Surviving cEDH
Posted on February 14th, 2024, and mostly not updated since then.
One of the most interesting formats to emerge over the past few years is that of competitive commander, or cEDH. Unlike Duel Commander a.k.a French Commander, cEDH is a multiplayer format, using the same banlist as regular, casual commander, but being run at Competitive REL with high prizes at stake. It's become quite popular, possibly eclipsing Legacy in total player count.
For a long time, my only knowledge of the format was limited to thirdhand accounts and online drama. It has no official judge program support, I never went to a convention that had a cEDH event, and almost all of the experienced TOs and judges tended to stay away from it, describing it only as "cursed".
This always struck me as a little off; our role as judges is to help players have a fun Magic experience, and who are we to tell players what formats they should and shouldn't enjoy? Just because a format is challenging to run doesn't mean we should refuse to support it if that's what the players want.
So when I was offered the chance last week to head judge a ~120 player cEDH event offering a Timetwister as 1st place prize, I was happy to accept. As I now write this article in recovery from that weekend, I can safely report that it is, indeed, cursed.
My overall opinion however remains unchanged. Thousands of players enjoy this format, and as judges, I think we should do what we can to accommodate them. Magic can't live without new players joining the community, and right now cEDh is one of the biggest sources of new blood, bringing in not only new players, but many new TOs and judges as well.
And for the experienced judges, cEDH provides a novel challenge that I expect many (myself included) to enjoy. The way we run two-player tournaments has been honed to near perfection in over 30 years of Magic history; cEDH is the wild west, open to innovation and in desperate need of taming.
So I write this as a hybrid tournament report and guide on what to expect from future cEDH events, hopefully allowing people to be a little more prepared than I was. (I'm not covering any event-specific issues we had, just talking about cEDH in general.)
Deck style
cEDH is the only format where the stack is one of if not the primary zone in which gameplay occurs.
The way the format works is that the first 2-5 turns of the game are spent setting up, with the players getting as many cards in hand, useful permanents on the battlefield, and open mana as possible. Then one player tries to go off with an infinite combo while the other three players try desperately to stop them. Repeat this process until someone successfully combos, or someone dies to chip damage from utility creatures.
Most decks are playing every 0 and 1 mana counterspell and swerve effect they can play, another with other cheap interaction like Endurance
It's basically legacy, but every player is simultaneously playing storm and control and has 12 cards in hand at all times.
This means you really need judges who can actually resolve a complicated stack. I've written before about the difference between heuristic and algorithmic understanding of a topic. In normal two-player tournaments, judges can generally get by with only a hazy understanding of how priority and the stack work, since the details of either one don't come up all that much, and their intuition will lead them to give the right answer in most cases.
In cEDH, that is not going to cut it. The judges need to be able to come over to a table at which all four players each control Trouble in Pairs and one of them has just drawn their second card of the turn, pull out a piece of paper, and walk through exactly what order everything happens in. (Keeping in mind that one of them could want to combo off at any point in the middle using Borne Upon a Wind to give all their mana rocks and tutors flash.)
Some decks play permanent disruption too. Teferi, Time Raveler, Grand Abolisher, Grafdigger's Cage, Torpor Orb, Blind Obedience, Sphere of Resistance, etc. There are 4 player's battlefields to pay attention to, so these are easy to miss. Maybe it's more like legacy combined with judge's tower.
Multiplayer dynamics
In two-player games, the players are always enemies. This keeps the game theory of the situation nice and simple: try your utmost to kill your opponent, and they will do the same to you.
In multiplayer, things are much more complicated. Alliances are made and broken on a dime, and players are often in a situation where they want someone else to do something that they could do themselves, in order to conserve their own resources.
The biggest issue is not the game theory though, but just the fact that there are more people involved at every point. Decisions take forever, there are way more permanents to pay attention to. It's easy for the players to lose track of the current state of the game, perhaps being in a conversation with another player about strategy while a third player plays through their turn. Miscommunications about whether a spell has resolved yet are common.
This means that many more infractions are committed per match than in a two-player tournament. Most of these the players fix themselves, without judge intervention. And some TOs change GPEs to upgrade on the 4th infraction rather than the 3rd so as not to be too harsh.
Time limits
Most TOs use one of two different EOR procedures: when the time limit is hit either the active player finishes their turn and then the game is a draw, or the active player finishes their turn and then there are additional turns equal to the number of players remaining in the game. (Usually with 75 minute rounds, sometimes a little more.)
Neither of these is acceptable in my opinion. We used the first option, and most rounds still went at least 25 minutes over. We had a single turn last for 49 minutes after time had been called, when three players all had Consecrated Sphinxes that they used to draw their entire libraries, and then one of them tried to use every card in their deck to combo off while the other two players used every card in their deck to try to stop them.
One alternative would be to do something like 90 or 100 minute rounds, and once time is called, the game is a draw immediately. The expected amount of time being used per round is about the same, but a hard cutoff ensures that long games can't mess with your scheduling, like running out of time to get all the rounds done in a day.
The downside of this option is that the three nonactive players are strongly incentivized to stall on the combo turn in the hopes of getting a draw. There's also a risk of surprising players who hadn't been paying attention with a sudden draw. Neither of these is ideal.
The surprise issue can be addressed with 30/15/5 minute warnings. The stalling issue is harder to solve, but even the EOR option we did use, of "finish the active turn and then it's a draw" still has strong stalling incentive, since it's often the case that a player taking the last turn of the game can try to combo off, so the player right before them in turn order would really rather not let them have that turn. This dynamic led to several heated debates from the players about exactly how long their extension was and exactly when time was called, since this was going to decide the game between a win and a draw.
Another option that some TOs have been trying in Europe is to use an extra turns option, but cap those turns at 20 minutes, which an immediate draw if they can't finish the turns in time. I like this, though it's effectively just another way of implementing a 20 minute warning.
The more complicated but theoretically best option is to use individual clocks for each player.
The general consensus among the judge community for many years has been "chess clocks bad", summed up in this article. It gives three main reasons why chess clocks would be a bad idea.
First, that chess clocks are expensive, they would be stolen, and they're a huge pain to distribute and ship. This was all true in 2009.
Secondly, it claims that clocks lead to stalling and angle shooting concerns, where a player forgets to hit their own clock and the opponent now wants to waste time. This entirely misses the point of chess clocks. The idea of individual clocks is that it turns the time into a strategical consideration rather than something judges have to subjectively enforce via slow play penalties. This subjectivity leads to all sorts of bad feeling and accusations of bias from the judges, as we have to police what amount of thinking time counts as "reasonable". With chess clocks, if a player forgets to hit their clock, that is a tactical error that the opponent can absolutely take advantage of. The player will quickly learn not to do that again. Skill issue.
The last reason it provides is that priority in Magic is much more complicated than in chess, and passes back and forth more frequently. This is true, and a legitimate issue. But in cEDH specifically, confusion and arguments over who has priority at any given time are already a big problem. Given the prevalence of countermagic and the relevant of who in particular is going to cast that countermagic, the better cEDH players tend to use explicit priority passes after any important spell is cast anyway. So I'm inclined to say that forcing players to learn how priority works and pass it explicitly wouldn't be as much of a change as it seems, and would be an overall benefit to the tournament scene even without the time savings. (The MTGO players have successfully been using explicit priority passes for 20 years. It's somewhat automated, but still shows that having to press an "ok" or "f6" button every time priority passes is not all that disruptive.)
One problem with chess clocks that doesn't exist in two-player is the existence of table talk. Alice might want to ask Bob a question, and if Alice's timer is ticking down the whole time, Bob now has an incentive to talk like a sloth at the DMV.
Adding individual clocks to paper play would certainly be a big change, and there would be a learning curve. But cEDH is a pretty new format, its players are new to Comp REL, and I think it could succeed with a coordinated push.
Software
The defining feature of cEDH is the fact that its games are 4-player rather than 2-player. This obviously completely changes the pairing algorithm. Neither Eventlink nor MTGMelee supports 4-player games, but luckily some other programs do. I used Command Tower for my event, and it was great. There were a few minor bugs here and there, but overall it feels about as polished as Eventlink, perhaps a bit more.
It doesn't allow for manual pairings yet, but I'm told that functionality is coming just a few weeks after I post this in February of 2024. One of the developers of the software was a player in our event, and they were super helpful; there was a bug that kept causing one player's name to be set to the empty string, and they were able to go into the back end and fix it in between matches. Command Tower comes with online pairings, the ability to import players from other software, to submit decklists online, to post a public event page that players can search for and see the event details, and pretty much every other feature I'd want it to have. It was well worth the $12.
The other multiplayer software options I'm aware of are Squirebot and Best Coast Pairings. I've used BCP for two-player events and it's quite solid as well, though it was designed primarily for non-Magic games and the terminology it uses may be a little unintuitive for Magic judges.
(Do note that Command Tower is closed-source and declines to provide the details of their pairing algorithm, which may frustrate players trying to do ID math.)
Tournament structure
In two-player events, the match winner gets 3 points, and a draw gets each player 1 point. This ensure that a draw is better than a loss, while making it so that the expected value of playing it out is still higher than agreeing to an intentional draw.
In cEDH, a player's a priori chance of winning each match is 25% rather than 50%, so the point structure necessary for the same incentive structure is that the winner gets 5 points, while a draw gets 1. This means that a win in cEDH is much more impactful than a win in two-player. My event had 6 rounds of Swiss on day 1 followed by a cut to top 40 for day 2, and some of the players who won round 1 simply drew the next 5 rounds in a row, guaranteeing them 10 match points at the end of the day, more than enough to make the top 40 cut at 7 points.
There are various ways to try to fix these incentives. What our tournament did was give the top 8 players after Swiss a bye in round 1 of day 2, encouraging people to actually play out the Swiss. (32 people play round 1, leaving 8 winners. Those 8 combined with the 8 who got a bye make up the quarterfinals
Some tournaments use more complicated point systems, such as awarding some number of points to second place, or giving points for eliminating other players. I haven't tried any of these, but I would be pretty hesitant to do so. Most cEDH games end with a combo killing the whole table, so there often isn't going to be a clear second place. And having the order of elimination matter leads to some weird incentives.
My event did have something similar, in that different prizes had been announced for each of 1st, 2nd, 3rd, 4th, 5-8th, and 9-16th places. This meant that in each of the 4 quarterfinals matches, there were going to be 3 losers, but one of them had to be crowned "second place". We ended up using the rule that if players are knocked out in order, we use that order, otherwise we go by their order in the standings after Swiss. This resulted in some really weird incentives in the finals, since the difference between third and 4th place was several hundred dollars, so players were encouraged to ensure specific other players didn't win even if that player lost. This is a terrible prize structure, never do this.
Awarding points based on player kills has its own problems, as the question of what counts as a player kill is very difficult to define. If I cast Approach of the Second Sun, do I get credit for 3 kills or 0? If I cast a pump spell on another player's attacking creature in order to kill the defending player, who gets credit for that kill? Granted I'm pretty sure not many people are playing pump spells in cEDH, but I'm not a huge fan of this. In practice I think it will usually lead to one player getting most of the points, since they'll kill all three other people at once with infinite mana and Walking Ballista or whatever.
Some tournaments also try to offset the disadvantage of going last in the turn order by awarding extra points to those players. In our event we didn't do that for day 1, and then in day 2 the turn order was determined by their standings after Swiss, as another reward for players who actually won games.
There's also the question of what happens if a player loses a game and then that game is a draw. If you don't give them the point for the draw, it would often be advantageous for the 3 players who are lowest in the standings to all gang up on the highest player, eliminate them, then agree to a draw between the 3 of them.
There's a lot of other finicky stuff too. If the number of players in a tournament isn't divisible by 4, you can either give 1-3 players byes, or have a 3-player or 5-player pod. If you do the latter, you may want to give the winner 4 or 6 match points respectively instead of 5. (Players will surely complain about this being unfair, but byes are always unfair in some direction, so we just have to pick the least bad option.)
Player experience
Two-player formats tend to draw from a pretty experienced playerbase. Most competitors have been to many Comp REL tournaments before, and gotten used to basic tournament structure and etiquette in events like FNM before that.
cEDH players, by contrast, seem to mostly come from a different background. Many of them are casual Commander players who kept making their decks better and better until it wasn't fun to play casual anymore and they needed a tournament. This means they're often unaware of stuff like "you should communicate clearly about what you're doing", "you should show up on time"
These are the sort of things that judges tend to take for granted, assuming that players will know them and not thinking to mention them ourselves. But in reality they're non-obvious, and if we don't make an effort to communicate these things to our players in advance, there's no easy way for them to find out on their own. I have a Comp REL primer here, which I plan to send out to players in advance of my next cEDH event. You're welcome to modify it for your own use.
Tardiness
Tardiness is a little weird, since you'll have three players waiting there on a fourth. What we did is have them wait for 10 minutes, then start with an extension. The lack of a player does make the game go a bit faster, so this isn't a problem for round time, but it does feel bad to make three players wait and then still have to play.
If the missing player does show up, a game loss is effectively a match loss, so most TOs use the penalty of skipping their first turn if the player is more than a minute late. This has a similar effect of being a meaningful penalty, though its severity is more deck-dependent than a game loss is.
One idea to speed things up would be to have the players start playing immediately, and any late players join the game whenever they get there, skipping however many of their turns have been passed by that point. This is elegant from a judge perspective, but I think it'd be undesirable from a player one; the other three players may have made mulligan decisions based on there only being two opponents. So I'm inclined to say that the "everyone just wait" solution is best, but perhaps this is worth testing out.
(Note that if two players no-show, or one player no-shows in a three-player game, the resulting game will only have two players remaining, in which case the question will arise of whether they still get a free mulligan and the first player skips their draw step.)
Deck checks
I was understaffed and under-timed at my cEDH event, which gave me an excuse to skip deck checks. But in an ideal event, they would happen; Cheating is just as much of a concern in cEDH as in any other format.
Checking a 100 card singleton deck takes quite a while, so I'd recommend not doing that. Check it for marked cards as usual, then check a random half or third of the deck against the list; whatever you can get done before the 8 minute mark. Deck checks don't need to check every card, they just need to have a reasonable chance of checking every card.
Games are 4-player, so you'd need 4 judges on the deck checks team to check every deck. That could work at a large event, you just check one match per round. But at a small event you probably won't have 4 judges available for deck checks, so you'd want to check fewer. I'd still take all 4 decks in the swoop, since taking only some of them would give the other players more time to make mulligan decisions, and lead to accusations of unfairness.
(Physically carrying 4 commander deckboxes is a challenge, so the swooping judge should bring a bag with them. Ideally one with 4 pockets marked 1-4 so you don't forget which box is whose.
cEDH matches are best-of-one, so there aren't going to be any midround deck checks. It's easier to slip cards into your hand or library without arousing suspicion in the middle of a 75 minute game; turns take a long time and players reach into their bags to get dice, tokens, etc. So in the 60+ minutes during the round when the deck check team would otherwise have nothing to do, I'd suggest doing noninvasive spot checks. Walk up to a table, pull up the player's lists on your phone
(In fact you could use this as your only method of checking decks, avoiding any delays at the start of the round. But that largely forgoes the ability to notice marked cards or counterfeits.)
Opponent choice penalties
In two-player events, the fix for Hidden Card Error is to have the opponent pick what cards should go where. In cEDH, most TOs go by either a majority vote among the opponents or a random opponent. Both of these options are bad.
Alice is in the process of trying to combo off. Bob cracks a fetch land and shuffles away his hand. Alice would like his replacement hand to be all land. Carol and David would like his replacement hand to be all counterspells. A majority vote will favor Carol and Dave. A random die roll has a 2/3 chance of doing the same. This means that it can significantly advantage Bob to shuffle away his hand, defeating the entire purpose of the HCE fix.
Missed Trigger has a similar problem. Alice could miss her trigger, Bob makes a decision based on that happening, then Carol and Dave vote to have it happen anyway, overriding Bob's objections. Same for Deck Problem, the GRV partial fix, and any other case where an opponent is given a choice under the assumption that they'll want choose whatever is worst for the infracting player. This assumption is false in multiplayer.
The idea behind all these fixes that have the opponent choose is that we don't want the judge making strategical decisions that could impact the game. This is a good idea in theory, since judges are sometimes poor at strategical reasoning, and even if their decision is good, could be accused of unfairness from a player who disagrees. But in reality, judges already make strategical decisions all the time: backup decisions, slow play judgement, advantage considerations in Cheating investigations, etc.
So I think the best solution is to just scrap that guideline. In multiplayer, have the judge make the choice that's worst for the player who committed the infraction. There are 3 other players who will happily point out any errors in the judge's reasoning, so game-impacting mistakes stemming from a judge's poor strategical evaluation seem like they'll be rare enough to be negligible. (In a situation where the judge is unsure what's worst and the players can't present a compelling argument either way, just pick randomly between the options that are tied for worst.)
Standardization
cEDH is currently a patchwork of different rulesets. There's the MSIPG and MSTR from Monarch Media, the cEDH IPG/MTR addendum from Topdeck.gg, the Competitive Multiplayer MTR/IPG Addendum from the @MTRA_IPGA Twitter account, the Multiplayer IPG/MTR addendum from the Portuguese judge community, and probably more similarly-named documents that I haven't yet found.
And several of these attempts at rulesets are actually multiple rulesets, taking the descriptive rather than prescriptive approach. (For example the MSTR provides two different EOR procedures, leaving it entirely up to the TO which one to use.)
This is not ideal. When thousands of dollars are on the line, players want to know what to expect going in. And some parts of the rulesets that some TOs are using are just bad, leading to a worse experience for players with no real upside. Having a single set of policy documents that are kept up-to-date with best practices and used by everyone worldwide would go a long way towards making cEDH just as "real" of a competitive format as Modern and Pioneer.
Ideally, Wizards would officially support cEDH, giving us a single set of policy documents to follow, adding multiplayer support to Eventlink, etc. But this seems very unlikely to happen. The EDH rules committee (which at this point functions as a de facto branch of Wizards) has been pretty clear that they think competitive EDH is against the spirit of the format, and they don't want support it. On top of that, Wizards has been pulling back their support for organized play in general, and even the regular MTR and IPG often aren't kept up to date for new mechanics.
So I think any standardization in the cEDH world has to be grassroots. Duel commander did this successfully. They have their own rules committee that publishes updates every 2 months, and pretty much everyone follows their rules and banlist. This is promising, and suggests that the cEDH community could in theory do this too.
The first hurdle is getting buy-in. This sort of thing is a classic collective action problem, where the community as a whole benefits from having a single standard, but individually it benefits people to do things their own way. The various TOs that are running events are for-profit businesses, so they'd be loath to announce that they're following one of their competitor's policy documents. And organizers are often players themselves, with their own personal preferences on how the game is played, and they don't want to listen to anyone else.
With Duel Commander, the standard is about something that players care about (a ban list), meaning they'll encourage TOs to use the same banlist so that their decks are always legal. With cEDH it's mostly behind the scenes stuff that's largely invisible to players, so most wouldn't care enough to push a TO into following some standard.
The second issue is that the things that need standardizing in cEDH are more complicated. Duel Commander just needs a ban list and a few simple rules like "the starting life total is 20" and "there's no commander damage". cEDH needs a complete overhaul of the MTR and IPG, and the filling in of several gaps in the Comprehensive Rules as well. This is not something that players or newer judges will be good at, and unfortunately the greater judge and TO community is taking the "don't touch it with a 10 foot pole" approach to cEDH.
Potentially as a result of this, the current cEDH community strikes me as kind of a mess right now. Monarch Media was one its biggest TOs in the US, publishing the most comprehensive policy documents, but then collapsed in October 2023, unable to pay their judges and artists. Most cEDH events aren't that bad, but they do tend to have a general theme of poor planning and disfunction. With most experienced TOs staying out of it, new ones are trying to fill the gap, but running a large convention is actually quite hard.
Similarly, the current state of rules and policy knowledge within the community is not great. Some of the most popular cEDH websites post articles saying things like "players can get priority while a spell is resolving" and "triggers can't be placed on the stack in response to a split second spell". This one purports to be a guide to common tournament issues in cEDH, relying on an interview with the head judge of some large cEDH events, but claims that Deck Problem and Marked Cards are the same infraction, claims that a Missed Trigger is a GRV, etc. A community whose tournament leadership doesn't understand the absolute basics of priority or Comp REL policy is clearly not going to be able to design competent tournament structures and guidelines. Things have gotten a little better since then, with some career judges being brought in to head judge more recent events, but there are still a lot of issues. The various rules documents I linked above contain several errors, omissions, and other very odd choices that will not lead to a good tournament.
There's also a lot of drama in the community about proxies. Most cEDH tournaments allow proxies, which is not that odd; many Vintage events do too. But proxy usage is so prevalent in the cEDH community that lots of players don't have any real expensive cards in their decks. And there are a lot of bad actors in the community pushing counterfeits as proxies and trying to confuse people as to the difference. (Topdeck.gg, one of the biggest organizers in the cEDH space, actually requires that players use counterfeits as their proxies.) These facts make vendors not want to attend cEDH conventions, and since vendors are one of the primary sources of funding for large conventions, this contributes to the difficulty of running quality cEDH events.
cEDH is quite new, so it's possible that it will sort itself out. Duel Commander, after all, has been around for more than 15 years. Any new format that doesn't have top-down support is going to be fragmented and disorganized, and if demand is there, I expect they can probably figure it out eventually.
There is another option though, which would be a lot faster: the judge program could do it directly. The actual amount of work involved in doing such a thing is relatively low; the barrier is experience and adoption. The judge program is low on resources right now, but high on experience and prestige. Judge Foundry
Looking at the history of both regular EDH and Duel Commander, it's interesting to note that its adoption was driven and rules committee formed largely by high level judges. This is probably not a coincidence. I'm sure Judge Foundry has no interest in running the format, like deciding on a separate banlist if such a thing is ever needed, but they would be a perfect fit for maintaining the underlying tournament policies.
But career judges right now are hesitant to get into cEDH. They risk not getting paid, getting blamed for all the bad decisions of the tournament organizer, and having to deal with a bunch of annoyed players.
So ultimately, I think the decision mostly lies with the cEDH community. If they want to be taken seriously as a format, they're going to have to learn how to play faster and communicate clearly, learn that someone has to pay for the convention center, hire people who actually know what they're doing to run the events, and be willing to follow a ruleset that they don't fully agree with themselves.
If they can do that, I'm really excited to see where the format goes. Despite all the issues in my event, I found it quite enjoyable trying to figure out how to best apply Comp REL ideas to multiplayer, and I think it has the potential to be a thriving Magic subculture. As terrifying as I'm sure most judges reading this find this sentence, I would love to one day have a cEDH RCQ season; it would be a great way to reach out to the millions of commander players in LGSs around the world and bring some of them into the tournament ecosystem.