Commander Focus: Tiamat, a rage of dragons

An article by    Competitive Play Gaming Magic the Gathering        0

Credit: Wizards of the Coast

Tiamat is one of those times that Gary Gygax & co decided it was cool to more or less co-opt something that already existed in real life. There’s even a first edition book that’s just D&D interpretations of various pantheons of historical gods. I actually really like the concept, because frankly, it teaches people, in the same way that people who play Civilization actually learn about world history. And that’s a fine upside to our made-up game about dragons, dungeons, and the dangerous murderhobos that pillage them. And while in the larger sense the whole Dungeons & Dragons crossover for Magic is essentially a marketing stunt/gimmick, there’s some upside to it as well. Specifically, having Tiamat ported from Babylon to Forgotten Realms to Dragonlance and back and then into a universe with people like Jace, where the Dragon god can eat him.

Also sometimes it’s just fun to build a deck around a nearly nonsensical commander who seems like a nearly unworkable card.

By Ben Pirard at nl.wikipedia, CC BY-SA 3.0,

With Tiamat we’re talking a 7/7 flier here but the real strong ability is to tutor 5 cards, which is an ETB but conditional on casting Tiamat. This means: you can’t reanimate Tiamat but also if your opponents counterspell or whatnot, you won’t get your search. It’s a very battlecruiser ability, and while a 7/7 flier is nothing to exactly sneeze at, especially as a commander, it’a also nothing to write home about because, if someone is going to let you get 7 mana and then hit them 3 times without doing anything about it, that’s a pretty non-interactive death, and frankly they deserve to pay dearly for their crimes.

Credit: Wizards of the Coast

In a 4-of format, you’d see the “up to five Dragons cards not named Tiamat that each have different names” as probably something of a downside. If you could fetch more Tiamats you’d get to keep grabbing dragons. If you were grabbing Goldspan dragons or something you might be able to set up a scenario where you can cast all of them, attack with all of them, maybe generate attack steps or something… fun times. You can’t do either with Tiamat, so you have to come up with a combo that works because of different dragons, which, given that most dragons are real big expensive creatures, probably won’t work. It’s almost like Wizards of the Coast expected this to be a weird commander.

So what we have is a 7 mana creature but also one whose cost includes WUBRG which is, let’s say “tough” to pull off. But the good news is that there is a really simple way to get there. A way that would make Sheldon Menery clutch at his pearls. Ramp into ramp. Run a ton of land tutors and 2 cmc rocks.

The deck should play out in almost exactly the same way every time: Turn 1 land that can tap for green. Hopefully, you have a burgeoning or a sol ring or a mana dork, but it’s not strictly necessary. Let’s lay out a typical scenario:

Turn 1: Maybe a deadish turn, but this is really a fun “battlecruiser” type deck. See the end of the article for a slightly more aggressive version that includes every mana dork in the book.

Turn 2, 3 mana available: Start with 2 lands, cast either a talisman or land tutor, depending on the tutor and

Turn 3, 5 mana available: Start with 2 -3 lands and 1-2 dorks or a talisman: you should be able to land tutor or talisman here one of the bigger land tutors here, leading to turn 4 or 5 casting Tiamat

Turn 4, 7 mana available + WBURG: Your goal is to turn 4 drop Tiamat.

The whole deck is built to get Tiamat into play so that you can recover your card disadvantage from heavily tutoring out lands, both thinning your deck and letting you recover if and when your dragons get blown up.

Let’s talk about land tutoring strategies:

MTG Land tutor table

Name# landsSupertypeStatusCMCSpecialBest Target
Three Visits1Non-BasicUntapped2ForestAny Forest shockland
Nature's Lore1Non-BasicUntapped2N/AAny Forest Shockland
Farseek1Non-BasicTapped2Not ForestAny shockland
Into the North1SnowTapped2SnowAny snow basic
Rampant Growth1BasicTapped2N/AAny non-snow basic
Emergent Sequence1BasicTapped2ManlandAny non-snow basic
Expedition Map1AnyHand2+11 handCavern of Souls/Boseiju
Wayfarer's Bauble1BasicTapped2+1N/AAny non-snow basic
Cultivate2BasicTapped31 handAny non-snow basic
Kodama's Reach2BasicTapped31 handAny 2 non-snow basic
Pir's Whim1AnyTapped4N/ACavern of Souls/Boseiju
Seek the Horizon3BasicHand43 handAny 3 non-snow basic
Shard Convergence4Non-BasicHand44 handAny shocklands
Tempt with Discovery01 to 05AnyN/A4N/ACavern of Souls/Boseiju

You need to get one of each mana, so your most important lands are: a green source, then Cavern of Souls, which is 5c for dragons and makes then uncounterable, big for getting Tiamat into play for sure so you recover your hand.  Finally, you need to fill out the 5 colors, then you want to grab Boseiju, Who shelters all, if possible, for your big haymakers like Urza’s Ruinous Blast or return all your dragons from the yard.

This table also does a decent job of breaking down the various “good” 2 mana tutors.  The common theme is that they put a land onto the board, with some variance, but the best is probably Three Visits and Nature’s Lore since both allow you to fetch a non-basic and it comes into play untapped.  Expedition map costs 2 and an additional 1, but allows you to fetch any land.

Your early game is ramping into ramp, essentially.

never stop

Midgame will consist of dropping Tiamat and fetching up to 5 dragons.  Fetch 5.  If your hand is too full fetch Eternal Dragon and Ebondeath, Dracolich.  Both have abilities allowing you to recur them, and you want to thin your deck to continue making land drops and ramping.


You probably want to grab Dragonlord Dromoka or Niv-Mizzet Parun, alongside Ramos, Dragon Engine and or Old Gnawbones or Klauth, Unrivaled Ancient.  Dromoka offers protection, and Niv-Mizzet will let you draw a lot of cards most likely, or else eat some removal.  Dragonlord Atarka or Piru, the Volatile can clear the board for you, The Ur-Dragon and Runehorn Hellkite and Iymrith, Desert Doom are significant draw sources.

Someone is probably going to clear the board of you, and the deck expects and is ready for that.  Several dragons either return from the graveyard or else return other dragons to the battlefield.  If by some miracle Klauth, Unrivaled Ancient or Old Gnawbones just sits on the board you’ll generate ludicrous amounts of mana and just drop your whole hand of dragons. Make sure you get at least one draw source so that you can refill your hand (again)


Did I make it clear you should not stop ramping? Even when you want to drop more dragons, wait and drop more ramp instead. It will let you recover and simply generate more threats than your opponents.  Once you can drop 2 big fat dragons, hard casting them, a turn, it’s going to be really tough for your opponents to deal with that. And in general you’ll want to drop a ramping dragon, like Old Gnawbones, rather than something else.

The deck runs minimal interactivity, instead choosing to just be a bunch of huge haymaker creatures and huge haymaker spells, like Patriarchs Bidding.  This is not a particularly good deck design, but it’s kind of fun. Piru, the volatile and Urza’s Ruinous Blast will let you clear the board, and almost every dragon in the deck is a legend, making you immune to both. Ryusei, the Falling Star should also clear out all the non-dragons for you, but only when it dies.

To make the deck faster and more consistent it would be possible to add in 1 drop mana dorks, but then it would lack the fun of huge board clears like Piru, which, for a deck like this, is more fun in my opinion. However, I’m going to also link the alternative, less fun, faster more consistent list, after the “real” list.

This is the list with dorks included, and less recursion and less big spells:

Have any questions or feedback? Drop us a note in the comments below or email us at


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.