Casualty Wants a Corpse for Every Spell. Sedgemoor Witch Pays You Two.
Casualty charges you a creature for every copied spell — and Sedgemoor Witch hands you two back, because magecraft reads "cast or copy". We walk Silverquill, the Disputant through all four official rulings (the copy resolves first, and it is never "cast", so your storm count doesn't move), then run a real 100-card Orzhov list through DeckCortex's combo engine: zero complete combo lines as built. One $1.47 Gravecrawler takes it to six at once and moves the bracket floor from 2 to 3. Whether those six lines are worth the table talk is the whole question.
Silverquill, the Disputant will copy every instant and sorcery you cast, and it charges you a creature each time. That bill is the entire deckbuilding problem, right up until you put a 3/2 Human Warlock on the table and the deck starts paying itself.
Four mana, , a 4/4 flier with vigilance. The text that matters: each instant and sorcery spell you cast has casualty 1. Casualty is an optional additional cost. As you cast the spell you may sacrifice a creature with power 1 or greater, and when you do, you copy the spell and may choose new targets for the copy.
So the dragon is a rate. Every removal spell exiles two things. Every drain drains twice. Every token spell makes double the tokens. The price is a body, every time, and bodies are not free.
The problem nobody warns you about
Read casualty as a cost and you build the deck wrong. You start counting creatures, decide you need thirty of them, and end up with a pile of 2/2s that do nothing except die politely so your removal can hit two targets.
The good version of this deck doesn't count creatures. It manufactures them faster than casualty eats them, and it gets paid at both ends of the transaction.
Here's the card that turns the corner:
Magecraft — Whenever you cast or copy an instant or sorcery spell, create a 1/1 black and green Pest creature token with "When this token dies, you gain 1 life."
Those two words carry the whole deck. A casualty copy is a copy, so magecraft sees it. Cast one spell with the Witch on board and you get two triggers: one for the cast, one for the copy. Two Pests.
Casualty took one creature. Magecraft handed back two. Every spell you cast leaves you one body richer than you started, and the next spell's casualty cost is already sitting on the battlefield waiting to be spent.
Run it once, slowly, with Village Rites — a one-mana instant that sacrifices a creature and draws two cards:
- You cast it. Village Rites has its own additional cost (sacrifice a creature) and Silverquill adds casualty 1, so you pay two bodies during casting. Both are additional costs, both paid before anything goes on the stack.
- Two triggers go up: casualty's copy trigger, and magecraft for the cast.
- The casualty trigger resolves and creates a copy of Village Rites on the stack. That copy is itself a copy, so magecraft triggers again.
- The Pests arrive. The copy resolves, drawing two. The original resolves, drawing two.
For one black mana you drew four cards, ended with the same number of creatures you started with, and gained two life when the Pests were sacrificed later. The spells that already ask you to sacrifice — Village Rites, Corrupted Conviction, Culling the Weak, Deadly Dispute, Diabolic Intent — break even on bodies. Everything else runs positive.
And the bodies you spend aren't wasted on the way out. With Blood Artist, Zulaport Cutthroat, Bastion of Remembrance or Elas il-Kor, Sadistic Pilgrim on board, each casualty payment drains the table. Mirkwood Bats is the greedy one: it triggers whenever you create or sacrifice a token, so a single casualty'd spell with the Witch out is three separate Bats triggers: two Pests created, one token sacrificed. Casualty stops being a cost and starts being a trigger.
Four rulings that decide how your turns actually play
Casualty's rules text hasn't moved since 2022, and Silverquill ships with the same four rulings Anhelo did. They're short, and each one changes how you sequence:
The copy resolves before the original. The copy trigger goes on the stack above the spell that made it, so it resolves first. When you copy an Exsanguinate, the copy drains, then the original drains. This matters for anything where the board changes in between.
You may sacrifice only one creature per casualty ability, and you copy the spell only once. No stacking two 1/1s to pay a bigger casualty number, no paying twice for two copies.
The copy is created on the stack, so it's not cast. This is the one people get wrong at the table. Abilities that trigger on casting won't see the copy. Your copies do not add to storm count — if you're building toward a storm payoff, casualty gives you nothing. Magecraft is the exception precisely because it reads "cast or copy" rather than just "cast", and that asymmetry is worth knowing before you argue about it in a pod.
New targets are optional. A copied Anguished Unmaking can exile two different nonland permanents, or hit the same one twice if the first gets countered. Note that the whole spell copies, downside included: you lose 3 life twice, not once.
One more that isn't in the rulings but follows from how copies work: X carries over. The copy of an X spell uses the same X you paid for. Debt to the Deathless at X=5 takes each opponent for two times X, so ten, and then the copy takes them for ten again. Twenty life out of each player, for nine mana and one 1/1.
Where the bodies come from
You need a creature available every time you want to cast a spell, which means recurring token production, not a one-time board. Four cards in this list do it on a timer:
Every one of them makes a body on a schedule whether you do anything or not. Ophiomancer is the neat one: it checks each upkeep, including your opponents', and makes a 1/1 deathtouch Snake if you control no Snakes, so it refills as fast as you can spend it. Jadar's decayed Zombie is a 2/2, the only fodder here with power above 1.
That threshold is the whole reason to prefer this commander. Anhelo, the Painter, the better-known casualty card, grants casualty 2, and only to the first instant or sorcery you cast each turn. Every 1/1 token in this deck is too small to pay for it, and you'd get one copy a turn regardless. Silverquill asks for less and gives more.
The one trap: Treasure is not a creature. Warren Soultrader, Pitiless Plunderer and Deadly Dispute make mana, not fodder. They're excellent here, but they don't pay casualty.
Then there are the X spells that feed themselves. Secure the Wastes makes X 1/1 Warriors at instant speed; copy it and you get 2X for the price of X and one body. Grand Crescendo does the same and hands your team indestructible. Blot Out the Sky makes X flying Inklings and, if X is 6 or more, destroys all noncreature nonland permanents. The copy checks that clause too, so a big one is a doubled army on top of a doubled wipe.
What's worth doubling
Removal is the obvious half: Swords to Plowshares and Path to Exile for one mana each, exiling two creatures. Damn cast for its overload cost, twice, when once wasn't enough.
The half that ends games is the drain:
Torment of Hailfire repeats its process X times, and each opponent either loses 3 life, sacrifices a nonland permanent, or discards. Copy it and they make that choice 2X times. At X=6 across three opponents, that's twelve rounds of a decision nobody survives making.
Two cards in the list multiply the multiplier:
Twinning Staff reads: if you would copy a spell one or more times, instead copy it that many times plus an additional time. Casualty's one copy becomes two. One corpse, three resolutions of the spell, and three magecraft triggers if the Witch is out. Elspeth, Storm Slayer doubles every token you create, so each magecraft trigger makes two Pests instead of one — four per spell, or six with the Staff also on board. At that point the deck is producing casualty fodder several times faster than it can spend it.
The list
This is a real 100-card Orzhov build, tokens and aristocrats rather than the creatureless version you'll sometimes see written up. It runs about $570 / €483 at Scryfall's paper aggregate across these exact printings, with five rows unpriced (the basics, plus the specific printings of Debt to the Deathless, Fellwar Stone and Godless Shrine that Scryfall has no USD figure for). Elspeth, Storm Slayer at $42 and Torment of Hailfire at $36 are the two most expensive cards in it.
Worth saying plainly: DeckCortex's scraped corpus is tournament-sourced, and casual Commander isn't in it. There is no metagame share, inclusion percentage or trend for this deck, and this article doesn't invent one. What our data can do is read the list itself.
One card, six combos, and a bracket you might not want
Run the list through DeckCortex's combo engine and it comes back with zero complete combo lines. Not one. For a deck stuffed with sacrifice outlets, recursion and drain triggers that's a surprise, and it's the good kind: the deck is winning on rate rather than on a combo kill.
The engine also reports the near misses, and one card sits on top of that list by a wide margin:
A one-mana 2/1 Zombie that can't block, and: you may cast this card from your graveyard as long as you control a Zombie. Power 2 clears casualty 1 comfortably, and it comes back for every time you spend it. Recurring casualty fodder that costs a card slot rather than a card. The list already controls Zombies through Warren Soultrader, who is a Zombie Goblin Wizard, and Jadar's decayed Zombie tokens.
It also completes six combo lines that are otherwise sitting in the deck unfinished:
| Line | What it produces | Combo popularity | |---|---|---| | Warren Soultrader + Gravecrawler + Blood Artist | Infinite death triggers, life loss, life gain | 43,718 | | Warren Soultrader + Gravecrawler + Zulaport Cutthroat | Same shape | 39,142 | | Gravecrawler + Pitiless Plunderer + Ashnod's Altar | Infinite colorless mana and death triggers | 36,650 | | Warren Soultrader + Gravecrawler + Bastion of Remembrance | Same shape as the Blood Artist line | 25,744 | | Warren Soultrader + Gravecrawler + Elas il-Kor, Sadistic Pilgrim | Infinite drain and sacrifice triggers | 5,642 | | Warren Soultrader + Gravecrawler + Kambal, Profiteering Mayor | As above, plus infinite life gain | 2,367 |
Combo popularity is DeckCortex's own figure for how many lists in our combo corpus run that line. It's a measure of how well-known each combo is, not a share of Silverquill decks.
Six lines, for $1.47. Those are the cheapest six infinite combos you will ever buy, and you should think hard before buying them.
Here's why. Adding Gravecrawler and re-running the same analysis moves the deck's bracket floor from 2 (Core) to 3 (Upgraded) in DeckCortex's bracket-fit model. Bracket 2 permits zero complete combo lines; you would have six. Nothing else about the deck changes: still no Game Changers, no fast mana, no mass land destruction, the same two tutors in Diabolic Intent and Dark Petition. A $1.47 card, a whole bracket.
That's a floor rather than a verdict, since the model's power-level simulation isn't part of this number and the real bracket can only sit higher. The direction is unambiguous though, and it makes Gravecrawler a decision instead of an upgrade. If your pod plays bracket 2 and likes it there, this card quietly makes you the person who brought a combo deck. If you're already at 3, it's the best card you can add: it fixes the fodder problem and gives you six ways to close a game you're winning anyway.
Want the fodder without the combos? Reassembling Skeleton returns itself from the graveyard to the battlefield tapped for , costs $0.35, and completes exactly one line rather than six. It's slower and it needs no Zombie. For a bracket-2 table it's the honest pick.
Rebuild the list in DeckCortex with whichever one you choose and read the combo count before you sleeve it up. A dollar separates those two cards. A whole conversation with your playgroup separates what they do to your deck.
Card text verified against live Oracle data; casualty rulings as published on Silverquill, the Disputant (2026-03-20). Prices are Scryfall paper market aggregate, not a live TCGPlayer feed, and move constantly. Combo lines and combo popularity come from DeckCortex's combo engine; bracket floor is our deterministic bracket-fit model, evaluated without its client-side power-level simulation, so it is a lower bound rather than a final 1–5 bracket. DeckCortex's deck corpus is tournament-sourced, so no metagame share, inclusion rate or trend figure exists for casual Commander and none is quoted here.