The Egg And Dragon Variant

Sheldon introduces the Egg and Dragon variant of Commander, chronicling from the initial idea to the final draft. Give it a whirl and let him know how it goes!

A few weeks back, tweeter Jake Fitch (@Milambus) suggested a variant that involves splitting your deck into two parts and playing two different phases of the game with them. There were a few loosely defined rules, but it looked like it had possibilities. I decided to run with it to see where it would take us.

I didn’t get to talk to Jake about his goals for the variant, but for me the attraction was threefold. First, it pretty much hamstrings combo decks since you can’t be guaranteed that your pieces are all in one deck or the other. Second, it creates a higher variance in the game, certainly cutting the value of tutors. Third, it simply has cool flavor.

Before we get to the details, I’ll tell you that this week begins a new adventure in my life. When this posts, I’ll be in my first week as an undergraduate student at the University of South Florida, where I’ll be majoring in English with a concentration in Creative Writing, thanks to the Post-9/11 GI Bill. What does that mean for this column? Not much (since I already know that "not much" isn’t a complete sentence because there’s no verb). I’ll still be playing in the Armada Games EDH League and bringing you all the tales of the wild and wonderful. I might have new and different tales and side stories about what university life is like when you’re older than most everyone, including the professors. We’ll see what chaos we can embrace there.

The First Draft

Jake’s original idea was basically this:

Randomize your deck.

Split it into a 40-card deck ("Egg") and a 59-card deck ("Dragon").

Set aside the Dragon deck.

Set your life total to ten.

When you lose, exile your hand, graveyard, and library. Set your life total to 30, reset any commander damage and poison counters to zero, make the Dragon deck your library, and draw a new hand of seven cards. Your permanents stay on the battlefield.

Play as normal.

That was the sum of it. Someone has to lose in both phases before they’re out. He suggested perhaps instead of exiling hand, graveyard, and library shuffling it into the Dragon deck, but that defeats the purpose of having separate decks in the first place.

I suggested the idea to some of the regulars at Armada Games. They thought it was worth a look. I decided to sketch out the rules in a little more detail before we ran with it. I noticed a few potential problem areas but decided to just go with it and see where the chips fell.

I broke the game into three distinct phases: the Egg phase, the Hatching phase, and the Dragon phase.

Egg Phase

Randomize your deck.

Count the first 40 cards off the top (the Egg deck), setting aside the other 59 (the Dragon deck).

Draw a hand of seven cards from the Egg deck.

Set your life total to ten.

When you lose, which includes getting decked, enter the Hatching Phase.

Hatching Phase

Your permanents stay on the battlefield.

Exile your library, hand, and graveyard.

Reset all commander damage and poison counters on you to zero.

Move to the Dragon phase.

Dragon Phase

Set your life total to 30.

Make your Dragon deck your library.

Draw a new hand of seven cards (no mulligans).


One of the first things we thought of was the possibility of building separate Egg and Dragon decks. It’d be cool to have a "setup" deck and an "execution" deck. It’d be cool to not have your eight-mana cards stuck in your hand for ten turns. I then realized that 40-card constructed decks, even in a singleton format, is just a bad idea. It would defeat the purpose of cutting out combo decks since that’s all the Egg decks would be. The banned list would have to be hideously long or the construction rules for the deck would have to severely limited. It just didn’t fit with the goals, so we discarded the idea immediately.

The first issue that came up was not having the correct mana in your Egg deck. Especially if your mana base is heavy on nonbasics, you might end up blanking on an Evolving Wilds or not having one of a particular type, which actually happened to one player. He had plenty of his other two colors of lands, but when he searched he ended up without a Swamp to fetch. We agreed that this wasn’t a crippling problem but was worth noting.

The second major issue came up when someone died in the Egg phase. We quickly came to the conclusion that there was a vulnerability issue. If you’re in seat 3 and get attacked and killed by seat 1, you then have to survive the attack of seat 2 (albeit while at 30 life). We also realized that if first strike or double strike were involved, you could actually die in two different phases during the same attack.

We decided on the fly to play with the idea that once you died in the Egg phase you were out of the game until your turn. You couldn’t be attacked or targeted. The problem is that all your permanents are still in play. We decided that they were still vulnerable and could be destroyed, stolen, or whatever.

This created a few more problems. Can you activate abilities? We decided no. Can your abilities trigger? We decided yes, and if choices were required (like with Blood Artist), you got to make them. But what about triggers that added (or subtracted) cards from your hand, like Harvester of Souls? We decided that your hand couldn’t be affected in any way since you don’t actually have one yet. We justified all this thematically somehow as not being able to do much while you’re breaking out of the shell, but the whole thing seemed way too hand-wavy to me. More notes made.

Life gain during the Egg phase became something to think about since, like it is intended to do, it makes you more difficult to kill. Unfortunately, there’s a strong possibility of eventually just getting decked—with your 40-card deck, if you’re spending all your time gaining life, it’s far less likely that you’re killing other players, obviously unless it’s with lifelink creatures. You’ve done all that work to get to a high life total, and when you hatch you’re set back to 30.

The more I thought about the vulnerability issue as it regards your life total, the less problematic it became. In the double strike scenario, you were likely going to die from 40 life anyway. The same goes for two different players attacking you. If you were that vulnerable, you would have taken 40 nonetheless. The same goes for your creatures. Especially if you were tapped out, you wouldn’t be able to save them from one of the other players anyway. We didn’t like the idea of being powerless for potentially two or three turns, but we lived with it. We resolved to give it more thought.

I considered two different options for clearing up the Hatching Phase problem. The first is basically what we played with, and the second is an attempt to resolve some of the difficulties option 1 created.

Option 1

During the Hatching Phase, you may not be targeted or attacked. Your permanents remain fair game.

You may not activate abilities, but if abilities of your permanents trigger, those triggers happen. If you need to make choices for them, you may.

The game otherwise continues without you until it comes around to your turn.

When it’s your turn, move to the Dragon phase.

Option 2

During the Hatching Phase, you may not be targeted or attacked. All your permanents phase out.

When it’s your turn, move to the Dragon phase, and your permanents phase in.

Option 2 solves the targeting and attacking problems. No one can steal your stuff, no one can target or attack you, your permanents won’t trigger, and we won’t have the awkward question of whether or not you can activate your abilities. It seemed like a reasonable solution with one niggling issue; if seat 1 kills you and seat 2 casts Planar Cleansing, you likely get to come back into the game way ahead of everyone else.

One of the problems I had with any version here is that by the rules of Magic there are certain things that happen in multiplayer when a player leaves the game (which, by definition, happens when you die): control effects end, all your permanents are exiled, and all your spells and abilities leave the stack. I’m fine with your spells and abilities leaving the stack, but one of the big things about the variant is that your permanents stay on the battlefield, so we’d have to work out something there. Additionally, what happens if you’ve stolen a creature via Control Magic? We can assume that the creature isn’t exiled due to whatever rule we come up with that says your permanents stay around.

Level 5 Judge and fellow Commander Rules Committee member Toby Elliott (the writer of many, many tournament rules) came up with a potential solution. His idea is to create a replacement effect that would be worded something like:

If a player would die in the Egg phase, exile that player’s hand, graveyard, and library, that player’s Dragon Deck becomes his or her library, that player draws seven cards, remove all commander damage and poison counters from that player, and set that player’s life total to 30 instead. 

Or maybe:

If a player would lose, do the following instead:

Exile that player’s hand, graveyard, and library.

That player’s Dragon deck becomes his or her library.

That player draws seven cards.

Remove all commander damage and poison counters from that player.

Set that player’s life total to 30.

This has possibilities and clears up many of the issues. None of the things that happen when a player leaves the game would come up since what’s happening is a replacement effect. We wouldn’t have to worry about things triggering, control effects, or any of the awkward stuff. As I see it, it would create three problems.

The first is that the player’s spells and abilities wouldn’t leave the stack, although we could just write that into the first line of our rule (" . . . library and all spells and abilities on the stack which that player owns or controls").

Second is the continued vulnerability problem, mostly involving the admittedly corner-case double strike scenario. I’m torn between it being awful or awesome to see someone get killed that way. My strongest inclination would be adding a provision removing that player from combat.

The third, and probably most significant, is that a player could suicide in order to reset their life total to 30 and fill his or her hand. This might not be problematic if a player wants to ping themselves with Prodigal Sorcerer in order to keep someone else from gaining a bunch of life from a lifelink attacker (valid since they still have to beat that player in the Dragon phase), but I wouldn’t want to see a scenario where it would foster the combo decks. Using Necropotence or anything that might have you pay life, you could either deck or kill yourself and then restart with a full hand, potentially continuing the combo.

That said, that situation is probably also a deep corner case—you’d have to have the relevant cards in the Egg deck in the first place. Add to that that the key combo cards that you want might be in the Egg deck, which is getting exiled. Writing it out leads me to the conclusion that this is less of a problem than I had originally thought it might be. I think I’m also getting more on board with the idea that getting killed in both Egg and Dragon from the same attack is more hilarious than it is bad.

All in all, I like Toby’s replacement effect solution. It gives us what we want—the transformation—and gets us away from any uncomfortable rules interactions. The problems are outliers at best, and if they demonstrate themselves to be worse than we thought, individual cards like Necropotence could easily be banned for the variant. The random nature of the Egg and Dragon decks leads me to believe that we wouldn’t need to.

I’d add a modified Gis mulligan to the variant because of the aforementioned random nature and limited number of cards in the Egg deck. The Gis mulligan (so named because it’s the brainchild of Judge Emeritus Gis Hoogendijk) is "continue to draw seven until you have a playable hand; don’t abuse it." You can understand why that’s never going to be the official mulligan of the format. For one, it would create too many arguments over what’s playable. Second, it’s a subjective method when an objective one is called for.

The modified Gis is the one we use locally. If you do not want to keep your first hand, set it aside. Draw seven new cards. If you have three lands, you must keep it. If you have fewer than three lands, you may set it aside again and draw seven more. Repeat until you get a hand with three lands in it. You must keep any three-land hand, even if the lands do not produce mana.

It’s that simple. It’s really hard to abuse, and it doesn’t waste time with a bunch of shuffling. There are only two issues that I’ve seen with it. First, we had to start making some people prove that they didn’t have three lands because they were abusing it by saying they didn’t have three when they did. Second, there have been some folks who have reduced their land counts to as low as 30 because they know they’ll be starting with at least three. Both groups are in the minority; we haven’t really had to make additional changes because of them. I think most of the land-short folks have seen the light and bumped their counts back up since you still want to hit your land drops in the first bunch of turns. I can’t imagine a deck where I wouldn’t start at 37 lands.

One of the things I like about the variant is players getting a second chance. In a normal game, if you can deal 40 damage to me in one combat, I’m done for good. This way, if you attack me for 40 during the Egg phase (setting aside for the time being the double strike scenario), I’ll then have an opportunity for a comeback. Sure, I might get killed by the next player, but I would have normally gotten killed anyway. Look at it this way. Normally, if seat 1 can deal twenty to me and then seat 2 deals twenty, I’m out. In this case, I’m still alive and kicking at ten life.

A problem area I hadn’t considered up to this point is the conditional "win the game" suite:  Azor’s Elocutors, Barren Glory, Epic Struggle, and the like. I don’t actually think they’ll be a problem. First of all, most of them are just bad anyway. Many of them, like Mortal Combat, are highly unlikely to work during the Egg phase, and even if they do their condition is shut off because of the hatching rules. Additionally, if you win the game, you stay in the Egg phase while all the other players hatch into the Dragon phase (and therefore get a full grip). In the end, I think cases where "win the game" cards would become relevant are also few and far between.

I briefly considered something like your commander not being available to you until you’ve hatched, but I quickly discarded that as kind of clunky and unnecessary. The reward of transforming—30 life and a full hand—is sufficient.

Final Draft

Taking in what we’ve discussed here, this is how I’d put together the variant (although I won’t repeat again the details of the mulligan rule):

Egg and Dragon is a Commander format variant that involves two phases of the game for each player. The first is the Egg phase; the second is the Dragon phase. Players play with a random portion of their deck in each phase. Each player’s commander remains available from the command zone during both phases. The commander tax is not reset when a player hatches to the Dragon phase.

Egg Phase

Each player randomizes his or her deck.

Each player counts the first 40 cards off the top, which will become the Egg deck, setting aside the other 59, which will become the Dragon deck.

Each player draws a hand of seven cards from the Egg deck, using the Modified Gis Mulligan.

Each player sets his or her life total to ten.

If a player would lose the game during the Egg phase, that player hatches to the Dragon phase instead.

Dragon Phase

When a player transforms to the Dragon phase, that player does the following in order:

Remove from the stack all spells and abilities owned or controlled by that player.

Exile that player’s hand, graveyard, and library.

That player’s Dragon deck becomes his or her library.

That player draws seven cards (without mulligans).

Remove all commander damage and poison counters from that player.

That player sets his or her life total to 30.

Play then continues normally, with all of a player’s permanents remaining on the battlefield. When a player loses during the Dragon phase, that player is out of the game.

The Egg and Dragon variant (which you can certainly play with any 100-card singleton deck) is intended to be loose, high-variance, and flavor-driven. It intentionally limits the power of tutors and diminishes the ability of dedicated combo decks to dominate a game, providing a change from "normal" Commander without losing any flavor. Give it a whirl and let me know how it goes.

Embracing the Chaos,


Facebook = Sheldon Menery

Twitter = @SheldonMenery

Food and Wine Blog = http://discoveriesinfoodandwine.com/

If you want to follow the adventures of my Monday Night RPG group (in a campaign that’s been alive since 1987), ask for an invitation to the Facebook group "Sheldon Menery Monday Night Gamers."