The challenge described here has had a number of attackers and contributors since it was first posed to a wide audience around early 2009.
The current page was last updated 2014/02/01 by Russell Jones (SadisticMystic), with significant support from Kaitlyn Burnell (metroid composite), whose explanation about a predecessor to this combo is still preserved here. Since my initial draft at writing for this challenge, we were also joined with contributions from Royce Peng (Deedlit11) and forum user "plopfill". Magic: the Gathering is a registered trademark of Wizards of the Coast, a subsidiary of Hasbro, Inc.

Your mission, should you choose to accept it...



I know my description of the challenge at hand involves a lot of technical jargon, but it's not going to get much better from here on. In fact, it'll likely be a good deal worse, because there will be numbers working their way in too. Still, if you're the type of person who wonders about theoretical Magic constructs like this, and if you really do have an interest in keeping up with such wizardry, reading on will be right up your alley. There's not going to be any Tarmogoyf among the cards here...they don't bite.

In essence, this challenge is the analog of the busy beaver problem, but using Magic decks rather than Turing machines as the unit of computational complexity (although...speaking of Magic decks and Turing machines...)


So how much damage can you deal?

"Mountain, Spark Elemental, swing for 3!"

Seriously now, you can do better than that.

"Add three Mutagenic Growth and a Bounty of the Hunt to make it 12?"

NEXT!

"Lotus, Mox, Mox, Mox, Ritual, Channel, Kaervek's Torch. Take 24."

You win! The game, that is. As for this challenge, you're still not even close.

"Exile Simian Spirit Guide and play Mana Clash..."

Sit down. Remember, this is Magical Christmasland, so no matter how much damage you want to deal, there's always the possibility that you flip only heads and the opponent flips only tails for more than that many flips, and indeed no guarantee the spell will finish resolving in any finite amount of time.

"Black Lotus, Forest, Channel, Primal Surge. Flip up a deck of Concordant Crossroads, Intruder Alarm, 4 copies each of Priest of Titania/Elvish Archdruid/Wirewood Channeler/Seeker of Skybreak (plus Wellwisher since life turns into mana too), a Feral Animist to put all that mana to use, and...Djinn Illuminatus. There are still three cards in hand, including such goodies as Cackling Counterpart..."

Now you're on the right track. Doubling is such an efficient operation.


Still, that leaves plenty of cards unutilized. If you added a Gaea's Cradle to the surge, you'd be able to get a similar use from cards like Argothian Elder, but that's quite a minor change compared to the power this combo already has. It's still the same fundamental sequence of events, and the damage output achieved by Feral Animist would hardly even budge: it's still only about 2^(27 septillion). That may be far more damage than you need to win a game of Magic, but you can still do better.

On a more basic level, a problem with Djinn Illuminatus is you have to commit to a number of replicates before any of them resolve, and pay them all up front, even though the subsequent resolutions of the spell will fuel a lot of mana production. It would be much more convenient if you could make copies of things on demand with something like Kiki-Jiki, Mirror Breaker. Of course you'd have to change things around to do that, as Kiki-Jiki would go infinite in the presence of either Intruder Alarm or Seeker of Skybreak. So sometimes using less powerful cards can end up better for the combo in the long run, because it means you have room for more cards that feed off each other without reaching that critical mass of power that lets you spiral off into the boundless realms (Not the Magic 2013 One).

Summoning Station Blasting Station Salvaging Station Grinding Station
Now that's some feeding. Too much so, in fact.

And Now, Our Feature Presentation

The current decklist for our attempt at this challenge is here, in convenient .dec form (or in TappedOut form) if you should ever want to play it. And you almost certainly don't. Seriously, 3 lands, 52 nonland permanents, cards of all 5 colors (even though the lands only make 2 colors of mana), so many terrible cards that do almost nothing outside the contrived circumstances we've created for them here, and an extremely narrow path that has to be followed before it can really get started. Needless to say, outside of the Magical Chrismasland rules allowed by this scenario, it'll have a devil of a time getting anything done at all.

And yet...it stakes out a higher claim for turn-1 damage, which is the question at hand, than any other possible deck the authors are aware of. It truly exemplifies the practice of each card in turn being used to fuel the one before it (but not the one after it, or else it could go infinite--this means that each engine piece in turn will carve out a bit more space that's disallowed to subsequent cards, kind of like creating a minefield, but the explosions are much bigger). The engine has enough different sections, plus assistive cards that serve to make each section as dense as it's possibly capable of, that the 60-card limit is stuffed to the gills.

The basis of the whole thing is the combination of Doubling Season and Minion Reflector. If you have X copies of one, Y copies of the other, effects that turn both of them into creatures (since that's what Minion Reflector triggers on), and an adequate source of mana to cover all payments demanded by all Minion Reflector copies, then playing either one of those permanents will raise its quantity beyond the greater of X or Y...no matter what those numbers are. For best results, playing them in alternation will cause both their ranks to lift each other up, at an exponential rate each time. All that's left is to see that "each time" is talking about a lot of times, and that's what the other 58 cards are for. It's also what the rest of this page is for.

Onward we go.

Section 0 (Getting Started)

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

Begin by playing Badlands, Mana Crypt, and Chaos Warp, shuffling Mana Crypt into the library. It's your lucky day, so the card that flips up is Omniscience, and this 1-in-54 crapshoot is enough to set you home free. Four cards remain in hand, and 53 in library.

The way the mana cost system is envisioned, sometimes the bluntest path to an end is the most efficient one once you have the power to ignore those costs, so go ahead and drop Enter the Infinite. It doesn't matter what the other 3 cards are, because after this spell resolves you'll have everything in hand, including a redrawn Mana Crypt that was shuffled in. Oh, you do have to put a card back, but even that card that stays in the library can be put to a vital purpose early on! Make it Seahunter, who we'll hunt down and see again in a very long time. Just not yet.

Now with all the cards in hand, and the ability to ignore mana costs, you can predictably do some very crazy things. But to pull off the craziest of those, you have to show some restraint, and walk a remarkably thin tightrope. Start with some artifacts and enchantments: Opalescence, Leyline of Anticipation, Mana Reflection, Copy Enchantment (copying Mana Reflection), Doubling Season, Minion Reflector, Rings of Brighthearth, and the redrawn Mana Crypt. It doesn't matter what order you play these in--Minion Reflector can trigger for the enchantments (but not itself or the other artifacts, since those aren't animated, at least not yet), but doing anything with those triggers requires a payment of 2 mana. Omniscience won't allow you to skip that payment, and even though you can tap Mana Crypt for 8 mana (its base of 2, doubled twice by the two Mana Reflections), it's not good to pay for any of them yet. Yes, even rich people need to be budget-conscious at times.

The reason why artifacts aren't animated yet is simple: March of the Machines sets power and toughness equal to the converted mana cost, and in the case of Mana Crypt, that cost is 0--which would immediately kill it, rendering it permanently unusable for the rest of the combo, and that's obviously not good. Additionally, even if it could live through animation, being a creature makes it subject to the summoning sickness rule, and therefore unable to tap. (Animating artifacts isn't all sunshine and rainbows, maybe with the exception of Pentad Prism.) We'll have to knock out these issues one at a time, starting with the toughness.

Play Elspeth, Sun's Champion. Thanks to the Doubling Season, her 4 starting loyalty counters are doubled to 8, and fortunately that's enough to use the -7 ability right away. But wait, there's more: Before the ability can resolve, Rings of Brighthearth will trigger, allowing us to pay 2 to copy the activated ability. Go ahead and do that, down to 6 mana left, and now Elspeth creates TWO emblems. All creatures from now on will have flying (twice), and they'll all get +4/+4. This is the last we'll ever hear of Elspeth, so give her a nice round of applause while you can, and behold as the "ultimate" can now freely unfold all because of those two emblems.

Now you can finally play March of the Machines, turning Mana Crypt into a 4/4 creature, Rings of Brighthearth into a 7/7, and Minion Reflector into a 9/9. Those numbers may sound big to you, at least in "normal" Magic terms, but before long you'll be looking down on them as absolutely tiny. There are also three more enchantments to play now: Cowardice, Cephalid Shrine, and Bloodbond March. Minion Reflector will trigger for all of these, and you should decline to pay for any of those triggers except Bloodbond March, so that's 2 more mana spent and 4 left in the pool. Minion Reflector wants to create a token copy of Bloodbond March, which Doubling Season doubles into 2 tokens. There are now 3 Bloodbond Marches on the battlefield in all.

Cephalid Shrine is normally a very weak card. The fact that it's being played in a singleton deck, against a goldfish opponent, is comical on its face. Yet it still has a very important role, one that already has a chance to manifest itself. The next card to play is Glissa Sunseeker, and before that can even resolve, four things trigger: Cephalid Shrine and all three Bloodbond Marches. You control them all, and can choose any order for them: put Cephalid Shrine on top, so that it resolves first. It wants to counter Glissa unless you pay...hmm, there are no other cards named "Glissa Sunseeker" anywhere in the decklist, let alone in the graveyard right now, so to avoid letting Glissa get countered, all you have to do is pay {0}, and you can pay {0} simply by assenting to such.

Don't assent.

The power of Cephalid Shrine lies in the fact that, because this is a goldfish game and only one player is playing spells, you have full control over what gets countered and what doesn't. And the reason you might want spells to be countered, as you might have guessed by now, is the existence of Bloodbond March triggers: just three right now, but many more later. As Bill Boulden can tell you, you only need to counter a spell once, and then Bloodbond March can return it many times, as long as it somehow ends up in the graveyard each time. The March received little fanfare upon its release, though the effect was much higher-profile when it returned a few years later on Demigod of Revenge (probably because the new version had a built-in way of smashing face, and that's all the company seems to want anyone to care about nowadays).

So the first March trigger resolves, and it returns all cards named Glissa Sunseeker from the graveyard. As it turns out, there's one of those, because it was just countered. Return Glissa, and get a Minion Reflector trigger on her. There's still 4 mana left, so use 2 of it (dropping to 2) and get a token...or rather 2 tokens, but it doesn't matter. There are 3 Glissas now, and they all share a name and the Legendary supertype, so all but one of them has to go. Pick either one of the tokens to keep, because Minion Reflector's tokens actually have haste; this also conveniently puts Glissa back in the graveyard for the next Bloodbond March, but don't let the stack clear that far yet.

You might as well take advantage of the token Glissa's haste, by tapping her to attempt to destroy Mana Crypt in response. Note that the mana pool is not a targeting restriction for Glissa (all it says is "target artifact", with the rest following an "if" clause), so since there's still {2} in the pool, that attempt will inevitably fail but it's still legal to try. In fact, we don't actually want Mana Crypt to be destroyed, but two things trigger from activating Glissa: Rings of Brighthearth attempts to copy the ability by demanding a payment of {2}, which isn't decided on until resolution time. Also, Cowardice kicks in and "rescues" the targeted Mana Crypt by bouncing it back to hand. Arrange the triggers with Cowardice on top.

Cowardice returns Mana Crypt, where you can immediately replay it in response to everything else (thanks, Leyline of Anticipation!) Cephalid Shrine triggers, but Mana Crypt is not a creature spell so Bloodbond March doesn't. The rule is "have Cephalid Shrine counter all creature spells, but no others"--this time, choose to pay the {0}, so Mana Crypt resolves. This time it does see March of the Machines, so it enters as a creature and Minion Reflector gets to trigger. Spend the last two floating mana to get 2 token copies of Mana Crypt, which now have haste, and the light of freedom from mana chains is just a bit further ahead. Hang in there.

Tap the first Mana Crypt token to replenish the pool with 8 more mana, and play Psychic Battle, paying {0} to Cephalid Shrine and {2} to Minion Reflector and winding up with 3 copies of Psychic Battle. (If you want to, you can also add Grip of Chaos here, and in fact the path to absolute maximum damage will need to have it out now and will need to get lucky on every single one of its random target redirections for the entire game, as the "perfect luck" clause does permit. As long as I'm just explaining how to get started from nothing, though, I'll describe a method that doesn't need its cooperation.)

Play Sinking Feeling, attempting to attach it to the animated Mana Reflection. Aura spells on the stack target what they're trying to enchant, so Cowardice triggers, as do all copies of Psychic Battle (and Cephalid Shrine, which is obviously inconsequential and which I'll ignore on noncreature spells from now on). Resolve Cowardice first as usual, bouncing Mana Reflection. Don't replay it just yet.

Whenever a Psychic Battle trigger resolves, you take part in something that's similar to a clash, except no one gets the option to put their card on the bottom. Remember how Enter the Infinite forced the return of one card to the library? That card, currently Seahunter, will be revealed for every such trigger until further notice; its converted mana cost of 4 will easily beat the goldfish who has only basic lands and who always gets a 0. This allows you to "win" every Psychic Battle and change the target. Each time you do, Cowardice will trigger (it acts upon something becoming the target, even if that target is just being changed from something else), allowing you to bounce something else...or even the very same thing, which has a reset object identity after changing zones and being replayed. For now, the first three battles should go as follows:
  1. Change target to Doubling Season. Cowardice triggers and bounces Doubling Season.
  2. Change target to Cowardice. Cowardice triggers and bounces itself.
  3. Change target to the tapped Mana Crypt token. Cowardice isn't on the battlefield, so it can't trigger.
That last one is important, because we don't want the Aura to fizzle. It still hasn't entered the battlefield yet, and if there's no target on resolution then it never will: off to the graveyard, lost forever. This is a Very Bad Thing for the goal we're after here, so whenever Psychic Battle is bouncing around something that's supposed to resolve, such as Allay, the second-to-last target change always needs to make Cowardice show its true colors (hint: it's blue) and hide back in your hand, then the last one will change to the proper target. In this case, Sinking Feeling can properly attach to the Mana Crypt.

So anyway, Cowardice, Doubling Season, and Mana Reflection are back in hand, and there's {6} left in the mana pool. Replay Doubling Season first, paying {2} to get 2 tokens of it, then Mana Reflection. Since there are now 3 Doubling Seasons, its Minion Reflector token gets doubled up three times into a total of 8, plus the original and the Copy Enchantment. Now since Mana Crypt is 4/4, and has the Sinking Feeling on it, activate Sinking Feeling, dropping to 1 mana and shrinking Mana Crypt to 3/3. Note that Doubling Season only doubles counters that are placed from effects, not costs; activating Sinking Feeling does not immediately flood the Crypt with a lethal dose of 8 -1/-1 counters, which is good because otherwise that card wouldn't be good enough to make it into the deck. The activation does trigger Rings of Brighthearth, which I'd like to pay for but can't since there's only one mana right now. Oh well. At least the original activated ability resolves, untapping Mana Crypt.

Immediately tap it for mana again and it will be doubled ten times by the Mana Reflections, producing a total of 2,048 mana. It should be clear now that this is easily enough to break free from the tyranny of small numbers, and to start picking up section 1-3 as described below in earnest. Most of the current 2,049-mana supply should be spent repeatedly casting Allay with buyback, bouncing and replaying Doubling Season. Before long, there will be so many tokens that when the supply of Psychic Battle triggers is exhausted, it'll be okay to change the target to a Doubling Season token and actually destroy it, while hardly noticing the loss. A quick rundown of the numbers: Yeah, that's a lot. 2^2,059 is a number with 620 digits. At least it isn't too obstructive, taking up only a few lines of text, so here you go.
66,185,228,434,044,942,951,864,067,458,396,061,614,989,522,267,577,311,297,802, 947,435,570,493,724,401,440,549,267,868,490,798,926,773,634,494,383,968,047,143, 923,956,857,140,205,406,402,740,536,087,446,083,831,052,036,848,232,439,995,904, 404,992,798,007,514,718,326,043,410,570,379,830,870,463,780,085,260,619,444,417, 205,199,197,123,751,210,704,970,352,727,833,755,425,876,102,776,028,267,313,405, 809,429,548,880,554,782,040,765,277,562,828,362,884,238,325,465,448,520,348,307, 574,943,345,990,309,941,642,666,926,723,379,729,598,185,834,735,054,732,500,415, 409,883,868,361,423,159,913,770,812,218,772,711,901,772,249,553,153,402,287,759, 789,517,121,744,336,755,350,465,901,655,205,184,917,370,974,202,405,586,941,211, 065,395,540,765,567,663,193,297,173,367,254,230,313,612,244,182,941,999,500,402, 388,195,450,053,080,383,488

Okay, now wave goodbye to the big number, because it's the last time the rapidly-expanding growth rate will be expressible using numbers that can be written out in full. (For comparison, our best estimate for the total number of elementary particles in the universe only has about 80 digits, which is barely one-eighth that length when written out, and far less than one-eighth of that value. If you can imagine taking "the square root of the square root of the square root", that's about how much smaller the value is.) To think, that's just from three more bounce cycles. 2,049 mana is enough to pay for about a thousand of them.

And of course, Allay is capable of slapping not just Doubling Season but any other global enchantment that needs bouncing (soon to be followed by replaying and getting another Minion Reflector trigger). There's one more "trick" you'll need to know before all the pieces come together, and that's the possibility of playing Copy Enchantment while choosing to copy nothing at all. If (and only if) you do that, then when the Minion Reflector tokens come in, you get to choose each copy separately, instead of being locked into what the original is copying. By itself, this isn't often useful; the real power comes from the fact that this maneuver, and only this maneuver, allows the deck to create additional copies of Auras. If Copy Enchantment does come in as an Aura, you simply "choose" something for it to attach to, and it doesn't "target" that thing so it can be done safely even on a board with Cowardice, not that such a fact is ever going to be relevant here. But if the nontoken Copy Enchantment came in as an Aura, well...Opalescence specifically does not animate Auras, and with good reason: by rule, an Aura creature is automatically dead. Then Copy Enchantment wouldn't enter as a creature, and wouldn't trigger Minion Reflector, and we'd have no way of bouncing it with Cowardice either. That would be bad.

So just from the current resource level, there's about a thousand enchantment bounces to be had. Then the tapped Mana Crypt is 3/3, so Sinking Feeling can be activated twice more...and now that the budget isn't so miserly, paying for Rings of Brighthearth on the untap isn't a problem. So the Mana Crypt can untap four more times before it ends up at 1 toughness and unable to use Sinking Feeling again. What then? Well, there's a second Mana Crypt token, of course! It's still 4/4, so slap a Sinking Feeling on it to allow it to untap SIX times! Then*, Glissa still has that unresolved Rings of Brighthearth trigger. She's also still 7/6, so a Sinking Feeling there allows at least ten more untaps. Beyond that, there's still two more Bloodbond March triggers on Glissa. Use the second one and get another Glissa token, which if everything has been done correctly, will be able to untap a lot more than just twice. (When the third Bloodbond March trigger returns Glissa yet again, you'll have to refuse the Minion Reflector token, but by then it'll be time to add some new cards to reach ever higher.)

*Oh yeah, there was a * in the previous paragraph. That signals when you get to it, there's one more minor step that needs taking before everything can be up and running at full power. Play Nature's Revolt to animate the tapped Badlands which helped pay for that very first Mana Crypt, making it a 6/6 after taking Elspeth's emblems into consideration. As a creature, summoning sickness obviously keeps you from tapping it even if it weren't already tapped, but what matters is that as a creature, it's now eligible to throw a Sinking Feeling on. Activate it 5 times, all in response to itself, and pay Rings of Brighthearth each time, getting a total of 10 untaps. But first, respond to all those activations and triggers by bouncing Nature's Revolt. Now lands cease to be creatures (and cease to be affected by summoning sickness), which makes Sinking Feeling fall off, but the untap abilities do still resolve. After each one, but before the next resolves, tap the Badlands again for a total of 9 batches of red mana plus 1 batch of black, which are undoubtedly multiplied many times by Mana Reflection. Now Nature's Revolt can come back and stay back; it'll be needed later on.

Also, now that there's a supply of red mana, play Figure of Destiny. Get a single batch of Minion Reflector tokens (ignoring Bloodbond March), and use just one {R} to activate the first ability on one of the tokens. Rings of Brighthearth does absolutely nothing to the ability; all that matters is that one lucky Figure is a 2/2...I mean, 6/6...Kithkin Spirit.

Section 1

6

20

15

10

11

You can play Allay basically any time you have the spare colorless mana for it. Omniscience allows us to ignore the mana cost, but not the buyback cost, so Allay still costs {3} each time. As you've seen above, all the target-changing allows us to send any enchantment on a brief vacation repeatedly, and more often than not it'll be right back for more (as long as you pay {2} for a Minion Reflector trigger each time--later on there will be multiple Minion Reflectors on the board, but you'll only ever need to pay for one at a time.)

As you keep looping Doubling Season, the numbers will just get more and more out of control, and out of mind. The first few batches of tokens were in quantities of 2, then 8, then 2,048, then 2^2,059 (the big 620-digit number from before). With a slight loss of precision, we can estimate the number after the fifth replay as 2^(2^2,059), then 2^(2^(2^2,059), and so on, with each replay just tacking a 2^ to the front of the expression. But even that notation is going to be unwieldy after a while.

Fortunately, almost 40 years ago, a guy named Donald Knuth came up with an infinitely extensible solution to that mess. It's well established that all of the natural numbers can be built from a single base case (called "zero") and a "successor" operation: 1 is "the successor of 0", 2 is "the successor of the successor of 0", and so on. On top of that structure, the common operations are really just shortcuts for loops of the successor function, the ultimate recursive building block:



Naturally, there's no reason why you couldn't continue such a pattern for deeper and deeper layers of recursion, which are the main engine that drives larger and larger numbers here. After a while, you might have trouble coming up with distinct symbols for each layer, but Knuth's idea was to simply use multiple up arrows, or carets, to represent further nesting: "a ^^ b" means "a ^ (a ^ ... (a ^ a) ...), b times.", and so on for 3, 4, and more arrows. Obviously those numbers get very big, very fast. Let's look at the first few values of 2^^b.

Since our base-case operation for blowing up the numbers during this combo is "get a batch of token copies of Doubling Season" and that takes the form of a ^ b (or really, 2 ^ b), it makes sense to say that exponents are our first layer, rather than deferring to addition and multiplication in front of it. This conveniently also allows the number of layers, and the number of carets (or arrows) in the representation, to be equal. 2,059 itself lies between 16 and 65,536, so 2^2,059 (the 620-digit number) is between 2^^4 and 2^^5. To avoid overestimating at any point, I'll round down to 2^^4, and then each replay of Doubling Season (plus the 2 mana for the associated Minion Reflector trigger) has the effect of adding 1 to the number of the right side.

There's a lineup of other "helper" enchantments--Mana Reflection, Clash of Realities, Rite of Passage, Bloodbond March, Grip of Chaos, Psychic Battle, and Copy Enchantment (in case you need to copy Auras)--which all have their occasional merits, and occasionally you'll run out of some resource and the last gating trigger in a batch should be used to bring back one of those instead of the usual Doubling Season. For example, if you're just about to tap something for mana, you'll want to bounce Mana Reflection and "top up" on that, spending 2 mana at a point where you're likely used to dealing in much larger quantities. As the combo winds on longer and longer, these "off targets" amount to a less and less noticeable proportion of the overall number of enchantment bounces, to the point that they're practically nonexistent. 2^^20, for example, is so monstrous that adding or subtracting 2, or even a billion, doesn't leave a dent in the number, so we can pretty much ignore the lost contribution of those bounces that don't result in more Doubling Seasons, and only focus on the asymptotic behavior. With numbers this big, there's no real way to talk about them with any more precision than that. And of course, they're only ever going to get bigger.

Section 1 Summary

Input: 2 colorless mana (and occasionally 3 more mana of overhead from Allay)
Output: Lots of copies of Doubling Season
Layer count: For each Doubling Season...1. Total 1.
Newly carved out space: Enchantments, sources that turn arbitrary creatures into mana

Section 2

14

2

23

24

9

25

26

21

Of course, Mana Crypt is what provides the colorless mana in this deck. You should always come into this section with at least 2 spare colorless mana, so that when you play Mana Crypt, you can afford to pay for the first Minion Reflector (and you only ever have to pay for one of them at a time). Other considerations, such as the imminent end of subsequent sections, may cause you to hold up more mana than that.

The point of the deck is to use multiple sets of triggered abilities that key off each other in a nice, orderly line, but always in the same direction so there are no infinite loops. In this case, every time a Minion Reflector trigger is about to resolve, you should use the last enchantment bounce to top up on Clash of Realities. Actually, that card hasn't even been played yet, so go ahead and play it for the first time and get one round of copies. Also, make sure Cowardice has gone away, since it's inexpensive to do so.

Clash of Realities contributes an ability to every creature, even the lowly Mana Crypt tokens; each copy of Clash imbues a separate instance of the ability, each one allowing Mana Crypt to deal 3 damage to target Spirit. As it turns out, there are no Spirit cards in the deck, and this is entirely deliberate because any such thing can easily go infinite with later cards down the line. The closest thing is Figure of Destiny, which can become a Spirit but only by spending 1 mana, like we made sure to do at the end of startup. In particular, it is absolutely not capable of becoming a Spirit fast enough to get Clash of Realities' first trigger instead of the second. The type change is not a copiable value either, so even if you activate the original to turn it into a Spirit in response to Minion Reflector triggers, all the tokens will still be regular old Kithkin, insistent on damaging their own brethren (as the only legal target) over, and over, and over for each Clash of Realities. Yes, it is possible to play Figure and activate its ability in response to the Psychic Battle triggers, then have one of those triggers redirect the target to itself, but this does nothing except waste a red mana. Only creatures with tap abilities can take full advantage of Clash of Realities.

When a Clash trigger is about to resolve, there are a couple things that also need topping up on, which haven't yet been played. Get a round of Rite of Passage, and use the Copy-Enchantment-copying-nothing trick to get a bunch of blank canvas tokens, all of which can come in as Guilty Conscience, attached to whichever creature is about to deal damage (except that if it's the first Clash of Realities trigger, one of them will need to become a Sinking Feeling instead, since the creature won't have one of those yet). After Figure of Destiny takes 3 damage, Rite of Passage triggers and promptly heals it with a boatload of overkill (or should that be...oversave?) In any case, keeping the Figure alive isn't going to be a problem.

Naturally, there are more productive things to do than just repeatedly pump Figure of Destiny. Each Guilty Conscience also triggers from the damage, and they each want to punish the unholy Crypt for being so reckless toward their own team's creature. The prescribed sentence is 3 damage to Mana Crypt, but these are overzealous judges so each one will continue meting out its own punishment without paying attention to the others which have done the same, and will continue doing the same until their numbers run out. Then when Mana Crypt deals damage again from the second Clash of Realities, there will be even more Guilty Consciences to join in the fun.

Mana Crypt starts out at 4/4 here, and this is why two Elspeth emblems are required, instead of one. (On a side note, toughness boosts are extremely dangerous to work with in a deck like this, especially on an activated ability; Elspeth simply happens to be the most compact mechanism that we can guarantee is only usable once...or I guess, twice.) Mana Crypt has to survive the first blast of 3 damage in order for the Rite of Passage triggers to do anything. The first one gives it a bunch of counters, which it can then use to untap and tap with Sinking Feeling, producing more mana until it's 1 toughness away from death, and using that mana to get tons more Doubling Seasons. Then the second Rite gives it even more counters, and so on until the last Rite trigger resolves. At that point Mana Crypt has to stop at four toughness away from lethal, and close out by topping up on Rite of Passage to get even more triggers from the next Guilty Conscience.

Each time you replenish the mana supply, it's time to immediately go out on a spending binge again, playing through section 1 until you're back down to just enough mana to cover the overhead of the following sections. Think of it as something like a ruler: as you go across, the first and smallest mark might be for a 1/16 inch, representing our section 1 here. Then you have a 1/8 inch mark that's slighly taller (section 2), and back to a 1/16, then a 1/4 (section 3), another 1/16, and so on. That's the kind of pattern that you'll keep up throughout the rest of the combo, because it's what generates the most exorbitant growth rates: the power of "do something for each X," where the supply of X's keeps getting bigger every time you come back to start that step anew, and where the future steps of "do something new for each Y" also involve coming back to all the previous steps, just as the definition of exponents involves coming back to multiplication, which itself involves coming back to addition.


Yeah, it's kind of like this, except for the whole "stopping at 12 feet" thing. We can go a lot further.

When you've exhausted a Crypt, or any other permanent, to the point where it can't get any more untaps, feel free to get rid of it if you like. Just make sure there's always at least one copy of Guilty Conscience and one copy of Sinking Feeling on the battlefield at all times, because Copy Enchantment can't produce more without drawing from a surviving specimen.

Section 2 Summary

Input: One playing of Mana Crypt
Output: Lots of colorless mana
Layer count: For each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Mana Reflection...6. Total 7.
Newly carved out space: Colorless mana sinks, Spirits, tap-for-damage effects

Section 3

16

17

18

12

27

19

So you've seen the first glimpse of how playing Glissa Sunseeker and deliberately allowing her to be countered was good for repeated returns. (I wonder if that infusion of Golgari magic from Bloodbond March was responsible for turning her into a traitor?) Anyway, now this section lets you attempt the same thing without being subject to the limitations of small numbers like 3 (copies of Bloodbond March) or 6 (toughness). Each time you're about to play Glissa, it makes sense to top up on Bloodbond March for maximum effect.

This is also where you'll start using Grip of Chaos and Psychic Battle to full effect. Reserve some more colorless mana from the section 1 and 2 loops, and activate Glissa. A few things trigger on this: Cowardice will bounce whatever you targeted, Rings of Brighthearth will allow you to get more copies of the ability, and all copies of Grip of Chaos and Psychic Battle will trigger on the act of targeting.

Grip of Chaos is a very unusual card. Its printed wording actually needed to be fixed by way of release-day errata: as written, it would trigger on any spell or ability at all. So something goes on the stack, which triggers Grip of Chaos, and the ability triggers Grip of Chaos again, and again, and the stack would fill up with a never-ending barrage of Grip of Chaos abilities. No one would even gain priority for the rest of the game! Needless to say, this behavior was undesirable.

They fixed it by introducing "intervening if-clause" technology into its text, similar to the way cards such as Battle of Wits are worded. Now Grip of Chaos can only trigger in the first place "if [the ability] has a single target", and since Grip of Chaos's own ability doesn't have a target, it neatly avoids the infinite loop.

Anyway, the reason Grip of Chaos is helpful here is that it triggers whenever a (single-targeted) spell or ability is put onto the stack, not just when it's cast or activated. This means that when Rings of Brighthearth produces copies of Glissa's ability (or any subsequent targeted ability), Grip of Chaos will trigger for those copies too.

Perhaps even more unusual is Psychic Battle, which is similar to Grip of Chaos in that a cursory reading of its printed text would have you believe that it could endlessly trigger itself, and that wouldn't be good for a deck that has to keep within some sort of finite bound. They "fixed" this problem by adding errata: "Changing targets this way doesn't trigger this ability." Seems all well and good, and it does stop a single copy of Psychic Battle from looping forever, which is normally all that's necessary. But this deck is anything but normal, so go figure that there will be a ton of tokens flying around (literally, thanks to Elspeth's emblems). Where Psychic Battle said "this ability", it was limited in scope--it really means "this exact ability, on this particular permanent". If there are two or more copies of Psychic Battle, they count as separate abilities, so they could endlessly bounce a target back and forth, and that's unacceptable.

All that changed on January 28, 2014, when the Born of the Gods Update Bulletin released. It was on this day, more than thirteen years after the release of Invasion, that the rules team finally revised the errata on Psychic Battle. It now reads "Changing targets this way doesn't trigger abilities of permanents named Psychic Battle," which is enough to prevent the infinite game of Hot Potato no matter how many tokens there are. Several months prior, I had alerted the team to the issue with multiple copies, and tried to lead them in the direction of the wording it ended up taking, only to be pleasantly surprised when they actually followed through on that request. Did I have some ulterior motive behind it, such as, I don't know...maybe putting the card to use in the very deck you're reading about right now? Sure. But the card undeniably works better for it, and everyone goes away pleased.

Even though Grip of Chaos has you choose its target at random, it still has you "[re]select a target", and that's good enough that in addition to having Psychic Battle trigger on the original activation of an ability, each Psychic Battle also triggers on every single Grip of Chaos: a good indication that when you run out of Psychic Battle redirects, and go through the lower layers, one of the last things you should do is top up on Psychic Battle, to get more triggers for the next Grip of Chaos. Likewise, when you run out of Grip of Chaos, it's also worth topping up on that for the next copy of the targeted ability.

Prior to the sanitization of Psychic Battle, this procedure fell back extremely heavily on the "Magical Christmasland" crutch of perfect randomness, to a far greater extent than almost any conceivable deck in the history of the game, because getting anything out of it was dependent on having every single Grip of Chaos trigger change the target exactly as needed, picking out one success from an ocean of failures. Now that Psychic Battle gives us full control over the targeting, Grip of Chaos's role diminishes to just another enabler. In fact, if you don't like the idea of depending so much on randomness, you can arrange it so that Cowardice goes away every time Grip of Chaos is about to come up, to guarantee that whatever Grip chooses is irrelevant--all it will accomplish this way is getting more Psychic Battle triggers, and then you can bring Cowardice back.

As far as what you'll want to target, the ideal scenario is to divide the targeting actions into groups of 3, and in each group change the target back and forth between the nontoken copies of Minion Reflector, Rings of Brighthearth, and Mana Crypt. (Obviously you only ever want to bounce nontoken copies, since they're the ones that can actually stick around in your hand.) With all those back in hand and 9 spare mana, you should pay {3} to Allay and make Cowardice go away, then replay Minion Reflector and pay for just one of its own triggers, replay Rings of Brighthearth and pay for one Minion Reflector trigger, and finally replay Mana Crypt and use all off its triggers to repeat section 2, bigger and badder this time. Taking a number that's big enough to use several arrow worth of Knuth notation and dividing it by 3 doesn't cause any appreciable difference, so this isn't a big loss even though the loop only repeats with every third targeting action.

All three of those replayed "creatures" will get triggers from Clash of Realities, even though Mana Crypt is the only one that can take full advantage of them. This is why Cowardice had to go away: we don't want to bounce the Figure of Destiny, only damage it. Go ahead and let Minion Reflector and Rings of Brighthearth deal some damage--it causes Rite of Passage triggers, after all, and making Figure big doesn't hurt (until combat, anyway, and then it only hurts the opponent). For the same reason, with every trip to section 1 from now on, you need Cowardice to go away each time you replay an enchantment.

Again, remember that Glissa is legendary, and since tokens are the only things that have haste here, Glissa will invariably have to sit in the graveyard while a token is doing its thing. Only Bloodbond March will return her, and after the last Bloodbond March of a batch, you have to decline Minion Reflector to have any hope of continuing.

Sure, the legendary aspect means only one token can exist at a time. This obviously isn't as good as having tokens from all the Doubling Seasons, but the loss isn't actually bad enough to miss out on any layers. Suppose there are X copies of Doubling Season, and Y copies of Clash of Realities. Then a token that enters the battlefield will turn into 2^X tokens, all entering simultaneously, and causing Y*2^X Clash of Realities to go on the stack...also simultaneously. As you move from one token to the next, the number of Clash triggers then can't help but remain constant between them, and only a new batch of tokens will be able to take advantage of additional Clashes. Y*2^X is clearly larger than just Y, but once you get up to 3 or more up-arrows, there isn't really much expressive difference between X and 2^X, and even less of a difference between X and X^2 or X*Y. As long as the Y is increasing alongside everything else in the loop, as it is here, it's still a layer even if it's not being multiplied by X or 2^X.

Section 3 Summary

Input: One playing of Glissa Sunseeker
Output: Lots of playings of Mana Crypt, Minion Reflector, and Rings of Brighthearth
Layer count: For each Bloodbond March, for each Minion Reflector, for each Clash of Realities, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 17.
Newly carved out space: Artifacts

Section 4

28


Elves are a well-known, venerable tribe with lots of support. Some of it isn't useful here, and some would go infinite. But what proves best for the combo is to reach out to an unexpected ally...Faeries? Yes, in Lorwyn block the Elves branched out into black, and they decided to include some cross-tribe cards to enable multi-tribe play in combinations that shared a color. Part of that effort included Nectar Faerie, an oddball card that helped black get accustomed to lifelink as part of its mechanical identity. We don't care about the lifelink, but it's another dummy effect that Cowardice eats right up. Actually, by throwing Cowardice away and letting Nectar Faerie go ahead and give Glissa lifelink, it's possible to gain life from the Clash of Realities triggers, so this does do one thing: prevent us from using "Pay 1 life:" abilities as a bottleneck in the future.

The activation cost on Nectar Faerie includes black mana. In this case, Nectar Faerie also requires a tap to activate, so you will need to get some Minion Reflector tokens, which have haste. Lest there be a potential problem of running out of copies of the Faerie before running out of black mana, note that it's possible to use a single instance to target the nontoken Nectar Faerie itself, bouncing and replaying it as necessary to ensure that black mana is the only bottleneck in this step.

When we untapped Badlands a few times during startup, with the main goal of powering up Figure of Destiny, there were 9 taps for red mana and 1 tap for black, which is enough to use Nectar Faerie a few times (but only a few, compared to the vast number of copies of everything else that's accumulated so far. After that...we'll need more backup.

Section 4 Summary

Input: One black mana and a Nectar Faerie
Output: Lots of playings of Glissa Sunseeker
Layer count: For each Rings of Brighthearth, for each Grip of Chaos, for each Psychic Battle...3. Total 20.
Newly carved out space: Elves, Faeries, life sinks

Section 5

22

29

So you have Nature's Revolt already. Despite it being yet another enchantment, you no longer have any reason to get Minion Reflector tokens of it, or bounce it with Cowardice. What it does allow is for lands to enter the battlefield as creatures, so they get Minion Reflector triggers.

Vault of Whispers is a pretty nice land, good enough to be banned in Standard at one point. And that's without it powering a deck that gets several up-arrow layers worth of token copies of everything--imagine how good it is here. Simply get it onto the battlefield, get some tokens that tap for black mana, which fuels Nectar Faerie and Glissa. Then Glissa can use one iteration of her ability to pop Vault of Whispers right back into your hand...spot the problem yet?

Yes, this would easily go infinite if it weren't for one thing: you can only play one land a turn, and Badlands was already it! Then when Glissa bounces Vault, as she's certainly capable of, it'll take yet another outside force to get it back to the battlefield. Sounds like we need more help then, from...another card!

Section 5 Summary

Input: Vault of Whispers entering the battlefield, in the presence of Nature's Revolt
Output: Lots of green mana
Layer count: For each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Mana Reflection...7. Total 27.
Newly carved out space: Black mana sinks

Section 6

30


Sakura-Tribe Scout is a card that has relatives all over the multiverse. It's the less popular cousin to Sakura-Tribe Elder, as well as a seven-year-belated remake of Skyshroud Ranger. As a matter of fact, Skyshroud Ranger doesn't work in this deck because of its sorcery-only restriction that even Leyline of Anticipation doesn't get around. When they remade it in snake form (but not in Snakeform, fortunately), they streamlined the wording by taking away the timing restriction, which means the new version is usable in an instant-speed combo.

You know the drill by now: get some tokens, pump them up, and copy the ability with Rings of Brighthearth. Whenever you get to the position where Sakura-Tribe Scout's ability is about to resolve, use Glissa to bounce the nontoken Vault of Whispers, and pick that as the land you put in. Get token copies of the land, and repeat.

The first time you play Sakura-Tribe Scout...well, actually, that's going to end up being the only time you actually get to "play" Sakura-Tribe Scout. Go ahead and let Cephalid Shrine counter it to get the Bloodbond March triggers while you still can, but all subsequent reuse of the Scout is going to involve dropping it directly onto the battlefield without passing through the hand, so we can't count on Bloodbond March for a layer here.

In fact, even though Scout does get one round of Bloodbond March, it takes a bit of trickery to make use of it. Each time it comes back from the graveyard, but before it takes damage or gets bigger, use Copy Enchantment to dump a Sinking Feeling on the nontoken copy. Of course, this is the one with summoning sickness so untapping it doesn't do a whole lot of good. The point isn't to untap it, though: it's to spend a few colorless mana and kill the Scout with -1/-1 counters! Ta-da, back to your grave. And for your next trick, I'm sure you'll just rise up again.

Section 6 Summary

Input: Sakura-Tribe Scout entering the battlefield
Output: Lots of copies of Vault of Whispers
Layer count: For each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability...7. Total 34.
Newly carved out space: Lands that can target or bounce themselves, or be bounced by effects prior to this point

Section 7

31

32

33

34

Izzet a bird? Izzet a plane? Izzet a mad scientist hellbent on perpetuating crazy combos? Okay, that's one out of three, and you've probably had enough Izzet puns over the years anyway. It's Izzet Guildmage, another creature that you only need the initial copy of.

Omniscience allows you to play spells for free, but only as an option, and for once you're going to decline that permission. First off we need some green mana, so play Overgrown Battlement and run all the tokens and triggers for all they're worth until the stack is clear, then get one more batch of green. Use a single green mana (along with a colorless) to play Green Sun's Zenith the "hard" way, choosing X=1. (If you attempt to cast an X spell using Omniscience, X is forced to be 0, which is undesirable here.) That spell isn't going to resolve for a very long time, though it will at least resolve. Hey, it's more than I can say about some of the other spells and abilities being thrown around.

Sure, it's possible to play Green Sun's Zenith for a lot more than 1, just by setting aside a few more mana out of the boatload available, but there's a good reason not to. The genius of X=1 is that it's the largest value that allows the spell to be a legal target for the red ability on Izzet Guildmage. Activate it once with a red mana, copy the ability with Rings of Brighthearth, and get another copy of Green Sun's Zenith, still keeping the same X value of 1. Before that spell resolves, though, we have to do something else to make sure it has some effect.

That something else is Mortuary. In fact, when you go through the Bloodbond Marches on that first Scout run, play Mortuary after the last March returns Scout, then go ahead and kill the Scout one more time. Now Mortuary will trigger, and move Scout from the graveyard to the top of the deck (which also means that Psychic Battle will start flipping Sakura-Tribe Scout instead of Seahunter, but 1 still beats 0 so there's nothing to worry about.) As soon as Scout is back in the deck, use Cowardice to get rid of Mortuary, because its effect is mandatory as long as it's the battlefield, and that would screw up Bloodbond March for Glissa and other creatures. Only a select few cards can ever be fetched from the library, so Mortuary needs to be used sparingly to make sure that those are the only cards that end up there.

So now that the library is properly seeded, let a Green Sun's Zenith copy resolve. More Scouts, and more mana. Cue your favorite impression of Maxwell the GEICO Pig. The original Green Sun's Zenith is one of those cards that can't be fetched from the library, and its effect includes shuffling itself in, so we can't ever allow the stack to fully clear from now until the end of the combo, or else there's won't be anything for Izzet Guildmage to make copies of.

We've already seen a step that calls for the production of colorless mana, then a similar step for black mana. Needless to say, this is where we need red. And you might guess that with 25 cards still unaccounted for in the deck, one of them is going to be magically useful for accomplishing just that. Actually it's going to be one of the cards that's already accounted for, but see? You're getting the hang of the deck already.

Section 7 Summary

Input: One red mana and Izzet Guildmage
Output: Lots of copies of Green Sun's Zenith, which fetch Sakura-Tribe Scout a lot
Layer count: For each Rings of Brighthearth...1. Total 35.
Newly carved out space: Green creatures with converted mana cost 1 or less, and sorceries with converted mana cost 2 or less

Section 8

1


Badlands is...well, it's certainly not a bad land per se, despite the name, but it's always been one of the least-used duals. So far you've seen it help power out the early Chaos Warp, and tap for a few more batches of red mana. Both of those things could be done by an ordinary Mountain.

It turns out that there are other considerations going forward that don't allow a basic Mountain to work, but based on the setup so far, if a dual land is going to be used to make red mana then it needs to be Badlands, rather than Taiga, Plateau, or Volcanic Island. The fact that black is the only other color we already have access to at this point means a more rigid order, and a greater number of nested loops, by forcing red to be the only color that's useful for Badlands to generate. In effect, this land offers you a fork in the road: tapping for black mana is equivalent to saying "repeat the lowest 21 layers", while tapping for red means "repeat the lowest 36 layers." I know what I'd choose.

If red mana is so great, why doesn't Sakura-Tribe Scout drop it more often, then? The key is that Scout can only drop a land card that's in your hand, and Badlands is now sitting on the battlefield. It's possible to move it to the graveyard by killing it with Sinking Feeling (or later on, when Cateran Overlord comes online, it'll be simpler to just sacrifice to that), and then Mortuary can put it in the library, but both of those options are stupid. Having a land on top of the library would mean we don't even get to do anything with Psychic Battle! Badlands needs a way to get back to the hand, and I have a pretty good sense of what the next card is going to do...

Section 8 Summary

Input: Badlands entering the battlefield
Output: Lots of red mana
Layer count: For each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Mana Reflection...7. Total 42.
Newly carved out space: Red mana sinks

Section 9

35


Although we don't normally care about the fact that Badlands can produce mana, Fendeep Summoner does care about the fact that it's a Swamp. Note that the ability allows choosing "up to two" Swamps as targets, but you should only choose one (as usual, the nontoken Swamp; all the others are as good as dead by the time you're ready for this step anyway). This is because Grip of Chaos only triggers on spells and abilities that have a single target, so by overreaching with your effort, you would miss out on a whole layer.

Obviously we don't actually care about the nominal effect of animating the land. Nature's Revolt does a good enough job of that already.

Section 9 Summary

Input: One playing of Fendeep Summoner
Output: Lots of playings of Swamp
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 52.
Newly carved out space: Swamps

Section 10

36


Every so often, the insiders at Wizards face a question such as "Why is plainswalk so much rarer than the other types of landwalk?" My theory is that part of this is due to the similar-sounding term "planeswalker," which is a high-profile term, so they attempt to alleviate ambiguity by writing "plainswalk" almost completely out of the game, to the cast-off pile alongside things like land destruction. The official, canonical answer is that it's difficult to come up with a concept for a creature that would exhibit plainswalk, since plains are clear, wide-open areas that aren't easy to hide in.

To date, the best concept for a creature with plainswalk that I've seen is Boggart Arsonists (which, coincidentally, also pulled land destruction out of the cast-off pile). "I dare you to block me while I'm burning the place down as I go!" Silly Goblins...what won't they do?

But this isn't about Arsonists. Rather, it's about their original, Lorwyn-shifted alter-egos, Boggart Loggers. Instead of destroying Scarecrows and Plains, an axe gives them an insatiable inclination to target Treefolk and Forests. And since there are no Forests in the deck, these guys will naturally try to keep targeting Treefolk. In particular, Fendeep Summoner. And as usual, the tree is a coward and will disappear at the slightest sign of danger, before receiving a single cut.

Black mana is already plentiful from the sections before this, so sparing one for a Boggart Loggers activation is practically an invisible operation. Note that Boggart Loggers don't have a tap ability, so they don't take advantage of +1/+1 counters or Sinking Feeling, but it's the best they can do. I figure they're so shocked by the sudden and repeated disappearance of trees in their "lumber yard" that they lose control of the follow-through swing and end up taking their own necks off. I wouldn't put it past a goblin.

Section 10 Summary

Input: One playing of Boggart Loggers
Output: Lots of playings of Fendeep Summoner
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Rings of Brighthearth on the ability, for each Grip of Chaos, for each Psychic Battle...6. Total 58.
Newly carved out space: Treefolk, Forests

Section 11

37


Unlike the other two lands, Goblin Burrows is a land that's used for a purpose other than directly generating mana. It's probably a good thing they hadn't invented the short-lived Tribal type as of Onslaught block, because this is the kind of card that would conceivably be printed as "Tribal Land - Goblin", which (combined with land animation) would allow it to target itself, allowing a loop that extends down to create more Sakura-Tribe Scouts among other things, and those could tap to dump Goblin Burrows yet again. This would be a textbook case of going infinite, and it shows how we have to be more and more careful as the minefield is slowly filled up with "no trespassing" signs. And, for that matter, mines.

I think the Burrows are where Goblins go to reattach their severed heads and sharpen their blades to be even more self-destructive. Or something like that. Red mana is of course plentiful with section 9 a sharply distant sight below, and of course colorless mana is the most abundant of all, so the mana portion of the cost here is again negligible, and you can use all the copies to get more hallucinating, axe-murdering goblins on the table, which is just what everyone needs. Isn't it?

Section 11 Summary

Input: Goblin Burrows entering the battlefield
Output: Lots of playings of Boggart Loggers
Layer count: For each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...9. Total 67.
Newly carved out space: Goblins

Section 12

38

39

This next step is complicated, and we have to skip over a step in the tightly controlled chain that links one card to the next, only to come back and pick it up in reverse. Play Skirk Alarmist, get it countered, get some tokens...yada, yada, yada. Basically the Alarmist needs to be on high alert, and in a position where it can use its ability (come to think of it, this is the one card in the deck that has innate haste, so even the original can tap right away). When it's ready, respond to everything with Dwarven Blastminer, but play it face down.

Dwarven Blastminer is the only creature spell that is not going to get countered by Cephalid Shrine. Since it's face down on the stack, Bloodbond March would be unable to find anything--cards aren't face down in the graveyard, and even if they were, it's impossible to "share" a name with a nameless card, just like Sever the Bloodline played to a board with two face-down creatures will only get rid of one of them.

The reason to play Dwarven Blastminer face down is so it can enter the battlefield face down, getting some Minion Reflector triggers. If you make a token copy of a face down creature, this is amazingly unproductive: the token will be considered face up, but will still be 2/2 (or in this case 6/6) and have blank characteristics otherwise. The important thing is that as a face down creature, it becomes a legal target to activate Skirk Alarmist. Get Cowardice, Grip of Chaos, and Psychic Battle as usual, but respond to Cowardice by paying a red mana to turn Dwarven Blastminer face up. Now there are no face down creatures on the battlefield whatsoever, so Grip and Psychic Battle have no effect, and Cowardice bounces Blastminer.

It's an unfortunate loss, but all this is necessary to get the stack cleared down to the Minion Reflector triggers on Dwarven Blastminer. Since Blastminer is no longer around, Minion Reflector creates tokens based on last known information, and wouldn't you know it? Just before Blastminer went back to hand, it had the good sense to turn face up, so its "last known information" includes the normal, face up characteristics. So now you get a bunch of Blastminer tokens, which can target Goblin Burrows many times for just one easy payment of {2}{R} and tap! They can target other nonbasic lands too (and there are no basic lands in the deck, so that might well read "any land"), but there's no sense slapping anything other than Goblin Burrows, which sits at a higher level than anything else.

And when all those Dwarven Blastminer tokens have been run into the ground, just get Skirk Alarmist untapped so it's in a position to activate again, then repeat the play-face-down procedure. Those two make such a great tag team.

Section 12 Summary

Input: One playing of Skirk Alarmist
Output: Lots of playings of Dwarven Blastminer, and even more copies of Goblin Burrows
Layer count: For each Bloodbond March on Skirk Alarmist, for each Minion Reflector on Alarmist, for each Doubling Season on a Minion Reflector batch of Alarmist, for each Guilty Conscience on Alarmist, for each Rite of Passage on Alarmist, for each Doubling Season on Rite of Passage for Alarmist, for each Rings of Brighthearth on Sinking Feeling attached to Alarmist, for each Minion Reflector on Dwarven Blastminer, for each Doubling Season on a Minion Reflector batch of Blastminer, for each Guilty Conscience on Blastminer, for each Rite of Passage on Blastminer, for each Doubling Season on Rite of Passage for Blastminer, for each Rings of Brighthearth on Sinking Feeling attached to Blastminer, for each Rings of Brighthearth on Blastminer's ability,for each Grip of Chaos on Blastminer, for each Psychic Battle on Blastminer...16. Total 83.
Newly carved out space: Nonbasic lands, face down creatures

Section 13

40


There are some facets of Onslaught block's tribal design that we may never get a good explanation for. If I heard that someone hailed from the Skirk Ridge and had a knack for spreading crazy conspiracy theories, my first inclination would be to believe that it was a goblin. But no, Skirk Alarmist is a Wizard. So much the better then, since it means Alarmist can be included in the deck without falling afoul of Goblin Burrows.

Better still, the Wizard type means it can be picked up by Ghosthelm Courier. Similar to Nectar Faerie, this costs a color of mana plus a tap, but it's capable of targeting itself if necessary so that the tap isn't an issue.

There are a couple things to point out by now. One is that Izzet Guildmage (remember him?) has a blue ability in addition to the red ability, and by using Ghosthelm Courier as the go-to outlet for blue mana, we lose the chance to get double duty from Izzet Guildmage's two abilities. We also lose the ability to make use of cheap instants higher up, since Guildmage can grab hold of them and spiral the ship out of control, out of this universe. Fortunately, since the deck was built with enough planning to foresee (Not the Future Sight (Not the Onslaught (Not the Exodus One) One) One) the consequences of these card choices, it's a simple matter to steer clear of them.

It's also notable that Ghosthelm Courier has another one of those toughness-boosting abilities. In fact, even in a world without Cowardice, all four other members of the Courier cycle are capable of going infinite (provided they're placed above the appropriate color of mana): it takes 3 copies of a Courier to start a cycle where each one is capable of supporting the other two with plenty of Rings of Brighthearth copies, and they keep building each other up higher and higher, never allowing themselves to slip back to the creeping tide of Sinking Feeling counters. But Ghosthelm is unique--as long as it's tapped down, the target has shroud, and any attempt to create another copy of the boost on the same target will fizzle. The only way to take away the shroud (untap Courier) also takes away the +2/+2, so no matter how many Ghosthelm Couriers or how much blue mana you have, none of them can ever subject a creature to more than one +2/+2 boost at a time. In that light, +2/+2 is clearly inferior to a bounce, so we might as well get the full power from each bounce.

Section 13 Summary

Input: One blue mana and Ghosthelm Courier
Output: Lots of playings of Skirk Alarmist
Layer count: For each Rings of Brighthearth, for each Grip of Chaos, for each Psychic Battle...3. Total 86.
Newly carved out space: Wizards

Section 14

41


Okay, so now we need blue mana to power Ghosthelm Courier, the third color sink to be refueled. For that--

Hold on, stop the presses. I just received word that there's going to be a targeting ability that actually does what it's supposed to do. See, none of the lands produce blue, and Tidal Warrior is out to change that, but similar to what happened with Goblin Burrows, it takes a lot of jockeying to make that happen.

When you're in need of blue mana...well, the first time that happens, play Tidal Warrior, and get some tokens. Keep going through the motions of the previous sections, and on the final trip to section 5 before running out of fuel, with the final Vault of Whispers token, stop when the Clash of Realities triggers are just about to have the token start dealing damage, and when it's all suited up with a full complement of Guilty Consciences. Make Cowardice go away at that point, and have Tidal Warrior tap to target that last land token, ignoring Rings of Brighthearth, Grip of Chaos, and Psychic Battle--the whole lot.

When the ability does resolve, Vault of Whispers turns into an island. It doesn't stop being an artifact, but it does lose its innate ability to tap for black, and in its place it gains the power to tap for blue mana. As it turns out, this is much more helpful, picking up Ghosthelm Courier and reigniting the flame with a far bigger blast...of water? Sure, it's magical water, why the hell not.

Because the newly infused land doesn't take full advantage of its ability until after the Guilty Consciences start piling on some damage and the Rites of Passage pile on even more counters, the stack will have to clear at least down to that level before moving on with the usage of blue mana. This is why you don't get to take advantage of the Rings of Brighthearth on the "Target land becomes an island" ability. When you go to untap Tidal Warrior with Sinking Feeling, you can still feel free to use the Rings on that, because all the lower layers become nested on top of it on the stack, and will be entirely cleared by time you need to pay for Rings again.

Tidal Warrior is another one of those creatures that only gets one round of Bloodbond March, and thereafter doesn't come from the hand. Oh well, it was good while it lasted.

Section 14 Summary

Input: Tidal Warrior entering the battlefield
Output: Lots of blue mana
Layer count: For each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience on Tidal Warrior, for each Rite of Passage on Tidal Warrior, for each Doubling Season on Rite of Passage for Tidal Warrior, for each Rings of Brighthearth on Sinking Feeling attached to Tidal Warrior, for each Guilty Conscience on the land, for each Rite of Passage on the land, for each Doubling Season on Rite of Passage for the land, for each Rings of Brighthearth on Sinking Feeling attached to the land, for each Mana Reflection...13. Total 99.
Newly carved out space: Lands, blue mana sinks

Section 15

42

43

44

45

Compared to some of the other venerable tribes in Magic's history, Merfolk are actually pretty difficult to work with while sticking to the confines of a deck like this. Streambed Aquitects and Merfolk Sovereign both immediately go infinite by targeting themselves, and while Sygg, River Guide does exist as a method of jumping to another color, that's going to be awfully wasteful of design space. We can do better, with a little hired help in the form of...Mercenaries.

Yes, Nemesis had a three-card mini-cycle designed with the aim of--and these are the words that Wizards themselves used to hype the cards--"giving blue, green, and red each a Rebel or Mercenary to call their own" by fetching a common creature type in that color. We've already hit Goblins and Elves, eliminating any role for Moggcatcher or Skyshroud Poacher, but if you'll remember, there's one card that's been waiting in the library all this time, and that card is Seahunter: an innate Mercenary whose role is to fetch Merfolk. Bring back Mortuary, kill off the original Tidal Warrior, and send it back in to be fetched.

So Seahunter needs to be fetched out of the library before we can run any more loops, but if there's one thing Mercenaries are good at, it's the process of buying out their comrades who doesn't ask for as much money, or mana, or whatever. Seahunter itself costs 4, so we can't start all the way down the Mercenary pecking order, but Cateran Enforcer does specialize in fetching 4-mana Mercenaries, and it has the good sense to cost 5 itself, making it unfetchable by its own effect (in contrast, Rebels search up the mana cost line, and every creature with a Rebel-fetch ability is capable of fetching copies of itself, so all of them are out of bounds). Cateran Enforcer can in turn be fetched by Cateran Slaver who costs 6, and the top of the totem pole holds the 7-mana Cateran Overlord. Each one raises the bar by a single mana after squeezing out several layers, making this an extremely efficient allocation of design space.

Unlike the other Mercenaries, Cateran Overlord can pick up regeneration shields (and clear the board of exhausted creatures at the same time). While everything else has to stop taking counters from Sinking Feeling when it gets 1 toughness away from having lethal damage marked, the Overlord can run itself all the way down to 1 toughness, period. Regeneration removes all the damage marked on it...and taps it, but you just have to untap it again complete with Rings of Brighthearth copies and be on your merry way. +1/+1 counters from Rite of Passage will completely dominate the relative pittance of damage that gets put on by Guilty Conscience regardless, but this is just an extra little something to take advantage of where the circumstances allow it, and why would you ever pass up a tiny opportunity like that?

Because this is where Seahunter finally comes out of the library, we need a way to put something else back in, lest all the Psychic Battle triggers get voided. Perhaps the simplest way to do this is by using a Copy Enchantment token to copy some static enchantment (probably Nature's Revolt or March of the Machines), then use a resolving Allay to kill off the original version in the presence of Mortuary. Even though it's an enchantment card in the graveyard, Mortuary still puts it in the library, and you can continue merrily winning every Psychic Battle for the rest of the game against a goldfish. Because that's totally an achievement.

Predictably, all this Mercenary fetching means there won't be much in the way of Bloodbond March at this level. Cateran Overlord does get to be cast from hand every time, but Enforcer and Slaver only get cast once each, and for Seahunter...zero times. Oh well.

Section 15 Summary

Input: One playing of Cateran Overlord
Output: Lots of copies of each of the other Mercenaries in turn, finally culminating in Seahunter and Tidal Warrior
Layer count: For each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability (repeated 4 times), and also for each Bloodbond March on Cateran Overlord...29. Total 128.
Newly carved out space: Mercenaries with converted mana cost 6 or less

Section 16

46

One thing about being a card company is that Wizards of the Coast has to print out physical cards. As such, producing a set with an infinite number of distinct cards would be an ill-advised idea. Instead, they make sets with a couple hundred cards, which means they have to stop somewhere.

In the case of Mercenaries, Cateran Overlord is that stopping point. There's nothing that fetches "Mercenary cards with converted mana cost 7 or less", let alone a continuation to higher numbers like 30. If they did, one possible plan in a deck challenge like this would be to clog as much of the deck as possible with nothing but Mercenaries, and coast on their easily-interlocking nature. This turns out not to be the best strategy, since each one is only worth 7 layers while peak efficiency under normal circumstances can get up to 10 layers per card; nonetheless, it's a fun thought.

So we have to try a different approach. Luckily, Mercenary Informer is a card that does exist, and it hits any Mercenary regardless of converted mana cost...including itself, but there's no need to waste activations on that (unlike Nectar Faerie or Ghosthelm Courier, it doesn't tap). As is typical for permanents with an ability whose cost only involves mana, it's redundant in multiples and you won't be paying for Minion Reflector.

Once you get hold of some white mana, pay one of that (along with some more spare colorless) to activate the ability, targeting Cateran Overlord (in nontoken form, as always--actually, this ability can't target a token even if you want it to). Putting it on the bottom of its owner's library isn't very helpful, but again Cowardice comes in handy.

It sounds great, if only there was a source of white mana to do that with...

Section 16 Summary

Input: One white mana and Mercenary Informer
Output: Lots of playings of Cateran Overlord
Layer count: For each Rings of Brighthearth, for each Grip of Chaos, for each Psychic Battle...3. Total 131.
Newly carved out space: Mercenaries (regardless of converted mana cost)

47

Tundra Kavu certainly seems like an odd choice to do anything sentient, let alone a specific task that plugs neatly into its place on one extremely long, beautiful line. But with a text box like that, that's exactly what it does.

In section 14, with Tidal Warrior, you saw what I refer to as the "landshaping" routine. Tundra Kavu does the same thing, but it landshapes for plains. It could do islands as well, yes, but with Tidal Warrior already covering that function at a lower level, it would be a waste of space. Just like tapping a Badlands for black.

Set aside another copy of Vault of Whispers when you need white mana, and the rest is easily if you've followed it this far.

Section 17 Summary

Input: One playing of Tundra Kavu
Output: Lots of white mana
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience on Tundra Kavu, for each Rite of Passage on Tundra Kavu, for each Doubling Season on Rite of Passage for Tundra Kavu, for each Rings of Brighthearth on Sinking Feeling attached to Tundra Kavu, for each Guilty Conscience on the land, for each Rite of Passage on the land, for each Doubling Season on Rite of Passage for the land, for each Rings of Brighthearth on Sinking Feeling attached to the land, for each Mana Reflection...14. Total 145.
Newly carved out space: White mana sinks

Section 18

48

If there's one thing Tundra Kavu advertises about itself right in the name, it's that Tundra Kavu is a Kavu. (I guess it also tries to advertise that it's a Tundra, and I'd love to pull my copies out of the junk box and sell them to Star City for 70 bucks apiece, but the folks there don't seem to believe me. Maybe getting them to read this page, and gain awareness of what it can do, will change all that.)

You know who likes seeing Kavu around? The Alpha Kavu, head honcho of them all but with a remarkably pacifistic tendency. We've seen this several times before: spend one mana of a new color, target a creature of the previous tribe, and get a few layers out of it.

Alpha Kavu's effect is another one of those toughness boosts that we have to be careful with. In fact, while we were compiling this deck, the original plan was to close out on Crystal Spray, and partake in a bit of "creature-shaping" similar to how Tidal Warrior and Tundra Kavu perform landshaping. But it turns out that Crystal Spray can change land types as well as color words, and if you rewrite a Tundra Kavu token to say "forest" and combine it with the use of Alpha Kavu for toughness instead of bounce, you'd get an infinite loop! This inadvertent combo had evaded all of our examinations until I was well into writing the outline of the deck--until I got to this very paragraph, in fact. It just reinforces that point that in decks like these, everything needs to be double checked or even more, lest you discover something nasty, maybe even intractable.

Section 18 Summary

Input: One green mana and Alpha Kavu
Output: Lots of playings of Tundra Kavu
Layer count: For each Rings of Brighthearth, for each Grip of Chaos, for each Psychic Battle...3. Total 148.
Newly carved out space: Kavu

Section 19

31

So finally, after puting colorless mana and all four other colors to good use, we come to the last mana sink. Overgrown Battlement is certainly an appealing choice to be the green mana producer, obviously better than Vine Trellis for instance, and a full layer better than lands (although lands have already been thoroughly closed off by this point). Too bad the bonus mana it gets for each other copy of itself really isn't all it's cracked up to be. Mana Reflection, for instance, far outshines the contribution of that phrase.

We already heard from Overgrown Battlement earlier, when it produced the first green mana necessary for hard-casting Green Sun's Zenith, but now it gets to see its proper place in line, and soon it will actually start getting reused.

Section 19 Summary

Input: One playing of Overgrown Battlement
Output: Lots of green mana
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Mana Reflection...8. Total 156.
Newly carved out space: Green mana sinks

Section 20

49


One of the reasons Overgrown Battlement is worded the way it is, is because Rise of the Eldrazi included a defender sub-theme, a conscious attempt to stall games out long enough for the eponymous Big Bad Guys to show up and wreck things. Funny, after working with this combo they don't look nearly so big (puny 8/8s and 15/15s), but maybe that's just me. Anyway, ever since Kamigawa block in 2004-2005, Wizards of the Coast has been decoupling the "defender" keyword from the "Wall" creature type, and to a more overarching extent, trying to stop producing new Wall creatures altogether. Nonetheless, the concept for Overgrown Battlement had it being a Wall, so that's what it's printed as.

That's the cue for the Dwarven Demolition Team to enter, stage right. I don't know if any of those dwarves are actually named Gorbachev, but I do know they have a knack for tearing down walls of all shapes and sizes. They don't actually get torn down, of course...only returned to hand, but after seeing everything that's played out so far, would you expect anything less? Disappearance defeats dwarves' destructive deeds, next on Sick, Sad World.

Section 20 Summary

Input: One playing of Dwarven Demolition Team
Output: Lots of playings of Overgrown Battlement
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 166.
Newly carved out space: Dwarves

Section 21

50


Like I was saying about a sick, sad world, Dwarven Pony. Seriously, this card exists. Someone, somewhere, decided it would be a good idea to call up Cartamundi over in Belgium and get them to spend their valuable ink and cardstock to print this card. And people wonder why Homelands is such a reviled set.

Oh well. If they're going to give us tools like this, we might as well use them. On the surface, Dwarven Pony looks like it has an utterly uninteresting ability: target Dwarf gains mountainwalk. (It would almost make sense if it granted horsemanship, but they weren't about to introduce a new sought-after keyword into a set like Homelands. Not even under the guise of calling it a stale rehash of an old keyword.) And in any other deck, that ability would be uninteresting indeed. But the fact that it has "Target Dwarf" makes it a valuable tool for us to latch onto. {1}{R} is once again a pittance of a cost, and the pony keeps inviting the dwarf to ride, only to find that the would-be rider is back in its owner's hand, whatever that means. And then for some reason it can't resist the allure (a PONY! Who can count to SIX!) and tries to come right back and bring some friends along, to no avail other than producing lots of Overgrown Battlements. I don't even know what it's up to any more.

Are we done yet? Of course not.

Section 21 Summary

Input: Dwarven Pony entering the battlefield
Output: Lots of playings of Dwarven Demolition Team
Layer count: For each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...9. Total 175.
Newly carved out space: Dwarves

Section 22

51


Thankfully, "target Horse" is not a string that appears in the text of any Magic card. As far as the combo is concerned, that means we have to reuse Dwarven Pony another way.

Start out with a well-deserved ritual sacrifice: throw away the nontoken Dwarven Pony to give Cateran Overlord a regeneration shield. Good riddance.

Or should I say bad riddance, because the next thing to come along will be Driver of the Dead. This doesn't have a tap ability, so it's far less efficient than a lot of the other cards we've seen so far. Still, once Driver has tokens in place, sacrifice one of those to Cateran Overlord, getting a death trigger.

Grip of Chaos and Psychic Battle will trigger, of course, but since Driver of the Dead targets a card in a graveyard instead of a permanent, those triggers don't do anything meaningful. After going through them all, the ability will resolve and return Dwarven Pony, now apparently sporting a new undead look. "Maaaaaaaaaanes."

Thankfully there are only 9 more cards to get through. Why are we even here any more? Seriously. Sooner or later most readers will undoubtedly pass off the numbers as "big enough" and turn their attention somewhere else, instead of trying to trudge through all this.

Section 22 Summary

Input: One playing of Driver of the Dead
Output: Lots of copies of Dwarven Pony
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch...3. Total 178.
Newly carved out space: Creatures with converted mana cost 2 or less

Section 23

52

Again, Driver of the Dead is quite inefficient compared to most of the cards being used, especially up near the top. The one saving grace it has is a creature type of Vampire (as though vampires can demonstrate any grace), which provides a hook into a sequence that would otherwise be difficult to access.

You know who else is a Vampire? Baron Sengir, though he didn't get that status until the Grand Creature Type Update of 2007. Since he can tap to target (another) Vampire, that means...well, it means there's a very good reason not to include Mirror Gallery in the deck. If a Baron token could be allowed to target its nontoken self, that's an infinite loop with a very short fuse. However, in a world where we follow the rules and can only keep one copy of a legend at a time, we can actually pull this off so this Black Baron provides yet another large, but well-bounded, contribution.

We have to make sure that no token copy can ever repeatedly target the nontoken. Clearly, both versions of Baron can't coexist on the battlefield for any length of time--if you have priority to activate an ability, then you've been forced to whittle the numbers down to one Baron already, and it had better be the one you're activating. The activation will set off a bunch of fun stuff like Psychic Battle and Rings of Brighthearth, and all of this will invariably take place on top of the next Bloodbond March trigger, so all of it has to clear from the stack before the nontoken Baron can return. There's no danger that a Psychic Battle might be capable of changing the target in a circular pattern. At least, not without help from an outside impetus at a higher layer, by which time you can rightfully return Baron Sengir anyway.

So it's settled then: all Baron Sengir does is slap Driver of the Dead, in a loose approximation of a vampire-eat-vampire world. There is one more trick that can be used for a tiny little boost: after Baron has received the last batch of +1/+1 counters it will ever be able to get in its current incarnation, and spent its way through as many as possible...remember how Clash of Realities kept causing Baron to deal damage to a Figure of Destiny token? Yeah, take that token and sacrifice it to Cateran Overlord, then spend 1 red mana to pick out another copy of Figure of Destiny and turn it into the lucky Spirit from then on (if you run out of Figure tokens, Driver of the Dead can spare one copy of itself to bring back Figure and create even more tokens). Even though it's a token, Figure of Destiny was dealt damage by Baron Sengir, and it did die this turn, so Baron Sengir triggers on the sacrifice, and gets...not +1/+1, but a +2/+2 counter! And since Doubling Season is not Primal Vigor, they can happily double the +2/+2 counter, giving the Baron one more toughness buffer to continue on his bloodthirsty rampage before he gets tired and takes a brief nap.

Section 23 Summary

Input: One playing of Baron Sengir
Output: Lots of playings of Driver of the Dead
Layer count: For each Bloodbond March, for each Minion Reflector, for each Clash of Realities, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 188.
Newly carved out space: Vampires

Section 24

53

The important thing about Baron Sengir is that he's legendary, which is perfect for Tsabo Tavoc. We would have liked to add Empress Galina in between the two, another card that's only allowed because of the lack of a Mirror Gallery, and furthermore the lack of haste on nontoken creatures (otherwise Galina could simply target herself and be replayed). She turns out to be unacceptable because Seahunter already cuts off the space of Merfolk, but anyway. All colors of mana are abundant by now, so use Tsabo to attempt to slay Baron outright, and look on completely unsurprised as the attempt falls short and the vampire winds up in hand again. (And again and again.)

Protection from Legends is a unique ability among all cards in existence. Even the token copies are incapable of targeting themselves, not that bouncing a token is useful at all. It would have been more important if Galina had made the cut, ensuring a one-way road where Tsabo could slap Galina but the reverse was impossible because of protection. It still gains several layers while consuming almost no design space, setting things up amazingly well for the final sprint.

Section 24 Summary

Input: One playing of Tsabo Tavoc
Output: Lots of playings of Baron Sengir
Layer count: For each Bloodbond March, for each Minion Reflector, for each Clash of Realities, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 198.
Newly carved out space: Other legendary creatures

Section 25

54


Willow Satyr is another card that targets legends, allowing us to squeeze even more efficiency out of this supertype that's unique in more ways than one. The relevant piece is that Willow Satyr isn't legendary itself, which allows it to target Tsabo without being targeted in return, by either Tsabo or itself. (The list of legendary creatures in Legends is exactly the same as the list of multicolored cards in that set.)

I guess there are still cards like Progenitus that can't be targeted by any of these last three sections, but suffice it to say that card isn't going to do anything to help.

Section 25 Summary

Input: One playing of Willow Satyr
Output: Lots of playings of Tsabo Tavoc
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 208.
Newly carved out space: Legendary creatures

Section 26

55


The good news right now is we're running out of card slots that need filling, which means we can start painting away broader swaths of design space. There's very little that stands out about Willow Satyr other than the fact that it's green. Well, it is a Satyr, but prior to Theros block there was only one other Satyr in the game (at least among those that aren't also Cat Rat Elephants), and even now with the sudden influx of Satyrs, they still aren't showing any signs of tribal support, especially not in the ways that workfor this deck. So let's go ahead and cut off the entire color of green: King Crab is as good a card for the job as any, reminding us of the power that enemy colors used to be capable of inflicting on each other.

Section 26 Summary

Input: One playing of King Crab
Output: Lots of playings of Willow Satyr
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 218.
Newly carved out space: Green creatures

Section 27

56


And if King Crab looked like an enemy hoser, then one look at Spinal Villain...wow. They sure don't make cards like that any more, but this one has already been printed, so I'll take good care of it and teach it how to slap a lot of crabs. Maybe take it to an underwater party and show it off as a neat trick, assuming everyone isn't immediately scared away by its mere sight.

Spinal Villain's creature type (after errata) happens to be Beast, which has a mild amount of tribal support. Nothing that can be used this high up and with this few card slots remaining, though.

Section 27 Summary

Input: One playing of Spinal Villain
Output: Lots of playings of King Crab
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 228.
Newly carved out space: Blue creatures

Section 28

57


So we're in the color-hosing sequence, and need to go after red now to continue drawing a pentagram on the color wheel. Southern Paladin is the obvious choice that comes to mind, and there's not really much interesting to say about it. Just another card humming along at peak efficiency, all in a good day's work.

Section 28 Summary

Input: One playing of Southern Paladin
Output: Lots of playings of Spinal Villain
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 238.
Newly carved out space: Red permanents

Section 29

58


Southern Paladin, of course, was made as an offshoot of Northern Paladin from the very first set, after people had been asking for years why such a card didn't exist. Fast forward another year and a half, and Wizards of the Coast decided to complete the compass with (almost-)mirrors in black: the Wicked Paladins of the West and East. These black versions only say they target "creatures", instead of generic "permanents", not that it matters because anything we might ever want to target here has already been turned into a creature. So the thugs from the west coast want to fight the thugs from the...south coast? Sure, why not. Boing, boing. Gang warfare with rubber bullets, very funny.

Section 29 Summary

Input: One playing of Western Paladin
Output: Lots of playings of Southern Paladin
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 248.
Newly carved out space: White creatures

Section 30

59


Fifty-eight cards down, two to go. As you know, almost all incoming creatures get tokens from Minion Reflector, so what better way to keep up the trend than by putting Minion of Leshrac so close to the top?

Like Tsabo Tavoc, Minion of Leshrac takes advantage of the "protection from itself" trick to declare a broad targeting condition--in this case, any creature or land--without falling afoul of it. It swings at the Western Paladin with a rubber mallet, I'm sure, and the crowd goes wild. Assuming they haven't all fallen asleep by now. Or died of old age, given how long this combo would take if you tried to play it out in reality.

Of course, 59 is not 60, so there's still room for one more card. But what can we possibly fit after taking such a broad wrecking ball like this to the landscape?

Section 30 Summary

Input: One playing of Minion of Leshrac
Output: Lots of playings of Western Paladin
Layer count: For each Bloodbond March, for each Minion Reflector, for each Doubling Season on a Minion Reflector batch, for each Guilty Conscience, for each Rite of Passage, for each Doubling Season on Rite of Passage, for each Rings of Brighthearth on Sinking Feeling, for each Rings of Brighthearth on the tap ability, for each Grip of Chaos, for each Psychic Battle...10. Total 258.
Newly carved out space: Other creatures

Section 31

60


Well, that's a nice trick. Devout Lightcaster is a creature, and--yes--a white creature, which causes Minion of Leshrac, and Western Paladin both to perk their ears up with interest. The creature types Kor Cleric haven't been touched yet, nor has converted mana cost 3, so those two cards are the only ones that think they have a basis for targeting it. Conveniently, both are black, so the Lightcaster scoffs at them as its protection is enough to stop any upward loops from being closed.

There is a slight snag with Devout Lightcaster in that it uses an enters-the-battlefield trigger, instead of an activated ability. The fact that it targets makes it especially problematic, because each time you get a batch of copies from Minion Reflector, all the targets have to be chosen at once and that means all the Psychic Battles are locked in at that point, with no way to increase the number that go off between Copy #1, Copy #2, and so on. It's better than nothing, and besides, I've had just about enough tapping, untapping, and bouncing for one deck.

Section 31 Summary

Input: One playing of Devout Lightcaster
Output: Lots of playings of Minion of Leshrac
Layer count: For each Bloodbond March, for each Minion Reflector, for each Grip of Chaos, for each Psychic Battle...4. Total 262.
Newly carved out space: Black permanents

Endgame

Savage Beating
(This card is not actually included. Only the sentiment.)

These numbers are big enough that they defy description by any means other than boring, rote expressions (which are even underestimated by a good deal). But finally, after going through all 60 cards, and somehow dodging penalties for Slow Play and Stalling in the process, everything that can be done, has been.

Except for one minor little thing. So far, the Clash of Realities, as well as everyone's Guilty Conscience, has caused a good deal of friendly fire to rain down on our own creatures (which they all shrug off), but other than that, there's been exactly 0 damage dealt over the course of the combo. Now we can finally move to the combat phase, and swing with a massive army of stuff: most of the creatures have tap or sacrifice abilities and will have run themselves into exhaustion, but the enchantments are still at full power: Doubling Season, Rite of Passage, Mana Reflection, Grip of Chaos, Psychic Battle, Bloodbond March, Clash of Realities, some artifacts like Minion Reflector and Rings of Brighthearth, and even a bona fide creature in Figure of Destiny that's been receiving potshots from absolutely everything. Rite of Passage dumps a good share of +1/+1 counters on Figure, which don't really make a visible difference since they can't be meaningfully distinguished from the number of Doubling Seasons (the most frequently-recurred permanent, with all those visits to Section 1).

Unfortunately, Devout Lightcaster can only be played once, and once it's out of Bloodbond March triggers, that's all. It would have been nice to use Yawgmoth's Bargain in place of Enter the Infinite, and Reito Lantern in place of Mortuary; then the initial deck would have to be drawn one at a time, but in return we'd be able to use a bunch of spare card draws at the end and redraw Chaos Warp every time, stacking two more layers on top of Devout Lightcaster. This fails because of Nectar Faerie's ability to grant lifelink, combined with all the Clash of Realities triggers that cause creatures to deal damage.

So how much damage is that? If you've been keeping track along with me, you'll see that my count comes to 262 nested layers of recursion. That means we're looking at a number of the form 2^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^^^^^^?, where the number of arrows is 262, and ? is essentially the number of Rite of Passage copies that trigger on the final burst of damage to Figure of Destiny. Of course this number is really big itself, and since writing over 200 arrows looks downright ugly, I vote that it's time to move past Mr. Knuth's notation, and on to a variation thereof developed by another mathematician, John Conway (yes, he of cellular automaton fame). Conway's system is based on right arrows instead of up arrows, and for now all you need to know about it is that the expression "a -> b -> c" means the same thing as "a ^^^^...^^^^ b, with a total of c arrows between them". (This notation can also be extended to expressions that use 3 or more right arrows, but those are beyond the scope of this page, and I doubt the system of Magic will ever be expressive enough to include the types of constructs that would call for their use, outside of cases that truly do go infinite.)

Instead of that long string of up arrows, then, we can write 2 -> ? -> 262, where ? is a number that is itself rather close (in a holistic sense) to the number of permanents; that number is also of the form 2 -> * -> 262. * is smaller than ? (it's controlled by the number of times the Devout Lightcaster was able to bounce Minion of Leshrac, creating more copies of everything else), and the Devout Lightcaster loop was bottlenecked by a number of Bloodbond Marches that were fixed as of the end of the first Minion of Leshrac step, so the best estimate I can give for * is a number of the form 2 -> # -> 258.

In any case, * is certainly bigger than 4, and since 4 can be expanded out to "2 -> 2 -> x" for any x value whatsoever, the damage potential can be expressed as something no less than 2 -> (2 -> (2 -> 2 -> 262) -> 262) -> 262. This is the Conway equivalent of the Knuth expression "2 ^^^... 2 ^^^... 2 ^^^... 2", which means we can collapse it and go up a level, expressing it as 2 -> 4 -> 263. (By the same token, it can also be expressed as 2 -> 3 -> 264, but advancing to the 264th layer is certainly unjustified at this point, and the apparent extra jump is merely a consequence of the small numbers involved at the top.)

Since there's no more recursion of anything beyond that, it looks like that's our final damage figure: about

2 -> 4 -> 263.

Even a googolplex, a number commonly cited as being absolutely monstrous, is nothing but a rounding error when compared to 2 -> 3 -> 4. That still doesn't make it very clear just how big 2 -> 4 -> 263 is, but it might give you a good idea how powerful and compact Conway's notation can be even at low values, and one thing's for sure. I wouldn't want to take 2 -> 4 -> 263 damage in a Magic game any time soon. Especially not on turn 1. Let's now give a moment of silence to the hapless goldfish opponent who had to suffer such a fate here.

Meanwhile, Elspeth has been sitting quietly this whole time, gazing upon the overcrowded battlefield with one solitary loyalty counter as though it's symbolic or something. Heck of an ultimate this has been.

Okay, time's up. Proceed to the second combat damage step.

"Game over. User wins."


That's about all that can be done, unless future sets bring out a card that can be used to improve this result while still not going infinite. Or, who knows. Maybe there's another hidden treasure already out there, waiting for you to find it and change around the combo to do even more! My first approach to writing a deck overview in this strain, as linked up at the top, came to 113 layers. Little did I know that the technology already existed to get over twice as much recursion, and average more than 4 full Knuth arrows for every single card in the deck! So will you get to stake your claim? You won't know until you try.

Oh, and there's also a planned version in the works that will go after the challenge with a larger deck, such as 100 cards with Commander rules. You can try that too.

If you have any questions for the authors or ideas for improvement, the best place to discuss them would be on Wizards of the Coast's own forum, in the established topic here.

Until next time, may your multi-layer combos prove as delicious as a multi-layer cake.