they affect every official tier. Clauses are also how tiering councils get around the stigma against conditional bans, where a move or ability is banned if and only if it is on a certain Pokemon or paired with another specific move or ability. Clauses take more time to create and implement than bans, and any metagame community has little to no say in the process. Moody was banned by clause when it was banned from Ubers in 2011. That process was done by suspect testing. The Evasion Clauses have been tested several times in Smogon’s history, and each time they’ve been found necessary to maintain the game. Clauses are complex bans that stabilize more than one metagame at a time, at the expense of limited player input.

Commander, in Smogon’s framework, looks like this: a format with a Singleton Clause, Commander Clause, Color Clause and Life Clause; one banlist for all levels of play managed by a rules council and an advisory council; and only one tier despite the huge variety of options available to players. The clauses are fine, since they work like additional rules that make Commander different from other formats. The single banlist does not seem to work, and the single tier hampers both competitive and casual play by pushing them to mix. Smogon’s tiering system can fix these two problems. Wizards of the Coast could establish five tiers, each with its own banlist. Whatever is in the top tier’s banlist would be permissible in an unrestricted tier above the rest. The presence of one restricted card in a deck would put it in the highest legal tier. For example, Jeweled Lotus is an OU card, so its deck would be rated OU, even if the rest of the deck is janky disharmony. This is true in Smogon metagames: a team with Mew and five basic-stage starter Pokemon is still an OU team, even if it’s not viable in that tier. A deck could have more than one banned card, in which case it would be in the highest tier that permits all of its banned cards. For some strategies, that would place it in Ubers, especially if it is a high-performance deck running the Power Nine. Each tier would have its own banlist, defining what’s allowed in the tiers below and inheriting the cards from the tiers above. A hypothetical PU tier would have a Pauper Clause limiting card rarities to Common only. That tier would still need to ban powerful Commons like All That Glitters and Lightning Bolt. The NU tier above it would have access to both of those cards, as well as all of the Uncommons. But maybe, in this hypothetical structure, Swords To Plowshares is too strong in UU, so it gets banned. It would then also be prohibited in RU, NU and PU. Again, these bans would define the tier of a deck regardless of its strategies and synergies. A flavorful Tawnos, The Toymaker deck with only birds and beasts would still be beholden to whichever banlist has Nadu, Winged Wisdom. A tiered banlist, rather than one monolithic banlist, would have three effects: it would establish a high end for competitive play; it would define deck power levels ahead of casual games; and it would encourage creative deck building.

High level players would self-select into groups and pick-up games where the other decks target specific high tiers. Those games would be as fast and intense as the participants like. Casual players wouldn’t be bamboozled into agreeing to get pubstomped by someone with a mild-seven that doesn’t always work. Competitive players who want to play with real cards can build for the tier that bans the Power Nine and other strong Reserved List cards, and their opponents will be equally limited. Proxy-friendly games could take place at the unrestricted level with unaffordable dream decks (which is an ethos in line with the Pokemon Showdown simulator). As a corollary, intentionally weaker games wouldn’t run the risk of rogue Free Spells or Rhystic Studies. Players targeting lower power levels only need to substitute banned cards for legal ones to play at that tier. Competitive players would have more choices of how and with whom they wanted to play Commander.

Players would have easier Rule Zero conversations at Commander nights in game stores and around kitchen tables. A pod can agree on a tier that they all want to play ahead of time, without getting into the weeds of what a mild-seven that doesn’t always work really is. A group doesn’t even need to have a Rule Zero conversation, as long as they agree on which tier they will play at. Each deck’s power level wouldn’t be defined by the subjective and nebulous 1-10 system many players currently try to use; it would instead be defined by the strongest card in the deck. The mono-blue turbo draw player needs to play at the appropriate level if a majority of the pod wants to play at a tier that excludes Rhystic Study and Mystic Remora. They could still play the deck with a few substitutions. A tiered banlist wouldn’t invalidate Rule Zero conversations. It gives players a framework for determining appropriate play styles and exciting matchups within consistent Commander groups. Part of a Rule Zero discussion could be moving a powerful budget deck up one tier. It also gives a pod the ability to ban or unban cards piecemeal to fit their local metagame, without removing the cards from the playable pool completely. A hypothetical Pauper Clause PU group could unban all of the Commons to make a higher power metagame. A similar group with different preferences could ban Commons that were ever printed at higher rarities. The tiered banlist would allow casual groups to better tailor the game to their preferences.

Players would be incentivized to get creative when deck building to fit their local metagame. As I pointed out before, a Little Cup team with one Mew is actually an OU team, regardless of its performance. Players who want to add a few high-power staples to fairly weak or inefficient decks will need to expect to perform at the appropriately higher tiers. If they don’t want to play against the agile and aggressive mild-seven that doesn’t always work, they would need to swap the high-tier staples for lower, more appropriately tiered alternatives. The tier system would push players to exclude good cards from bad decks, and to remove bad cards from good decks. The inverse of the Mew team example is an OU team with one much lower-tiered monster. The player can replace that weaker Pokemon with a stronger or more functional monster. If they want to play with their favorite low-ranked Pokemon, they can build around it at the appropriate tier. Low-tier pet cards encourage building within the limits set by the banlists. High-tier pet cards encourage building synergistic decks that show the best aspects of the pet card. Less enfranchised players who rely on netdecks and databases like EDHREC would have a better idea of how powerful a decklist is before they get to play it by looking at the deck’s highest tier. They may not want to build that wicked Ur-Dragon deck if it includes cards of a tier to which their friends don’t have access. It would also prompt players to look outside of lists of common and widely-played staples for less-restricted similar cards. All together, card and deck variety would increase to meet the variety of tiers available. Commander players would get more creative, and sometimes strategically better, at deck building with more nuanced restrictions.

This system is not a perfect fit for Commander. First, the Rules Committee would be in charge of the official tiered banlist. Unlike Smogon’s system, players would have almost no influence on what gets banned or unbanned. Ban decisions would be made the same way they are for the other formats. The most accessible usage stats for Wizards are official tournament results, but there are currently no official tournaments for the format. The current closest thing to a sanctioned tournament is WotC’s offer to sponsor prizes for the cEDH European Tournament sometime in the future. The initial banlists would be based on the existing list, and the RC would need to make additional bans in the future based on either community anecdotal input, the metrics from unofficial organized play, the judgment of the RC members, or a combination of these. Second, there is no clear distinction between tiers of play currently. The RC would need to invent and define tiers by intended gameplay and not existing patterns of play. The 1-10 power scale is useless in the endeavor, as it is ill-defined noise based on individual player opinions. They could solve this by defining tiers by average match length in turns, but that can be as dependent on the players as it is on the cards. The starting point would be to make a distinction between competitive Commander and casual Commander, and anything banned from the competitive scene would be the Uber-equivalent top tier. Further defining the casual tiers would be difficult, as not all casual decks are equally slow, grindy or weak. Third, the RC would need to choose early in the process how to ban commanders. Currently, all banned legendaries are excluded from the command zone and the library. Banning a card as only the commander would be a complex ban, similar to Smogon’s clauses. Would these clauses be tier specific? Would they be modular the way Smogon’s clauses sometimes are? Could the RC make the inverse ban, where a card can only be a commander? The easiest solution to this problem is to avoid complex bans, which is already a policy position within the Smogon policy review system (they make complex bans as a last resort). But that brings up a fourth issue, which is that a tier might need to remove a combo that uses three or more cards. In that case, the RC can either make a complex ban on one combo configuration, a clause barring infinites, or a simple ban of one combo enabler. How to do this, when to do it, and why to do it are tricky questions. With the huge size of the card pool, high-complexity interactions will be hard to remove or prevent. These are conflicts between Magic’s design and Smogon’s system that need to be solved before the Rules Committee can make any meaningful progress on a tiered banlist.

Commander has much to gain and little to lose by adopting a tiered banlist similar to what Smogon’s official metagames use. It would better define the function of the banlist as a series of limits on intended or implied playstyles. These limits would separate competitive and casual play in a way that is conducive to all players having the fun they seek by playing Magic. It would help players communicate those expectations to each other, and it would set expectations by official rulings for players who do not want to have Rule Zero conversations. All players would have a better understanding of what is expected at each tier of play, and they can better curate their cards to fit their preferred style and power level. Establishing and maintaining a tiered banlist comes with challenges. Those challenges have potential solutions, and the group responsible for solving these problems has the resources of the game’s publisher, Wizards of the Coast. Building and establishing a tiered banlist would take time, and the work would never stop. It would keep Commander a living format, which is better for the game than letting its current small banlist stagnate.