A few months ago I wrote a series about how to get into Legacy. It focused on identifying pivotal cards that players can buy and build decks around, playing competitive Legacy matches throughout the process of building a deck. I wrote about format staples Lion’s Eye Diamond, Wasteland, and Ancient Tomb, laying out a pathway from owning a playset of a given card to owning a complete competitively viable deck.
Most of the decks I discussed are known quantities—Dredge, Storm, Delver of Secrets tempo decks with Wasteland, black attrition decks with Wasteland, and Show and Tell decks have all done well at major Legacy tournaments in the recent past. One deck that I discussed has not put up the numbers that other decks have, despite playing four copies each of two previously-banned cards.
It also just so happens that in the informal Facebook poll that I set up last week the color with the plurality of votes was red. Technically, since no option hit my arbitrary benchmark of 50% of total Facebook likes, I committed to writing about artifact decks, but last week’s video was also something close to the most liked video article of StarCityGames.com history, which I didn’t exactly expect. I am a man of my word so this week is about artifacts first and foremost, but I’ll work on giving a good nod to those of you who want to see a red deck in Legacy.
Let’s talk about Metalworker and Grim Monolith.
In 2013, three decks with Metalworker made Top 8 of an SCG Legacy Open. Here they are:
Creatures (20)
- 4 Metalworker
- 1 Sundering Titan
- 4 Goblin Welder
- 3 Lodestone Golem
- 1 Wurmcoil Engine
- 1 Steel Hellkite
- 4 Kuldotha Forgemaster
- 1 Myr Battlesphere
- 1 Blightsteel Colossus
Lands (23)
Spells (17)
Creatures (20)
- 4 Metalworker
- 4 Goblin Welder
- 4 Lodestone Golem
- 3 Wurmcoil Engine
- 1 Steel Hellkite
- 3 Kuldotha Forgemaster
- 1 Blightsteel Colossus
Planeswalkers (1)
Lands (20)
Spells (19)
Creatures (22)
- 4 Metalworker
- 1 Sundering Titan
- 4 Goblin Welder
- 4 Lodestone Golem
- 4 Wurmcoil Engine
- 4 Kuldotha Forgemaster
- 1 Blightsteel Colossus
Lands (20)
Spells (18)
- 4 Grim Monolith
- 1 Crucible of Worlds
- 2 Lightning Greaves
- 2 Voltaic Key
- 2 Mox Diamond
- 4 Chalice of the Void
- 2 Mox Opal
- 1 Spine of Ish Sah
Sideboard
They all have the same rough core. Point by point, let’s start with the mana base:
- Their mana bases all play the maximum number of Wasteland, Ancient Tomb, City of Traitors, and Great Furnace.
- All of them play at least two Cavern of Souls, but can play as many as four.
- Decks with fewer than four Cavern of Souls play four Darksteel Citadel.
- If a deck has twenty lands, it also has two Mox Opal.
- All of them play four Grim Monolith and at least one Voltaic Key, but most have two.
- None of them play more than two Mox Diamond, which may surprise those of you who haven’t looked at an artifact control deck in several years. The card’s stock has fallen a lot, which makes sense in a deck where you don’t want to play more than 23 lands.
So that’s how these decks have built their mana bases. What do their threats look like?
- Every deck has all four Goblin Welder and Metalworker. Goblin Welder is pretty clearly the payoff for playing red in an artifact deck.
- Two of the three decks each play four Kuldotha Forgemasters and four Lodestone Golems, while the odd deck out plays three. It’s also pretty clear that these decks are focused on activating Kuldotha Forgemaster early in the game and disrupting opposing spells via Lodestone Golem.
- There’s a good amount of variation in the remaining threat slots, but options range from Wurmcoil Engine and Steel Hellkite (the castables) to Spine of Ish Sah, Myr Battlesphere and Sundering Titan (the mostly-uncastables).
- All of the decks play exactly one Blightsteel Colossus and two Lightning Greaves. Two of them have Staff of Domination. For those of you unfamiliar with these cards roles, it goes something like this:
Step 1: Play Lightning Greaves on an empty board.
Step 2: Cavern of Souls naming Construct, cast an uncounterable Kuldotha Forgemaster, equip Lightning Greaves, activate Forgemaster (likely sacrificing itself and two non-Lightning Greaves artifacts), and find Blightsteel Colossus.
Step 3: Equip Lightning Greaves to that.
Step 4: Attack for eleven poison. Win.
The long way goes as follows: tap Metalworker, revealing three artifacts and Staff of Domination. Cast Staff. Use three mana to untap Metalworker, use one mana to untap Staff, and proceed to generate infinite mana—six from each Metalworker tap minus four for each Staff tap/untap dance, netting two mana per activation. Draw your deck with Staff of Domination, cast all of your Lodestone Golems, cast your Lightning Greaves, cast your Blightsteel Colossus, tap all of their creatures, and kill them with poison.
No matter how you slice it, the Kuldotha Forgemaster kill is a huge threat. How do decks tend to back it up?
The Stompy style lockdown package never gets fully played. When I use the term “Stompy,” I use it in the same vein as ancient Legacy deck naming convention: Ancient Tomb plus City of Traitors ramp decks with three to five mana creature threats and—relevant to this usage—playsets of Chalice of the Void and Trinisphere. A decent reference point is this list.
Getting back on point, the artifact control decks nowadays don’t play all four Chalice of the Void and four Trinisphere—a decision that I really like. One of the easiest ways to lose to today’s blue decks is to fall behind on board. All it takes is one Stoneforge Mystic and an untap step, one Deathrite Shaman on the play, one Delver of Secrets and a quick flip with a few counterspells and your grip of cards that are good against blue decks have no beneficial card text. We live in a world where successful Legacy decks focus on establishing board presence as fast and powerfully as possible. Chalice of the Void and Trinisphere aren’t that great anymore, and that goes double for a world where you want to play your own Goblin Welder, Voltaic Key, and Mox Opal.
I don’t think Chalice of the Void and Trinisphere are bad—clearly they have value against combo decks that can’t beat you with creatures. I just don’t like them against blue decks that can attack for three on turn two with Spell Pierce backup. They’re just narrower than they used to be, and it’s important to recognize that they aren’t going to give you the sort of free win capability that, say, Blood Moon gives an Imperial Painter deck.
It is telling however,that people building these decks want to get some free wins—two of the three lists have Blood Moon in their sideboards despite having at most eight red sources between Great Furnace, Mox Opal, and Mox Diamond. Adding a few basic Mountains helps cast Blood Moon, but at what point would you rather just be an Imperial Painter with a cheaper two-card kill? Painter’s Servant plus Grindstone plus activation is six mana versus Lightning Greaves plus Kuldotha Forgemaster being seven mana and requiring more mana in play at once.
I see you there, rhetorical audience member device. I see you. Yeah, go on. I know you want to.
Hey Drew, you tell people to not play a worse something else all the time. Why is a Metalworker deck not a worse Painter’s Servant deck? It has worse disruption—Lodestone Golem and maybe Chalice of the Void and maybe Trinisphere versus Blood Moon, a tutorable Magus of the Moon and Phyrexian Revoker. It has a worse kill, which you just went over. And finally, it has a worse plan B. Painter can play Koth of the Hammer or Chandra, Pyromaster or Jaya Ballard, Task Mage, whereas Forgemaster has to sit around and wait for its draws to sort themselves out. What gives? Why talk about this? Just play Painter, right?
Well, not so fast. Those are all excellent arguments for playing Imperial Painter, but that’s not where the story ends. It’s not like a Metalworker / Goblin Welder deck can’t also play Sensei’s Divining Top; it just hasn’t because people have defaulted to historic tendencies involving quad Chalice of the Void, and we’re only now beginning to understand that the card isn’t great in a world where people are focused on hooking up Stoneforge Mystic and True-Name Nemesis. Mana curves have spread out, but artifact control decks are still shooting at targets from literal years ago. This archetype can evolve, and it should. Metalworker and Grim Monolith were banned at the outset of Legacy because they were phenomenally powerful cards, and they haven’t just stopped being incredibly good—people just stopped updating the deck because it plays an inconsistent game of Magic. That can change.
One way to change it is to tighten up the mana curve and play fewer cards like Spine of Ish Sah and Sundering Titan – cards that are conditionally good but terrible when you draw them. Truly successful decks tend to find ways to make use of “dead cards.” For instance:
Reanimator, a deck that wants creatures in its graveyard, plays Careful Study and Show and Tell when it draws creatures instead of Entombs. Omni-Tell plays Dream Halls, which it can cast when it doesn’t draw Show and Tell. Same with Sneak and Show and its eponymous four-mana enchantment.
On top of all of that, this Metalworker sub-archetype is basically a Dream Halls deck – Kuldotha Forgemaster does cost five – that plays no Show and Tells and a tutorable Emrakul, the Aeons Torn that can be Swords to Plowshared. There’s a reasonable argument to be made for the power level of the endgame being too low.
I considered most of these arguments in proposing a more streamlined and aggressive Metalworker deck in my Getting into Legacy article. For reference:
Creatures (20)
Lands (24)
Spells (16)
One of the major problems with the deck is that it has pretty bad Metalworkers. Like Affinity, any Metalworker deck has to intentionally suppress the number of non-artifacts it plays. I forget where I heard this, but a rule of thumb that managed to ingrain itself into my Magic-playing consciousness is that an Affinity deck wants at least forty-eight artifacts. I built my Legacy Affinity list with that “rule” (guideline?) in mind, playing four Thoughtcast, four Ancient Tomb, a Tundra, and three Tezzeret, Agent of Bolas as my only non-artifacts—48 artifacts on the nose.
So what happens if we work to increase artifact count? We clearly want Ancient Tomb and City of Traitors, but what if we just played Mountains, Goblin Welders, and Faithless Lootings as the remainder of our non-artifacts? How would this deck look?
Creatures (23)
Lands (19)
Spells (18)
This deck may look familiar, even to those of you who haven’t done much reading about Metalworker before today. That’s because the logic that drives card selections in this deck drove a lot of what was going on in Cleveland Rhodes’ deck from April 2013:
Creatures (20)
- 4 Metalworker
- 4 Goblin Welder
- 4 Lodestone Golem
- 3 Wurmcoil Engine
- 1 Steel Hellkite
- 3 Kuldotha Forgemaster
- 1 Blightsteel Colossus
Planeswalkers (1)
Lands (20)
Spells (19)
The way my thought process worked is as follows:
“If the point of Metalworker is to activate for a lot of mana and then use that mana, you want both a lot of artifacts and a good density of artifacts that cost enough to make Metalworker worth it. We need a lot of artifacts. Let’s start by figuring out how much the top end threats should cost.
Metalworker taps for even numbers all the time—two per artifact. We have eight two-mana lands, so that’s additional even mana production. Let’s shoot for six-drop artifacts, so we can reveal that plus two other things and cast our fatty. Ideally, we can follow up with something else, but we’ll get to what those should be later.
The best six-drops in artifact decks are Steel Hellkite and Wurmcoil Engine. There is no real option for anti-combo threats that also happen to be live against decks with creatures—why in the world does Nullstone Gargoyle cost nine mana?
If we’re playing red, we obviously want Goblin Welder. I remember one of the big problems with Goblin Welder from my experience playing it in Painter decks is that sometimes, you don’t have anything to do with it. It just sits there and doesn’t activate, which sucks. Having Intuitions was a big part of why Welder was good in Painter, but we’re not playing blue or Intuition—what else can we do?
I remember John Cuvelier had a sweet Welder Reanimator deck back when the SCG Invitational was in Richmond. What did that look like?
Creatures (16)
- 1 Platinum Angel
- 2 Sundering Titan
- 4 Goblin Welder
- 4 Hapless Researcher
- 1 Inkwell Leviathan
- 1 Sphinx of the Steel Wind
- 1 Iona, Shield of Emeria
- 1 Steel Hellkite
- 1 Platinum Emperion
Lands (18)
Spells (26)
Jeez, that deck could use a Faithless Looting.
Hey, Faithless Looting goes well with Goblin Welder, and if you can cast one, you can cast the other. Get some Lootings in there.
How do we cast all these red cards? We’re going to need more Mountains and we definitely want some Moxes. What kind of Moxes? Well, Mox Opal is legendary, so that’s a drawback…
…actually, no, wait, new legend rule, it’s GREAT. We can play one Mox, tap it, play the other, sacrifice the first, tap the second, Welder the first back in, and build a sort of Goblin Birds of Paradise. Also, as soon as we have a second Mox, our Welder turns on and never turns off because we always have at least a Mox to Weld in. This is great!
Also, let’s check Mox Diamond’s rules text. Yup, it has an if clause on entering the battlefield, not casting, so Welding that one in is going to be a nonstarter. Mox Opal it is, and definitely four.
I remember having a lot of fun building a draw engine with Sensei’s Divining Top and Goblin Welder by responding to the draw ability with Goblin Welder – Goblin Archivist – and this deck could probably use some more card selection anyway. Let’s start with three but be open to playing four.
What am I missing? Oh, right, Lodestone Golem. Four of those, and let’s fill things in with some Phyrexian Metamorph—not too many since we don’t want to draw a ton with nothing to copy—and a few Lightning Greaves to protect and haste up our Metalworkers and Welders and Hellkites. Might as well get a Staff in here. We can’t one turn kill them, but we don’t have to draw our entire deck, just enough to put an overwhelming amount of stuff in play. This should be fun.
How does this sideboard work? Where are we weak?
Combo looks like it’s going to be a nightmare. We don’t have counters or discard or mana disruption, just Lodestone Golem. We are going to need a lot to beat them. We also don’t actually have a great way to beat True-Name Nemesis outside of flying over with Steel Hellkite and exploding their board for three. Let’s start there.
I want a card like Elesh Norn that can just Plague Wind someone. I need it to kill True-Name Nemesis and ideally small creature decks in general.
So Contagion Engine, which also costs six—ideal for Metalworker to cast and kills a Nemesis on sight. Cool, next.
Blood Moon is legitimately castable in this deck, although we aren’t turn 1ing it like Painter. I still think it’ll play against Delver and so on.
We need some low-cost disruption against combo. Phyrexian Revoker is a classic go-to and it works against a decent range of decks. Ensnaring Bridge shuts off Reanimator and Sneak. We could Mindslaver some decks, but it’s garbage against others. We could Thorn of Amethyst or Sphere of Resistance, but we basically need a bunch of those. We could Tangle Wire, which goes well with Goblin Welder . . . Okay, that’s probably the best we’re going to do. Not sure that we shouldn’t just maindeck Tangle Wire since I did advocate for maindeck Wire in my Getting into Legacy article, but I’m not sure what to cut. The forums will probably straighten that out.
So we have Revokers and Wires and maybe Moons against combo, Moons against tempo, and Contagion Engine against creatures. What else? A few Ensnaring Bridges? They probably can’t hurt—even against Abrupt Decay decks, they still have to kill Welder and Metalworker, and we can Metamorph it, so we’re probably fine there. We want it against Tarmogoyf and Griselbrand, which is probably enough of the metagame to justify its existence.
I wish Torpor Orb were worth it. Maybe Batterskull is worth it? Probably not—it’s just a worse Wurmcoil where it counts. Too bad Trash for Treasure isn’t good enough. God, Goblin Welder is so filthy. Okay, let’s put some numbers on these.
We need three Engines if we’re going to draw and cast them enough. Same with Moons, and we probably can’t have enough Revokers in a world where we want any. So three/three/four, four Wires, and that only leaves room for one Bridge? We can probably shave a Revoker if we have four Wires. So that leaves us with:
4 Tangle Wire
3 Contagion Engine
3 Phyrexian Revoker
3 Blood Moon
2 Ensnaring Bridge
Are we missing anything important? Our plan against Delver is to drop sixes on them until they die. Our plan against Tarmogoyf decks is to either slam a fatty and race with lifelink/kill their board from above, or hide behind Bridge, build up a huge board, and weld Bridge out and kill them. Our plan against combo is to board in our Wires, maybe Moons, Revokers, and maybe Bridges for our sixes and a few other cards and kill them with as many 5/3s as we can find.
We can definitely cover a Griselbrand, although it’s not the easiest card to beat. True-Name Nemesis isn’t even that hard to beat with our creatures and long game.
Not entirely sure on Contagion Engine, but if it’s good, it’ll be real good. If it’s bad, we have Ratchet Bomb or Powder Keg or whatever. I’m sure the folks in the comments will tell me what I missed.